U. S. Food and Drug Administration
Center for Food Safety and Applied Nutrition
FDA Prime Connection
Proceedings of the 1994 Vibrio vulnificus Workshop - 06/15-26/94
PROCEEDINGS
OF THE
1994 Vibrio vulnificus WORKSHOP
June 15-26, 1994
Cohen Building
300 Independence Avenue
Snow Room -- 5051
Washington, D.C.
Second Printing
Sponsored by:
Department of Health and Human Services
Public Health Service
Food and Drug Administration
Department of Commerce
National Oceanographic and Atmospheric Admininstration
National Marine Fisheries Service
Interstate Shellfish Sanitation Conference
Foreword
This document is an abridgment of the presentations and
discussions that occurred at the 1994 Vibrio vulnificus Workshop
held June 15-16, 1994, in Washington, D.C. The Workshop was
jointly sponsored by the U.S. Food and Drug Administration, the
National Marine Fisheries Service, and the Interstate Shellfish
Sanitation Conference. This document was constructed based on
information obtained from an unabridged Transcript of Proceedings,
produced by Miller Reporting Company, Inc., Washington, D.C. The
scientific presentations in this Second Printing of the Proceedings
have been reviewed by the respective speakers for accuracy and
comprehensiveness.
This Printing has been assembled, edited, and produced by the
FDA's Office of Seafood to provide essential information acquired
during the Workshop to participants and other interested parties.
The skillful technical assistance of L. Tomlinson and W. Horwitz,
FDA, Center for Food Safety and Applied Nutrition, are gratefully
acknowledged. Copies of the Second Printing are available by
request from the FDA, Office of Seafood, 200 C Street S.W.,
Washington, D.C. 20204.
William Watkins, Ph.D.
and
Susan McCarthy, Ph.D., Editors
C O N T E N T S
Page
Foreword ................................................... 2
Contents ................................................... 3
Part I. SCIENTIFIC AND TECHNICAL UPDATE, June 15, 1994
Introductory Remarks: Mr. John Marzilli .............. 5
Welcoming Address: Mr. Thomas Billy .................. 5
Vibrio vulnificus Overview: Dr. William D. Watkins ... 9
Epidemiology: Dr. Cynthia Whitman ................... 13
Risk Assessment and Current Data: Dr. Gary Hlady .... 27
Case Studies: Mr. Charles Kaysner ................... 39
Pathogenicity: Dr. Glenn Morris ..................... 43
Methodological Considerations: Dr. Anita Wright ..... 55
Animal Models: Dr. James Oliver ..................... 57
Viable But Nonculturable State: Dr. James Oliver .... 63
Ecology of Vibrio vulnificus: Dr. Mark Tamplin ...... 75
Time-Temperature Factors: Dr. David Cook ............ 87
Oyster Purification and Intervention Measures:
Dr. Mark Tamplin ............................... 101
Relaying: Dr. Steve Jones .......................... 105
Other Gulf Coast Oyster Research:
Dr. Marilyn Kilgen ............................. 111
Icing at Harvest: Dr. Mark Tamplin ................. 117
Consumer Education and Health Advisory Programs:
Ms. Ruth Welch ................................. 119
Retail Food Safety: Mr. Larry Edwards .............. 131
Concluding Remarks: Dr. William D. Watkins ......... 137
Discussion .......................................... 143
Part II. INFORMATION NEEDS AND APPROACHES, June 16, 1994
Today's Focus: John Marzilli ....................... 147
Panel 1. Epidemiology and Risk Factors ............. 147
Panel 2. Causative Factors: Pathogenicity,
Virulence, VBNC, Lethality ................ 151
Panel 3. Causative Factors and Controls:
Time/Temperature, Harvesting,
Handling, Processing ...................... 155
Discussion .......................................... 161
Panel 4. Causative Factors: Ecological and
Environmental Influences .................. 171
Concluding Remarks: Dr. William D. Watkins ......... 173
P R O C E E D I N G S
Part I. SCIENTIFIC AND TECHNICAL UPDATE, June 15, 1994
INTRODUCTORY REMARKS
Mr. John Marzilli
Division of Field Science
Office of Regulatory Affairs
Food and Drug Administration
Rockville, Maryland
Good morning. My name is John Marzilli. I am with the Food
and Drug Administration and I would like to welcome you to the 1994
Vibrio vulnificus Workshop. I am the facilitator for the workshop
sessions.
We are recording the sessions, and a summary of the technical
presentations and information needs identified during the Workshop
will be produced in a Proceedings document which will be available
at the Interstate Shellfish Sanitation Conference meeting this
August 6-12 in Tacoma, Washington. Following that meeting, the
Proceedings will also be available through the National Marine
Fisheries and through the Office of the ISSC.
We have a full schedule for the next two days, and we will try
to keep speakers on track and follow the agenda closely so that
people arriving for certain presentations will be able to obtain
the information they may specifically want at the times designated.
Now, I'd like to call on the Director of the Office of Seafood, Mr.
Thomas Billy, to welcome us to today's meeting.
WELCOMING ADDRESS
Mr. Thomas Billy
Director, Office of Seafood (HFS-400)
Center for Food Safety and Applied Nutrition
Food and Drug Administration, 200 C Street, S.W.
Washington, D.C. 20204
I would also like to give a warm welcome to all of the
speakers and researchers, officials of the state and federal
government organizations, and everyone in the audience. We have
two full days scheduled. During this time, we believe that all of
us will become well informed about the nature of the problem facing
consumers, industry, and regulatory officials in dealing with
Vibrio vulnificus.
I am delighted that we have been able to attract an impressive
array of experts in this area. It is a very important public
health issue that we are addressing. The workshop has been
arranged to provide an update on the scientific and technical
progress that has been made on V. vulnificus since the 1988
workshop and to identify further information needs.
It is hoped that the format for this workshop will provide those
attending with a present day perspective on the hazards presented
by this naturally occurring pathogen and will enable researchers,
public health officials, and managers to assess future needs and
further actions that may be warranted.
V. vulnificus in raw molluscan shellfish presents us with a
tremendous regulatory, scientific, and societal challenge. Its
effects can be mild in humans or it can be a deadly pathogen that
kills very quickly. It claims lives every year and most of these
fatalities occur around this time of the year.
V. vulnificus is a natural part of the ocean's ecosystem but
in many respects it is a mystery. Until now, it has defeated our
best efforts to adequately understand and control it. We are here
today to review what is known and what we still need to learn about
V. vulnificus.
The FDA has been and continues to be deeply concerned about
the illnesses and the deaths that have occurred from the
consumption of raw shellfish, and particularly those that have been
caused by V. vulnificus. We have made the control of this
naturally occurring pathogen a high priority.
This workshop is being jointly sponsored by FDA, by the
National Marine Fisheries Service, and by the Interstate Shellfish
Sanitation Conference. It is evident that we are collectively
seeking to arrive at a control strategy that is scientifically
based and appropriate to the level of consumer risk that exists.
As public health officials, we recognize our responsibilities to
the consuming public. Although some will argue that this
particular issue is extremely complex, I have found in my last
couple of years with FDA that there are rarely any easy answers.
I trust the information resulting from this workshop will
contribute to our knowledge and will enable us to determine the
best course of action based on the available scientific
information. It is my expectation that the outcome of this session
will serve as a catalyst for the ISSC to take responsible action on
this issue at their annual conference to be held in Tacoma,
Washington, this August.
Each of the invited speakers will be summarizing the
scientific and technical information on a particular topic. Some
of the topics related to V. vulnificus being covered today include
the epidemiology, pathogenicity, and ecology of the organism, as
well as time and temperature factors, purification and intervening
control measures, consumer education, and health advisories.
I also want to emphasize that this workshop is for the
exchange of scientific and technical information. It is not a
forum to hear public testimony or to provide advice to the Agency.
Again, I welcome all of you, and I hope that the workshop will
be both stimulating and successful. Thank you very much.
OVERVIEW OF THE Vibrio vulnificus PROBLEM
1988 Perspective and the 1994 Vibrio vulnificus Workshop Approach
Dr. William D. Watkins
U.S. Public Health Service
Northeast Seafood Laboratory
Food and Drug Administration,
Building S-26, School Street, Davisville
North Kingstown, Rhode Island 02852
Introduction
Vibrio vulnificus is naturally occurring and, as such, it
presents a particular problem for the National Shellfish Sanitation
Program (NSSP), a program initially devised to protect human
consumers against enteric pathogens principally derived from human
waste and sewage. V. vulnificus has been recognized as a pathogen
since the late 1970s, when it was referred to as the lactose-
positive Vibrio species. The major concern the NSSP has for this
organism is that it causes sepsis in humans, and this can result in
serious disease and even death. Death results from a rapidly
progressing, fulminating septicemia. V. vulnificus can be
transmitted by the consumption of raw shellfish, and in this
country the vehicle has been oysters. It is estimated that nearly
200 people have succumbed to this infectious disease since the late
1970s, an average of about 15 per year. No significant change in
mortality figures has occurred to date. One consequence of the
recurring cases of mortality is that a great deal of unfavorable
publicity has been generated, critical of both the shellfish
industry and the state and federal authorities charged with
oversight responsibilities. Another consequence of the growing
number of victims is a considerable amount of litigation. Needless
to say, the occurrence of illness and mortality caused by ingesting
V. vulnificus has served to decrease public confidence in molluscan
shellfish generally and to repress the U.S. oyster industry. The
problem is a difficult issue for regulatory officials and the
shellfish industry to address, as there are many operative factors,
both known and unknown.
Foremost among the known factors involved is the natural
occurrence of the causative organisms in estuarine waters
worldwide. They are not associated with fecal contamination and,
therefore, their distribution cannot be controlled. Moreover, this
species occurs as normal flora in oysters. Routine refrigeration
prevents the growth of this mesophilic halophile, but does not
eliminate viable cells of the organism.
V. vulnificus mortalities occurring in the United States, to
date, have had a decidedly regional source and seasonal occurrence;
essentially all have involved raw Gulf Coast oysters, with warmer
periods accounting for the large majority of cases. Also,
virtually all cases involved oysters which had been in the
distribution chain for at least a day; no cases were known to be
caused by oysters freshly harvested and consumed. This suggests
the possibility that a lethal density or dosage is related to
proliferation of the pathogen after harvest and during handling
without refrigeration (temperature abuse).
Victims have invariably been found to have some underlying
health problem, most frequently blood and liver disorders and
immune deficiencies. This high-risk group of people has been and
continues to be the target of an extensive FDA educational advisory
effort, principally via health care professionals. Warning labels
are required for raw oysters in some states. The efficacy of these
warning measures for reducing morbidity and mortality is unknown.
Lawsuits stemming from deaths continue to threaten the collapse of
the industry.
1988 Perspective
In 1988, a jointly sponsored workshop on V. vulnificus was
held in Washington, D.C., to address concerns of public health
officials, researchers, and the oyster industry. Approximately 60
participants met over three days, and the workshop broke into four
work groups. A quick review of some of the outcomes from that
workshop will give a better understanding and recognition of the
achievements during the past six years. The four work groups were
assembled to identify the state of knowledge and the unknown
factors, address issues and concerns, and provide recommendations
for obtaining answers and solutions in improving shellfish safety.
A prevention and control panel also addressed needs facing
government, industry, and consumers. A brief synopsis of the
products from these groups follows.
1. Environmental work group
The group unanimously raised the questions, "Why is V.
vulnificus found only in Gulf Coast oysters in this country? What
are the factors related to oyster-borne transmission of V.
vulnificus?" Discussions and deliberations resulted in the
following recommendations: a) education of the medical community,
particularly physicians, and of shellfish consumers, especially
those that are at high risk due to recognized underlying health
concerns; b) research on the efficacy of depuration for eliminating
the risk from V. vulnificus; c) research on the ecology of this
estuarine species to understand why it causes a problem seasonally;
d) research to learn of the numbers of V. vulnificus involved in
illness and in death; and e) research on the regional differences
found for this organism, in its abundance and occurrence, possible
strain differences, and particularly those related to the
organism's virulence.
2. Epidemiological work group
This group recommended the following: a) determine
contributing factors related to the morbidity and mortality of this
organism; b) develop educational measures for physicians and
consumers; c) intensify reporting of illnesses caused by V.
vulnificus because delayed reporting has seriously hindered the
ability to determine the impacts of the distribution system and
shipping, the time intervals between harvest and consumption, and
the roles of imports; d) use a standard complaint form to enhance
reporting; e) cooking of oysters during all dangerous periods.
The group also provided a list of options for prevention of
illness: a) cease all oyster harvesting; b) cease some harvesting
during certain periods in certain regions; c) market only shucked
oysters during certain periods from certain regions.
3. Time-Temperature work group
The group generally concurred that temperatures of
refrigeration recommended by ISSC appeared to be reasonably
effective in preventing the growth of V. vulnificus in oysters. It
was queried, however, that if bacterial growth via temperature
abuse is the causative factor, why are illnesses from other
microorganisms not being reported; why only V. vulnificus? The
work group recommended the following: a) examine temperatures of
oysters from harvest to consumption and the times involved; b)
determine the adequacy of truck refrigeration systems being used;
c) determine the growth rates of V. vulnificus in oysters; d)
develop time-temperature curves; e) evaluate depuration of vibrios;
f) stress education, including the harvesters, the shippers, the
processors, and the retailers; g) conduct a trace lot study to
determine V. vulnificus levels in oysters at harvest and thereafter
in the distribution chain; h) determine the effectiveness of
cooling rates; i) develop ways to estimate doses that cause
infection and deaths.
4. Analytical Methods work group
This group deliberated the merits and deficiencies of various
methodological approaches and procedures and made the following
recommendations: a) adopt the alkaline peptone water (APW) MPN as
an interim standard enumeration method to provide comparability
among research data; b) develop more rapid enumeration methods; c)
develop more rapid confirmation methods to specifically identify V.
vulnificus isolates, especially the refinement and use of a
cytotoxin-hemolysin gene probe; d) determine the virulence factors
of V. vulnificus; e) investigate the number of V. vulnificus
organisms normally consumed and those involved in morbidity and
mortality cases; and f) develop educational measures for health
care professionals and consumers.
5. Prevention and control panel
This panel provided the following recommendations: a) educate
all parties concerned, especially health care professionals and
high-risk individuals; b) use a standard quantitative method; c)
establish uniform surveillance and reporting of illnesses and
cases; and d) evaluate the influence of current industry practices
on V. vulnificus densities in oysters.
1994 Workshop Approach
A considerable amount of activity has occurred during the six
years since the 1988 Workshop. FDA, National Marine Fisheries
Service (NMFS), and ISSC have convened this workshop to gauge what
has been accomplished and what further activities are needed to
bring this problem under control, being mindful of protecting a
particularly vulnerable high-risk consumer group, and aiding an
important seafood industry.
The format for this workshop differs from that used
previously. Participants in 1988 separated into work groups and,
consequently, everyone missed direct involvement in the three
quarters of the discussions and deliberations. This year, experts
in the field have been invited to provide scientific and technical
updates in specific areas of concern to all those attending. This
will enable everyone present to receive the information provided
and afford ample opportunity for audience participation. This is
intended to encourage individuals to offer issues that have not
been raised, bring forth information that may not have been
presented, and ask questions and enter into discussion as
appropriate.
Following the scientific and technical update, panel groups
formed from among the invited speakers and officials will identify
and discuss remaining information needs and how these might be
satisfied. Again, there will be ample opportunity to contribute,
question, and help to resolve relevant concerns and needs. It is
intended that this approach will enable fuller participation from
the experts and at the same time allow for ample audience
participation and interaction as well. Welcome on behalf of the
FDA, NMFS, and the ISSC.
Reference
1. Miescier, J. (ed). Vibrio vulnificus Workshop. 1988.
Shellfish Sanitation Branch, Division of Cooperative Programs,
Office of Compliance, Center for Food Safety and Applied
Nutrition, U.S. Food and Drug Administration, DHHS,
Washington, DC, 74 pages.
EPIDEMIOLOGY
Overview of the Important Clinical and Epidemiologic
Aspects of Vibrio vulnificus Infections
Dr. Cynthia Whitman
Foodborne and Diarrheal Diseases Branch
Bacterial and Mycotic Diseases Division
National Center for Infectious Diseases
Centers for Disease Control and Prevention
Atlanta, Georgia 30333
Case History
In August, 1992, the Florida State Department of Health
notified the Centers for Disease Control and Prevention (CDC) that
a Georgia resident had died while on vacation in Florida. The
victim was a 40-year-old man with hepatitis C who was being
evaluated for a liver transplant. On July 23rd, he consumed two
dozen raw oysters at a local restaurant as part of his evening
meal. Four days later, on July 27th, he presented at a local
emergency room in hypotensive shock with nausea, vomiting, and
diarrhea; blistering, painful wounds covered both legs. He died
within 24 hours despite aggressive antimicrobial therapy and
admission to the intensive care unit. The hospital laboratory
identified Vibrio vulnificus in a blood specimen drawn from the
patient. This history is representative of the cases related to
oyster-borne V. vulnificus, the focus of this workshop.
Introduction
In 1976, Vibrio vulnificus, a ubiquitous, gram-negative
bacterium found along United States coasts, was first identified
and described by the Enteric Diseases Laboratory of CDC (5). It
was initially referred to as "halophilic Vibrio X" and later as
"lactose-postive Vibrio." In 1979, a CDC epidemiologist first
described the clinical syndromes and epidemiology of V. vulnificus
infection, and gave it its current name (1). Between 1979 and
1988, knowledge of the clinical and epidemiological aspects of V.
vulnificus infection increased through case control studies, case
series, and case reports. A summary of this information is
presented in the next few sections before the preliminary results
from the first six years of regional surveillance along the Gulf
Coast are described.
Overview of Clinical Syndromes
Infections with V. vulnificus (referred to here as
"vulnificus") are associated with three distinct clinical
syndromes.
1. Primary septicemia, the most severe syndrome
Primary septicemia occurs when food containing vulnificus
organisms is consumed and when the bacteria invade the bloodstream
and disseminate throughout the body. It is characterized by fever
and chills, usually accompanied by nausea, vomiting, and diarrhea.
A sharp drop in blood pressure commonly occurs, and this can lead
to intractable shock and death. The majority of patients also
develop painful skin lesions. The skin initially appears red and
blisters quickly develop and erode into necrotic ulcers.
More than 50% of patients with primary septicemia die despite
appropriate medical care. One-third of patients present for
medical attention in hypotensive shock, and 90% of them die (7,8).
Like the transplant candidate described above, more than 95% of
patients with primary septicemia have a pre-existing, chronic
illness, usually involving the liver (1,7,8).
2. Gastroenteritis
Ingesting food containing vulnificus organisms can also cause
gastroenteritis. Patients with gastroenteritis have a relatively
milder syndrome consisting of vomiting, diarrhea, and abdominal
cramps. Other pathogens along with vulnificus are often isolated
from cultures obtained from patients with gastroenteritis.
Patients with gastroenteritis may require hospitalization but
gastroenteritis, rarely, if ever, results in death.
3. Wound infection
In contrast to the ingestion syndromes just described, wound
infections are acquired when skin lacerations and abrasions come in
direct contact with seawater or during acute, penetrating marine
injuries. Vulnificus wound infections typically begin with
swelling, redness, and intense pain around the infected site.
Fluid-filled blisters often develop and progress to tissue necrosis
in a process resembling gas gangrene in its speed and severity.
Fifty percent of patients with vulnificus-infected wounds require
surgical debridement or amputation. In some patients, infection
spreads to the blood stream, and, in such cases, death commonly
occurs.
The overall mortality for vulnificus wound infections is
approximately 25%. However, for patients with a pre-existing
chronic illness, the outcome is worse; mortality rates of greater
than 50% have been reported in some case series (2,8).
Medical Treatment
The mainstays of medical treatment for vulnificus infection
are prompt antimicrobial therapy and supportive care. Vulnificus
is sensitive to most of the antimicrobials used in the initial
treatment of bloodstream or severe wound infections. Tetracycline
is the drug-of-choice. Historically, it has been an effective
agent against other, more common Vibrio species. In addition, some
case series have suggested that seriously ill patients treated with
tetracycline have better outcomes than those treated with other
antibiotics (7). It is essential to keep in mind, however, the
high case fatality rate associated with primary septicemia and
severe wound infections. Case reviews have shown a median time
period from hospital admission to death of 48 hours or less (2,4,
7). This emphasizes the limited effectiveness of treatment and the
importance of prevention.
Host Susceptibility
Early case control studies and case reviews revealed a
population-at-risk for vulnificus infection with a striking
profile. Over 70% of patients were male, more than 70% were white,
and greater than 90% were over age 40 (7,8).
Primary septicemia and gastroenteritis, the two ingestion
syndromes, are both associated with consumption of raw or
undercooked seafood, particularly oysters. Eighty to 90% of cases
with an ingestion syndrome consumed raw or undercooked oysters in
the week prior to illness (1,7,8).
Host susceptibility plays an important role in primary
septicemia and severe wound infection. Almost all patients with
primary septicemia have a pre-existing, chronic illness. Although
people with chronic illness do not appear to be at greater risk for
acquiring wound infections, once infected, they are at higher risk
for spread of infection to the bloodstream and possibly death.
Liver disease and chronic alcoholism are the two underlying
conditions most frequently associated with severe vulnificus
infection. Between 60 and 75% of cases with primary septicemia
have pre-existing liver disease or report heavy alcohol
consumption. Another 15 to 20% of cases of primary septicemia
occur in persons with iron storage disease, such as hemochromatosis
or hemosiderosis, and in patients with chronic diseases of the
immune system, such as diabetes, cancer, or kidney disease, or in
those who take immunosuppressive drugs, such as steroids (1,7,8).
Host factors associated with increased susceptibility to
primary septicemia among patients with pre-existing illness,
particularly liver disease, are not completely understood.
Patients with liver disease may have a decreased ability to remove
vulnificus from the blood due to shunting of the blood around the
liver, defective white blood cell function, or reduced humoral
immunity. In addition, patients with alcohol-related liver
disease, such as cirrhosis, as well as patients with iron storage
disease have increased availability of iron in the serum through
increased transferrin saturation, which markedly enhances the
growth of vulnificus (10).
Epidemiologic Surveillance
Prior to 1988, vulnificus was not a reportable disease, except
in a very few states, such as Florida. In 1987, four states along
the Gulf Coast--Alabama, Florida, Louisiana, and Texas--along with
the FDA and CDC, proposed a regional surveillance system to report
vulnificus infections. This regional surveillance was initiated in
1988. One year later, the system was expanded to include all
Vibrio species. Shortly afterward, Mississippi became the fifth
state to join the coastal surveillance system. The information
presented here is a result of this unique surveillance
collaboration.
Collection of the Gulf Coast surveillance data is a multi-
stage process. When vulnificus is isolated from a patient's
specimen, the case is reported to state health officials. The
health department staff then conducts a detailed case investigation
on the patient's clinical course, underlying illnesses, and
potential sources of infection. Information regarding seafood
consumption and exposure to seawater in the week before illness is
obtained. Data concerning illness among eating companions are also
collected. After the state completes its investigation, the FDA is
notified. A traceback of any implicated shellfish is initiated.
The data presented today are based on information from the
state investigation forms and the FDA traceback reports for cases
occurring between 1988 and 1993. This summary is based on an early
preliminary analysis of the vulnificus surveillance data.
Geographic Distribution of Cases and Case Fatality Rates
Between 1988 and 1993, 138 cases of vulnificus infection were
reported through the Gulf Coast surveillance system (Table 1).
Florida reported the greatest number of cases, contributing over
40% of the region's total. Louisiana accounted for one-quarter of
cases, and Alabama and Texas contributed roughly 15% each.
Mississippi has reported only three cases since joining the
regional surveillance system. Louisiana reported the highest rate
of infection, and Mississippi and Texas the lowest.
The overall Gulf Coast rate for vulnificus infection is only
0.5 per 1 million persons; however, this rate significantly
understates the risk to persons in the high-risk populations, such
as those with pre-existing illness. A recent review of Florida
cases indicated that the risk of infection for adults with liver
disease who eat raw oysters is 80 times that of oyster eaters
without liver disease (3).
Of the 138 reported cases, 45 had a fatal outcome, resulting
in an overall case fatality rate of 42%. Among states, the case
fatality rate varied from 0% (0/3) in Mississippi to 57% (13/23) in
Texas. This mirrors the case mix reported by states. For
instance, Texas reported the highest proportion of cases of primary
septicemia and the lowest proportion of wound infections. If case
reports from outside the Gulf Coast are included with the regional
figures, a total of 145 vulnificus infections, including 48 deaths,
were reported to CDC between 1988 and 1993.
Temporal Distribution of Cases
The incidence of cases and deaths was highest during 1988, the
first year of regional surveillance. Since then, the number of
cases reported each year has remained steadily in the range of 15
to 30 per year. Based on this data, vulnificus infection appears
likely to remain an ongoing, persistent problem.
Mode of Transmission
Ingestion syndromes account for 58% of reported infections but
83% of fatalities. In contrast, wound infections accounted for
almost one-third of cases but only 8% of fatalities. (The
remaining 10% of infections and 8% of fatalities occurred in cases
with insufficient data to characterize the syndrome.) Among
patients whose outcome is known, 58% (40/69) of those with
ingestion syndromes died compared to 11% (4/37) of those with wound
infection (p <0.001).
Primary septicemia accounted for over 80% of ingestion cases
and almost 100% of ingestion fatalities. (In one fatality case the
type of ingestion syndrome was unknown.) The case fatality rate
for primary septicemia was 72%. Only ten cases of gastroenteritis
were identified and reported during the six-year period, and no
fatalities were associated with this syndrome.
Risk Factors
Data from the Gulf Coast surveillance system confirm the risk
factors associated with vulnificus infection in earlier studies.
In patients with primary septicemia, the at-risk population is
older (median age 56 years), 94% male, and 90% white (Table 2).
Among patients whose medical background was known, 75% of these 69
patients had a history of liver disease or chronic alcoholism, and
100% had at least one underlying illness. Of the 65 patients whose
food history was known, 95% ate raw oysters in the week before
their illness. Patients with wound infections had an almost
identical demographic profile: median age 57 years, 92% male, 87%
white. Thirty-two percent of the 47 patients whose medical history
was known had pre-existing liver disease. Surprisingly, 80% of
these patients had at least one underlying condition, a higher
percentage than previous studies reported.
The ten patients with gastroenteritis had a slightly different
profile (Table 2). They were young (median age 36 years), and only
50% were male. Only one patient had a history of underlying
illness. However, as with primary septicemia, a high percentage
(89%) of the nine patients whose medical history was known reported
consuming raw oysters in the seven days prior to their illness.
Among patients who acquired their vulnificus infection by
eating raw oysters, 74% of those with liver disease died and only
26% survived. Among patients without liver disease, the outcomes
were reversed. This association was statistically significant (p
<0.01).
Severity of Illness
For primary septicemia, illness was severe: 100% of patients
were hospitalized, 72% of those with known outcome died, and
deterioration was swift (Table 3). The median incubation period,
the time between ingestion and onset of symptoms, was one day.
Among patients who died, the median time period between hospital
admission and death was two days, confirming the rapid progression
to death reported in earlier studies. Those who survived had a
median hospital stay of 14 days. Given the speed and
aggressiveness of this infection, interventions aimed at prevention
rather than treatment appear to have the greatest potential for
reducing the number of cases and deaths due to primary septicemia.
Wound infections can also cause severe illness and swift
deterioration (Table 3). Eighty-nine percent of these patients
were hospitalized, and more than one in ten whose outcome was known
died. Like primary septicemia, the median incubation period was
one day. The time from hospital admission to death was only one
day for those who developed fatal wound infections.
Although gastroenteritis is the mildest of the three
syndromes, 50% of reported cases required hospitalization (Table
3). The incubation period was only one and one-half days. The
high hospitalization rate most likely reflects incomplete
identification and reporting of milder cases.
Vehicles Associated with the Ingestion Syndromes
Among persons with known food histories, 95% (70/74) of
patients with ingestion syndromes reported eating raw oysters in
the week before illness; 78% of these patients did not report
consuming any other seafood. Only one patient who ate oysters in
the week before illness reported eating only cooked oysters.
Another three patients reported consuming seafood which did not
include raw oysters in the week prior to illness. One of these
patients ate cooked shrimp, another cooked crab, and a third,
seafood gumbo containing cooked shrimp, crab, and fish. These
findings emphasize the importance of raw oysters as a source of
infection, and the relatively minor role of other seafoods in
transmitting vulnificus infections in the United States.
Infectious Dose
Fifty-six percent of infected oyster eaters with known
outcomes died. A dose-response was found between the quantity of
raw oysters consumed and fatal outcome. The number of raw oysters
consumed by patients who died varied from one to a pint of shucked
oysters. It is important that consumption of even one oyster
proved fatal.
This raises the question of infectious dose. In high-risk
populations, infection, even fatal infection, has occurred after
only one oyster was eaten. (The number of vulnificus organisms
present in these oysters is not known, however.) Given the large
number of raw oyster consumers and the rarity of vulnificus
infection among healthy people, it may be that the infectious dose
varies among subpopulations.
Sporadic Illness
V. vulnificus causes sporadic disease. During the entire six-
year surveillance period, there was not a single outbreak of
vulnificus infection (i.e., two or more culture-confirmed cases
linked to a common meal or lot of oysters).
Sources of Oysters
Among the 60 patients with data on where their oysters were
obtained, 95% purchased their oysters at commercial establishments;
only 5% of oysters were acquired through informal systems, such as
self-harvesting or roadside stands.
FDA traceback reports reveal that Florida and Louisiana each
supplied 43% (13/30) of the oysters whose harvest site was
identified through traceback. Other Gulf Coast states produced
only 13% (4/30) of the oysters with an identified harvest site.
Seasonality
There is a marked seasonal increase in vulnificus infections
during the warm summer months. Eighty-eight percent of all
vulnificus infections occur between May and October. Wound
infections show the greatest concentration during the summer
months, with 94% of cases occurring between May and October.
Eighty-eight percent of cases of primary septicemia occur during
this time period. Gastroenteritis is least affected by
seasonality; 70% of these cases occur between May and October.
The marked seasonality of vulnificus infections provides a key
time period for intervention. If consumers could be persuaded to
avoid raw oysters during the warm summer months between May and
October, 85% of reported ingestion infections and 88% of their
associated fatalities could be avoided.
Consumer Knowledge
Are high-risk patients, particularly those with liver disease
and heavy alcohol consumption, aware of the hazards of eating raw
seafood? In a small Florida study, high-risk patients, including
those with liver disease, were interviewed to determine if they
were aware of the risks from eating raw seafood. Only 14% of
patients were aware, and only 7% received this information from
medical personnel (6).
Last year, CDC conducted a survey of liver transplant units
around the country to see whether patients were being educated
about the dangers of raw seafood. In fact, only 23% of transplant
coordinators themselves were aware of the hazard to their patients,
and only 16% of centers provided patients with information about
raw seafood (9). In a follow-up to this study, CDC is attempting
to inform transplant personnel about this hazard through articles
in trade journals.
Mandatory consumer alerts placed at the point of sale may
enhance public awareness, including that of individuals in the
high-risk groups. California, Louisiana, and Florida now require
these mandatory labels. Florida established these requirements
within the last year. Last month, CDC, which oversees vessel
sanitation, required consumer advisories on all cruise ships.
Conclusions Based on the Gulf Coast Surveillance Data
1. V. vulnificus infections are an ongoing, fatal problem. Data
from the Gulf Coast show no indication that these infections
are declining.
2. Persons at greatest risk for vulnificus infection are those
with liver disease and heavy alcohol consumption.
3. The severity of disease and the rapidity of its progression
suggest that prevention activities rather than treatment
should be the focus for intervention efforts. However, the
group at high-risk for vulnificus may be the most difficult to
access.
4. Raw oysters are the major vehicle involved in ingestion
syndromes. Other foods, including other raw seafoods, play
only a small role in the transmission of V. vulnificus
infection in the United States.
5. The majority of cases and deaths occur between May and
October, signaling an important time period for intervention.
If vulnificus-containing raw oysters were not consumed during
these months, approximately ten cases and eight deaths each
year could be prevented.
Recommendations
1. Target preventive measures rather than treatment.
2. Focus intervention on high-risk populations, such as those
with liver disease and heavy alcohol consumption.
3. Evaluate the effectiveness of a variety of interventions in
reaching these target groups as they may be difficult to
access.
4. Decrease raw oyster consumption during the warm summer months
between May and October. A variety of methods to achieve this
goal should be considered, including closing oyster beds
during this time, and posting consumer advisories.
5. Strengthen regional surveillance so that intervention effects
can be measured. For example, surveillance may aid in
assessing the impact of Florida's new consumer advisory
labels.
Table 1. Reported V. vulnificus infection by state, 1988-1993
Rate
State Cases % of total (per million)
AL 21 15 0.9
FL 57 41 0.7
LA 34 25 1.3
MS 3 2 0.2
TX 23 17 0.2
Total 138 100 0.5
**********
Table 2. Risk factors for V. vulnificus infection, Gulf Coast,
1988-1993
Primary Wound
septicemia infection Gastroenteritis
Median age (yrs) 56 57 36
Male 94% 92% 50%
White 90% 87% 86%
Liver disease 75% 32% 0%
Any chronic disease 100% 80% 10%
Ate raw oysters 95% --- 89%
**********
Table 3. Severity of V. vulnificus infection, Gulf Coast,
1988-1993
Primary Wound
septicemia infection Gastroenteritis
Median incubation 1 day 1 day 1.5 days
Hospitalized 100% 89% 50%
Dead 72% 11% 0%
Median hospital duration
fatal outcomes 2 days 1 day ----
nonfatal outcomes 14 days 7 days 4 days
References
1. Blake, P.A., M.H. Merson, R.E. Weaver, et al. 1979. Disease
caused by a marine vibrio: clinical characteristics and
epidemiology. N. Engl. J. Med. 300:1-5.
2. Bonner, J.R., A.S. Coker, C.R. Berryman, et al. 1983.
Spectrum of vibrio infections in a Gulf Coast community. Ann.
Intern. Med. 99:464-469.
3. Centers for Disease Control and Prevention. 1993. Vibrio
vulnificus infections associated with raw oyster consumption--
Florida, 1981-1992. Morbid. Mortal. Weekly Rep. 42:405-407.
4. Chuang, Y.C., C.Y. Yuan, C.Y. Liu, et al. 1992. Vibrio
vulnificus infection in Taiwan: report of 28 cases and review
of clinical manifestations and treatment. Clin. Infect. Dis.
15:271-276.
5. Hollis, D.G., R.E. Weaver, C.N. Baker, et al. 1976.
Halophilic Vibrio species isolated from blood cultures. J.
Clin. Microbiol. 3:425-431.
6. Johnson, A.R., C.R. Anderson, and G.E. Roderick. 1989. A
survey to determine the awareness of hazards related to raw
seafood ingestion in at-risk patient groups. Proc. of the
13th Annu. Conf. Trop. Subtrop. Fish. Technol. Soc. Am. Gulf
Shores, Alabama, October, 1988. Gainesville, Florida:
University of Florida Sea Grant College Program, 1989.
7. Klontz, K.C., S. Lieb, and R.A. Gunn. 1988. Syndromes of
Vibrio vulnificus infections: clinical and epidemiological
features in Florida cases, 1981-1987. Ann. Intern. Med.
109:318-323.
8. Tacket, C.O., F. Brenner, and P.A. Blake. 1984. Clinical
features and an epidemiologic study of Vibrio vulnificus
infections. J. Infect. Dis. 149:558-561.
9. Tuttle, J., S. Kellerman, and R.V. Tauxe. 1994. The risks of
raw shellfish: what every transplant patient should know. J.
Transpl. Coord. 4:60-63.
10. Wright, A.C., L.M. Simpson, and J.D. Oliver. 1981. Role of
iron in the pathogenesis of Vibrio vulnificus infections.
Infect. Immun. 34:503-507.
Questions and Answers
Q. Ms. Carolyn Smith-DeWaal, Public Voice: Your data show that
older people are more likely to get primary septicemia, but
younger individuals are more likely to get gastroenteritis.
Is that accurate from what you've shown us, or how would you
explain some of the differences?
A. Dr. Whitman: Yes. The median age for people with primary
septicemia is 56, and for gastroenteritis it is 36. Looking
at it a different way, just to get a sense of the
distribution, more than 80% of people with primary septicemia
were over the age of 40. The reverse was true about
gastroenteritis; just about 80% were under age 40.
Q. Ms. Smith-DeWaal: Is that because the older population is
most likely to have developed these underlying conditions that
might not be evident in the younger population?
A. Dr. Whitman: No. I don't believe that to be the case. We
looked at age to see whether it was independently associated
with infection, and it was not.
Q. Ms. Smith-DeWaal: One other question. I came in a few
minutes late, but the first thing I heard was 95% of the
patients you found with septicemia had pre-existing
conditions, and I'm not sure whether that was wound cases or
raw oyster consumption cases. Did you do any review of that
5% that fell outside of the group with pre-existing
conditions?
A. Dr. Whitman: There were actually a couple of different
figures there. For people with primary septicemia, we were
actually surprised; 100% had an underlying condition. By
"underlying condition," we're talking about the conditions
that were described earlier; immune compromised, iron overload
disorders, liver disease. What was surprising was that in
wound infections, 80% had an underlying condition. That is
higher than usual. So those were the two figures, I think,
that we were talking about. In terms of that remaining 20%,
we haven't actually investigated those.
Q. Mr. Tom Billy, FDA: Do you have any other information about
the folks at risk, the profile of the white males, certain age
group? Is there any other information about them? Could you
characterize them as risk takers? Do you have any type of
information like that?
A. Dr. Whitman: We did not do a specific study. We looked at
surveillance data. The uniform reporting forms used by the
four states, for instance, had demographic data, but there
wouldn't be a way for me to assess risk taking. There was
nothing relating to occupation other than wound infections;
people in fishing and marine activities were at high risk.
But other than the demographic information, no.
Q. Mr. Billy: Do your data show any positive effect of the
point-of-sale warning either in the State of Louisiana or
California?
A. Dr. Whitman: California isn't within the regional system, so
we wouldn't have any way of looking at that.
Q. Mr. Billy: You have a general set of data as well. Was there
anything you could pick up from that?
A. Dr. Whitman: I would say no, not from what we have so far.
And I think that would be a real argument for increasing and
enhancing surveillance.
Q. Mr. Tom Schwarz, FDA: You mentioned consumption of shucked
oysters, and that is the first time I've heard of that. Is
there more than one occurrence where shucked oysters have been
involved, because most of the ones I've heard about, all of
them, in fact, have been shellstock.
A. Dr. Whitman: The reporting form gives people the option of
describing whether the oyster is served on the half shell,
whether it is shucked, etc. I tried to look at that data,
but, unfortunately, the data are very incomplete. There were
cases with shucked oysters, but the problem with that was
really basically incompleteness of reporting. A third of
these people are in shock, so trying to get that kind of
information is not easy. Relatives may know if someone went
to an oyster bar in the last week. You may be able to get
some information, but that level of information has been
difficult to obtain. Also, I think that sometimes people
filling out the forms have not understood it.
Q. Mr. Ken Moore, ISSC: Again, I just want to make sure that
I've heard you correctly. You said that only four deaths have
occurred from wound infections from 1988 to present?
A. Dr. Whitman: Four that were reported to us.
Q. Mr. Moore: Of the reported illnesses, were all of your data
obtained from patient specimens?
A. Dr. Whitman: Yes. In order to make the case definitions that
we had, in order to be considered a Vibrio vulnificus
infection, you had to have a specimen isolated from the
patient. In the case of primary septicemia, it had to be
blood. In the case of gastroenteritis, it had to be a stool.
In the case of wound infection, it could be either wound or
blood.
Q. Mr. Moore: You mentioned a dose-response relationship with a
significant p value. Could you elaborate on that just a
little bit? How did you calculate that?
A. Dr. Whitman: That was done through logistic regression.
Q. Dr. Glen Morris, University of Maryland and VA Hospital,
Baltimore: What were your endpoints? Was it death? Was it
nature and severity of illness or ...?
A. Dr. Whitman: It was fatality. It was death.
Q. Mr. David Heil, Florida Dept. of Environmental Protection: I
noticed you mentioned in your data between 1988 and 1993 from
the five Gulf Coast states that the overall infection rate was
0.5 per 1 million, and you mentioned that that certainly
underestimates the risk for high-risk individuals. It would
certainly conversely overestimate the risk for non-high-risk.
Is that correct?
A. Dr. Whitman: Yes.
Q. Mr. Heil: Do you have any idea what the risk factors would be
for the high-risk and for the non-high-risk or what the
overall estimate of risk would be for the people at high risk
and not at high risk?
A. Dr. Whitman: We didn't look at that, but I think that one of
the other speakers today will actually address that.
Q. Mr. Larry Edwards, FDA: I think one of your consumption
slides mentioned that three other patients ate gumbo or
shrimp, and were infected with vibrios. Did they eat oysters
at all?
A. Dr. Whitman: No. That is why I singled them out. They
reported neither cooked or raw oyster consumption in the week
prior to illness.
Q. Mr. Edwards: Was there follow-up done on how those infections
could have occurred; cross contamination or something else?
A. Dr. Whitman: No; that wasn't done.
RISK ASSESSMENT AND CURRENT DATA
Florida Risk Assessment and Factors Associated with Risk
Dr. Gary Hlady
Epidemiology Program
Florida Department of Health and Rehabilitative Services
1317 Winewood Blvd., Building E
Tallahassee, Florida 32399
Vibrio vulnificus infections
Shellfishing is a major industry in Florida, and Florida has
the nation's largest number of reported illnesses and deaths from
consumption of raw oysters contaminated with Vibrio vulnificus.
The data and information presented here stem from infections
acquired only from the consumption of raw oysters. Included are
infections leading to both septicemia and gastroenteritis. To
estimate the risk consumers face for these disease syndromes
requires a large amount of information.
Oyster landings in Florida, and presumably oyster consumption,
have declined since 1981 (5). Recent harvests average a little
over two million pounds per year. In contrast, the proportion of
oysters harvested during the warmer months has gradually increased
over the same time period. This is important because nearly all
oysters harvested in Florida during these months contain V.
vulnificus.
From 1981 through 1993, 82 cases of infection attributable to
V. vulnificus from raw oysters were reported in Florida (4).
Thirty-eight of these (46%) were fatal, making V. vulnificus from
raw oysters one of the most deadly infections on record, and the
leading cause of reported deaths from food-borne illness in the
state.
The mean age of patients was 57 years, with ages ranging from
19 to 90 years old. Eighty-two percent were male. This averages
as six cases and three deaths each year. A slightly increasing
trend can be seen since 1981, though this may be partially due to
earlier reporting errors, as V. vulnificus infections were only
first required to be reported in 1981.
As noted previously, V. vulnificus infections are highly
seasonal, with 85% of cases occurring from May through October, and
93% from April through October (Figure 1). So far in 1994, four
additional cases (not included in these data) have occurred, two in
April and two in May. Since 1981, no cases have been reported in
January or February, and only one has been reported in December.
This seasonal pattern very closely resembles the pattern of
isolation of V. vulnificus from raw oysters.
This species appears to infect raw oyster consumers with
certain underlying conditions (1). Of all reported victims, 66%
were found to have underlying liver disease due to infection,
malignancy, cirrhosis, alcoholism, or metabolic disorders. Four
percent had only gastric illness characterized by low stomach acid,
such as peptic ulcer disease under treatment, pernicious anemia, or
surgical resection. Three percent had only diabetes as a risk
factor. But 50% of the patients with liver disease also had one or
both of these other two conditions. Some patients without liver
disease, diabetes, or gastric disease, had a variety of other
disorders of lesser frequency.
Not only are people with liver disease overrepresented among
cases, they also suffer greater mortality. The case-fatality rate
among patients with preexisting liver disease was 57%, more than
twice the rate of 25% of patients without preexisting liver
disease. The case-fatality rate among the 20 cases without any of
the three identified risk factors was 15%.
In 1988, the Florida Behavioral Risk Factor Study (2), a
random telephone survey of 1,483 adults, found that 25% of the
adult population in Florida ate raw oysters at least once a year,
and that 2.4% of these people were aware that they had liver
disease. These findings, along with our case reports and mid-1987
population figures, were used to calculate the average annual
incidence rates for illness and death due to V. vulnificus during
the 13-year period from 1981 through 1993 (Table 1).
The risk for V. vulnificus infection among raw oyster
consumers without liver disease was 0.8 cases per million adults
per year. This is a small but statistically significant increase
over the risk of 0.6 cases per million adults who did not eat raw
oysters, and who became infected through wounds or other means. At
74.1 cases per million adults per year, the risk of illness from V.
vulnificus among raw oyster consumers with liver disease was 88
times greater than that for raw oyster consumers without liver
disease. Because of the uncertainty about the true prevalence of
liver disease among the population, the 95% confidence interval for
this rate is broad, extending from 2.7 to 145 cases per million
adults. Still, the lowest limit of 95% confidence is greater than
the upper limit of 1.0 for the risk among people without liver
disease who eat raw oysters.
The relative risk for raw oyster consumers with liver disease
is even more striking with regard to the risk of death from V.
vulnificus. With a rate of 45.3 deaths per million adults per
year, the risk for raw oyster consumers with liver disease is 192
times greater than that for raw oyster consumers without liver
disease. There was no overlap of the 95% confidence intervals for
the risk of death between these two groups. The lower limit of the
95% confidence interval on the risk of death for raw oyster
consumers with liver disease was 1.6 deaths per million adults per
year. This was still almost 5 times greater than the upper limit
of risk among raw oyster consumers without liver disease, which was
0.3.
Of all oysters associated with illness from V. vulnificus, 85%
were purchased in restaurants, 11% were purchased in retail
markets, and 4% were purchased wholesale. All were harvested from
approved sites and none were self-harvested.
In recognition of the information presented so far, the State
of Florida now requires restaurants serving raw oysters to display
a warning message in plain view of all patrons. A similar message
must appear on all wholesale containers of oyster shell stock, and
retail markets will soon be held to a similar requirement. The
message addresses people with other chronic conditions and immune
disorders in addition to liver disease.
In conclusion, it is useful to emphasize the following points.
1) V. vulnificus is an important, preventable cause of severe
illness and death; 2) illness due to this species is seasonal, with
greatest incidence from May through October; and 3) people with
liver disease are at greatly increased risk (3).
Recommendations
1. Warn high-risk individuals nationwide, especially those with
liver disease, to eat oysters fully cooked. Gulf Coast
oysters are consumed throughout the United States, and cases
of infection have been reported from at least 14 states. A
national policy is needed to ensure that all consumers are
aware of their risk.
2. Until some method is developed to process oysters in a way
that will eliminate V. vulnificus yet preserve the raw
product, consideration should be given to restricting oyster
harvest to cold weather months, or to requiring that all
oysters harvested during warm weather months be fully cooked
before they reach the marketplace.
3. Reliable methods should be developed to predict or detect V.
vulnificus in raw oysters and safe tolerance or action levels
established to ensure that all oysters reaching the
marketplace meet consumer expectations for food safety.
One additional point is important to mention. V. vulnificus
is only one of several naturally occurring Vibrio spp. which
contaminate raw oysters and are not detected or eliminated by
current quality control procedures. These non-vulnificus vibrios
have caused almost three times as much reported illness as has V.
vulnificus, and have resulted in at least seven deaths in Florida
since 1981. It is notable that illness from non-vulnificus vibrios
in raw oysters occurs throughout the year, and unlike V.
vulnificus, no special risk groups for this infection have been
defined. These observations suggest that control measures for V.
vulnificus may not be as effective in preventing illness from other
vibrio species found in oysters. Any discussion of the safety of
raw oysters should consider these facts in addition to the threat
of V. vulnificus.
Figure 1. Month of onset for V. vulnificus infections associated
with raw oysters, Florida 1981-1993.
Table 1. Average annual incidence ratesa per 1,000,000 adults
18 and over and rate ratios (RR)b with 95% confidence
intervals (CI)c for illness and death caused by
V. vulnificus by host factors in Florida 1981-1993
Incidence Incidence
Host illness RR death RR
factors (95% CI) (95% CI) (95% CI) (95% CI)
Raw oyster consumer
Yes 2.60 1.32
(2.14-3.06) (1.09-1.55)
4.33 10.08
(3.39-5.27) (7.71-12.45)
No 0.60 0.13
(0.56-0.64) (0.12-0.14)
Raw oyster consumer with liver disease
Yes 74.05 45.30
(2.70-145.40) (1.63-88.97)
87.95 191.95
(10.70-165.20) (14.56-369.34)
No 0.84 0.24
(0.67-1.01) (0.19-0.29)
aBased on 1988 Florida Behavioral Risk Factor Survey, Florida
Office of Vital Statistics population data and V. vulnificus cases
reported to the Epidemiology Program, State Health Office, Florida
Dept. of Health and Rehabilitative Services.
bRate ratios may not compute with the rates given because of the
rounding effect.
cConfidence intervals calculated according to Cochran W. Sampling
Techniques. Wiley, 1977.
References
1. Centers for Disease Control and Prevention. Vibrio vulnificus
infections associated with raw oyster consumption --Florida,
1981-1992. Morbid. Mortal. Weekly Rep. 42:405-407.
2. Desenclos, J.A, K.C. Klontz, L.E. Wolfe, and S. Hoecherl.
1991. The risk of Vibrio illness in the Florida raw oyster
eating population, 1981-1988. Am. J. Epidemiol. 134:290-297.
3. Hlady, W.G., R.C. Mullen, and R.S. Hopkins. 1993. Vibrio
vulnificus from raw oysters: leading cause of reported deaths
from foodborne illness in Florida. J. Florida Med. Assoc.
80:536-538.
4. Mullen, Robert. 1994. Epidemiology Program, State Health
Office, Florida Department of Health and Rehabilitative
Services, Tallahassee, FL. Personal Communication, March 1,
1994.
5. Norris, Martha. 1994. Marine Fisheries Information System.
Florida Department of Environmental Protection, Tallahassee,
FL. Personal communication, June 7, 1994.
Questions and Answers
Q. Dr. Angelo DePaola, FDA: As I recall, you had 74 cases per
million people who ate raw oysters that had liver disease and
came down with vibrio infections. That would mean that almost
99.99% who were in the high-risk group who ate raw oysters did
not come down with vibrio infections even though you mentioned
that probably 100% of the oysters during this time of the year
are contaminated with Vibrio vulnificus. Why do you think
that the 99.9% did not become ill and that those 74 out of a
million did?
A. Dr. Hlady: We don't know much about that. It's clearly a
given that what we are talking about is a rare condition.
It's rare that we see people getting sick and dying from V.
vulnificus. But I think the important point is that the
relative risk for these people with liver disease is
astoundingly high compared to other people who may be sitting
at the oyster bar or sharing the same meal. We don't know.
It may be a dose-response phenomenon. It may have something
to do with host factors. But in face of relative risks on the
order of 100 to 200 times, any refinement of that definition
of a high-risk group--I don't know how much practical
importance that would have.
Q. Ms. Ivette Aguirre, FDA: You have data showing that in 1990-
1991 the cases of infection were real low. Then, in 1992, it
spikes real high. Is there a reason that you think it's high
risk individuals who are not eating in 1990 and 1991? Were
restaurants not serving raw oysters? I mean, any explanation
why there is a slump between those high peaks of vibrio
infections?
A. Dr. Hlady: Yes. We did have a particularly bad year in 1992.
In fact, that is what really sparked my interest in this whole
problem. We had 11 cases, and 9 of them died. I'm not sure
exactly why 1992 was such a bad year. It may have had
something to do with the concentration of V. vulnificus that
was in the oysters. Again, I think that Dr. Tamplin may have
some information to bear on that, looking at temperature and
salinity at the harvest sites.
Q. Mr. Wittman: You mentioned the confidence intervals on your
relative risk ratios for both disease and death. Could you
tell us what those were again?
A. Dr. Hlady: For illness among raw oyster eaters with liver
disease, the confidence interval was from 38 to 216. Among
raw oyster eaters without liver disease, the confidence
interval was from 0.76 to 1. For non-raw oyster eaters, the
confidence interval was from 0.56 to 0.64. For death, because
we are dealing with smaller numbers and more uncertainty, the
intervals were wider. For oyster eaters with liver disease,
the intervals were from 78 to 6,738. For raw oyster eaters
without liver disease, it was from 0.2 to 0.3. For non-raw
oyster eaters, it was from 0.12 to 0.14.
Q. AUDIENCE: Was this last death or illness?
A. Dr. Hlady: Death.
Q. AUDIENCE: Wasn't your mean 45?
A. Dr. Hlady: It's not a mean. It's a point estimate.
Q. Mr. Moore: In your data, you mention that you have been
tracking illnesses, or at least your data included illnesses
from 1981 to 1993. Did you use the same data from 1988 to
1993 that Dr. Whitman referred to?
A. Dr. Hlady: I believe so. I don't think they had any other
source.
Q. Mr. Moore: The data that you acquired from 1981 to 1988, was
that acquired from patient specimens?
A. Dr. Hlady: Yes. We had the same case definition. We had to
have a culture positive for the organism to call it a case.
Q. Mr. Moore: How did you go about determining the number of raw
oyster eaters?
A. Dr. Hlady: Through the random telephone survey that was done
in 1988, which was approximately the midpoint of our time
period here, and then extrapolating the proportion to the
population figures that we got.
Q. Mr. Moore: You said that V. vulnificus, I think you put it
this way, was responsible for the largest number of foodborne
deaths in Florida from 1981 to 1993.
A. Dr. Hlady: The largest number of reported deaths from
foodborne illness. Now, that requires that the death is
reported, that it is identified with a foodborne pathogen, and
that the food source is identified; yes. Out of something
like 45 deaths that met that definition, 35 of them were due
to V. vulnificus in raw oysters.
Q. Mr. Moore: Do you know what numbers two and three are?
A. Dr. Hlady: Of the 45 deaths that we had, 36 were from V.
vulnificus in raw oysters; 3 were from V. parahaemolyticus in
raw oysters; 2 were from V. cholerae non-O1 in raw oysters; 1
was from V. cholerae non-O1 in raw clams; 1 was from V.
cholerae non-O1 in clam chowder; 1 was from V. fluvialis in
raw oysters. The final death was organophosphate poisoning
from mustard greens.
Q. Mr. Moore: Let me rephrase my question. Do you know what the
number two and number three pathogens were that caused the
second largest and third largest number of foodborne illnesses
in the State of Florida over the same period of time?
A. Dr. Hlady: Not for illness. I just have the number for
deaths.
Q. Mr. Moore: Do you know what numbers two and three were for
deaths?
A. Dr. Hlady: Number one was V. vulnificus; number two was V.
parahaemolyticus; number three was V. cholerae non-O1.
Q. Mr. Moore: Of all foodborne deaths, those were one, two, and
three in the state of Florida?
A. Dr. Hlady: Of reported foodborne deaths, and, again, with an
identified foodborne pathogen and an identified food source;
that's right.
Q. Ms. Smith-DeWaal: What percentage of this population of
Florida falls into this high-risk group that you think is at
risk here?
A. Dr. Hlady: From the survey that was done in 1988, for people
with liver disease, the point estimate was 2.4%. That had a
confidence interval that went from 1 to 4%.
Q. Ms. Smith-DeWaal: Aren't those from among people who said
they ate raw oysters?
A. Dr. Hlady: Who were aware that they have liver disease and
ate raw oysters.
Q. Ms. Smith-DeWaal: Do you have a figure for how many people in
Florida might fall into that high-risk group of liver disease,
cancer, diabetes, AIDS?
A. Dr. Hlady: If you are going to throw together all of those
conditions, we can only estimate. I've heard estimates as
high as 25%.
Q. Ms. Smith-DeWaal: Of the cases you've documented in Florida,
how many were for people who were visiting or nonresidents of
Florida?
A. Dr. Hlady: I don't have that. There were several. I don't
have the numbers.
Q. Mr. Richard Thompson, Texas Dept. of Health and ISSC: Dr.
Hlady, you indicated that these other foodborne illnesses,
compared to the deaths related to V. vulnificus, had to be
identified and reported before you could count them; is that
correct?
A. Dr. Hlady: Yes.
Q. Mr. Thompson: Are these other pathogens that cause other
foodborne illnesses required to be reported in the State of
Florida as are vibrios?
A. Dr. Hlady: Most of them are. It depends on which ones you're
talking about. Some of the more recently identified ones like
the E. coli O157 are being added to the list, so we don't have
any historical data on that.
Q. Mr. Thompson: You indicated there was a very high relative
risk for the consumption of raw oysters and V. vulnificus
deaths for this at-risk group. Did you do any other risk
comparisons for other causes of death for the at-risk group
and come up with relative risk for other causes of death?
A. Dr. Hlady: No; I haven't.
Q. Mr. Thompson: And I have one ironic comment since Chuck
Kaysner got to make one. I have a gastric problem. I'm well
aware of it and my physician is well aware of it. Both of us
know the information and both of us eat raw oysters.
Q. Dr. George Hoskin, FDA: The issue about what physicians
consciously know and the information they have received does
require a little attention. With all the attention in the
news media, and at least two FDA Drug Bulletin publications
dealing with the issue, I wonder how many physicians haven't
received information and how many don't know they've received
such information.
CASE STUDIES
Mortality Scenarios and Pathogenicity Implications
Mr. Charles Kaysner
Seafood Products Research Center
U.S. Food and Drug Administration
22201 23rd Drive, S.E.
P.O. Box 3012
Bothell, Washington 98041-3012
Mortality Scenarios
What follows is an abbreviated review of two relatively recent
reports of primary septicemia caused by Vibrio vulnificus. Each
case is quite typical of those reported in the past 15 years or so.
This approach exemplifies the factors associated with the
infections and the information that was not obtained.
The two cases involved males: one aged 52 from the coastal
state of Alabama (1); the other aged 62 from the inland state of
Kentucky (4). Both individuals were alcohol abusers. Cirrhosis,
a medical condition frequently seen in other victims, was not
evident in Case one. The second patient's condition was
complicated by alcoholic hepatitis, and he had a prior
cholecystectomy. Alcohol abuse is reported in 65% of cases, and it
is likely the underlying condition causing each of these
individuals to be at risk.
Case one occurred during October; that information was not
available for Case two. The vast majority of infections generally
occur from April through November, peaking in August and September,
with a significant number occurring in the Gulf Coast states. Each
had consumed raw oysters prior to onset of symptoms. Case one ate
one-half dozen; no information was available for Case two. Other
reports indicate that as few as three individual oysters to several
dozen are consumed by patients. But usually consumption of one-
half to one dozen are reported and have been lethal. Each had
purchased the oysters from commercial sources. Case one obtained
the oysters from a retail market in Florida, and Case two at a
local restaurant in Kentucky. The onset of symptoms occurred two
days after raw oyster consumption for Case one, and just one day
following consumption for Case two. Generally, symptom onset has
been reported from one to three or four days following consumption.
Both patients sought medical attention two days after onset of
symptoms. When admitted to hospitals, both were experiencing
diarrhea, fever, chills, with pain and septic lesions of the lower
extremities. In many cases, a delay in seeking medical attention
for several days after symptom onset has been critical. Often,
patients are admitted with fulminating septicemia which has
progressed too far to respond favorably to antibiotic therapy.
Fever and chills occur in 85 to 90% of patients with primary
septicemia (3). Nausea is reported in 60% of cases. Diarrhea is
reported only in about 30% of cases. Only the patient in Case one
experienced vomiting, which has been reported in approximately 35%
of cases. One patient complained of abdominal pain, which is a
complaint in about 45% of reported cases. Hypotension is reported
in 45% of patients, but neither of these two patients presented
with an indication of septic shock.
A distinct condition of many reported V. vulnificus cases is
the presence of secondary lesions on the lower extremities, which
occurs in about 70% of patients. Each of these patients had such
lesions. The presence of secondary lesions seems to set apart this
disease from other bacterial foodborne infections. V. vulnificus
was cultured from the blood of both patients.
Aggressive antibiotic therapy was unsuccessful in Case one.
The patient died two days after admittance, an event that has been
reported to occur in one to four days. However, during treatment
of Case two, the lesions slowly resolved over a 10-day period. He
was released from the hospital after two weeks, 18 days after
consumption. One year later he was said to be doing well. This
individual now reportedly abstains from alcohol and avoids eating
shrimp and oysters.
The oysters that were consumed in these cases have generally
been traced to harvest areas of the Gulf of Mexico. Unfortunately,
tracing the origin of the suspect oysters is often difficult when
an illness has resulted several days after consumption. Also
unfortunate is that rarely do any of the suspect oysters remain for
testing to determine levels of V. vulnificus, and to possibly
obtain an indication of infectious dose. The suspect oysters in
these two cases were harvested anywhere from two to 17 days before
consumption, which indicates that the time from harvest to table
can vary greatly.
West Coast Epidemiology
No cases of V. vulnificus have occurred to date in either the
states of Washington or Oregon. Thus far there have been no
documented reports or cases of primary septicemia caused by V.
vulnificus from the consumption of the Pacific oyster, Crassostrea
gigas, a species different from the eastern oyster, C. virginica.
In California, from 1983 through August of 1993, a 10.5 year
period, 24 cases of V. vulnificus infection were reported to the
Department of Health Services, including 18 deaths (2, 5). Of
these deaths, 15 reported eating raw oysters before infection, and
16 reported alcoholic cirrhosis or other liver disease. Five cases
were reported in 1993 through August 13, and four of those patients
died. All five had consumed raw oysters. Only limited information
was available regarding the origin of the suspect shellfish in all
of these cases, but there are indications that the suspect
shellfish were harvested from the waters of the Gulf of Mexico.
Implications and Conclusions
1. A segment of the population is highly at risk.
2. Risk stems from the consumption of raw oysters from the Gulf
of Mexico.
3. The infectious dose for the individuals at risk is unknown; if
known, it could possibly aid in developing control measures.
4. Risk is greatest during the warmer months, when very high
levels of V. vulnificus are likely to be present in Gulf coast
oysters.
5. Consumption of a dozen raw oysters during the summer may
contribute a significant load of this pathogen to an
individual at risk. Such consumption is frequently
accompanied by an alcoholic beverage.
6. A very large population also is consuming the same raw oysters
without becoming ill.
7. The attack rate is normally one individual in a group who has
consumed oysters, very unlike the attack rate for other
foodborne gastrointestinal illnesses.
8. This attack rate may be a clue to the attending physician if
the pertinent information is obtainable.
9. Many at risk individuals are not being alerted to their risk.
References
1. Creasy, R. 1993. State Cooperative Programs, FDA Southeast
Region, Atlanta, GA, October/27/1993.
2. Department of Health Services. 1993. To eat or not to eat
(raw shellfish). Part I. Vibrio vulnificus infections in
California --a ten-year review. California Morbidity, DHS,
Berkeley, CA, August 13, 1993.
3. Oliver, J.D. 1989. Vibrio vulnificus. In: Foodborne
Bacterial Pathogens. M.P. Doyle, ed., Marcel Dekker, Inc.,
New York, NY, pp. 569-600.
4. Stahr, B., S.T. Threadgill, T.L. Overman, and R.C. Noble.
1989. Vibrio vulnificus sepsis after eating raw oysters.
Kentucky Med. Assoc. J. 87:219-222.
5. Wang, Mary. 1994. Food and Drug Branch, California State
Department of Health Services, Sacramento, CA, personal
communication.
PATHOGENICITY
Pathogenesis of Vibrio vulnificus
Dr. J. Glenn Morris
University of Maryland and
Chief, Infectious Disease Section
Baltimore Veterans Administration Hospital
10 N. Green Street
Baltimore, Maryland 21201
Factors Relating to Pathogenesis
A number of factors have been implicated as contributing to
disease caused by Vibrio vulnificus (6): extracellular products
such as cytolysins, proteases, phospholipases, and collagenases;
strains having the ability to utilize transferrin-bound iron;
encapsulation; and ability to stimulate a tumor necrosis factor
(TNF) response.
Extracellular Products
1. Cytolysin. The cytolysin produced by this organism is an
extremely potent protein. It lyses red blood cells and is
lethal for mice in nanogram quantities. It can cause skin
lesions in animals. However, genetic constructs which remove
or mutagenize the cytolysin gene from V. vulnificus do not
affect virulence, at least not in animal models (11). This
suggests that, while the cytolysin may play some role in the
disease process, it is not "the" factor, because deletion of
the gene does not seem to alter pathogenicity.
2. Protease/elastase. Similar to the cytolysin, deletion mutants
for the protease do not appear to decrease virulence.
3. Phospholipase/collagenase. Although data are limited, these
factors also do not appear to be essential for virulence.
None of the extracellular factors identified to date appear to
be essential for virulence. They may play a role in the disease
process, but they do not determine whether or not disease occurs.
Utilization of Transferrin-Bound Iron
Strains of V. vulnificus require iron for growth. Strains
that are virulent in animals can extract iron from transferrin (7).
Interestingly, the ability of these strains to grow in serum from
alcoholic patients was not found to be related to the total iron
present in the serum. That is, the absolute value of the iron was
not critical. What was critical was the percent saturation of the
transferrin.
In humans, transferrin is normally between 25 to 30% saturated
with iron. V. vulnificus does not grow in serum with this level of
transferrin saturation, or in sera with transferrin up to 65% iron-
saturated. In contrast, at higher saturation levels strains grow
almost exponentially.
How does this relate to alcoholics? It was found that
alcoholics do not have an absolute increase in their serum iron
levels. However, because of their underlying liver disease, they
have a decrease in the amount of their transferrin. Consequently,
although the iron remains constant, the transferrin levels drop,
and an increase in saturation levels result. Thus, a malnourished
alcoholic will have an increased level of iron saturation of
transferrin and, consequently, falls into the high-risk category
(1).
Encapsulation
Encapsulation is the most critical element of V. vulnificus
virulence. The ability to produce a polysaccharide capsule sets
this species apart and enables it to cause the devastating diseases
described.
Two morphologic forms of V. vulnificus isolates grow on Luria
agar. One is opaque and the other is translucent, and the
differentiation of these forms is very obvious to someone with
experience. Strains can shift spontaneously from one form to the
other at rates of approximately 10-5. The capsule is associated
with this phase variation. On electron microscopy, strains that
are opaque have a heavy capsule layer surrounding the organism,
whereas translucent cells show only a thin residual capsule.
A transposon mutant of V. vulnificus can be generated,
rendering it incapable of expressing the capsule entirely.
Encapsulation in V. vulnificus is controlled by genes which show
homology to the viaA complex of Salmonella typhi. Dr. Anita Wright
has now cloned many of the genetic elements involved in the
expression of the capsule.
There is no doubt that encapsulation is strongly associated
with virulence. It confers resistance to complement-mediated
bactericidal activity and to phagocytosis, and experimentally
decreases the 50% lethal dose (LD50) by more than 4 logs.
Encapsulation appears to be coordinately regulated with but
separate from the ability to utilize transferrin-bound iron.
In our laboratory, certain V. vulnificus strains have been
acquired and genetically constructed to investigate the roles of
encapsulation and the ability to utilize transferrin-bound iron in
virulence (5). These strains are briefly described (Table 1) as
follows:
a) Opaque strain MO6-24/0--a fully-virulent strain isolated
from a patient with septicemia, and its translucent phase
variant (MO6-24/T).
b) Transposon mutant CVD752--a translucent strain which is
locked into the translucent phase and which, as verified by
electron microscopy, does not appear to produce any capsular
material. It also has lost the ability to utilize
transferrin-bound iron.
c) Transposon mutant CVD737--a strain which is locked into
the translucent phase, but retains the ability to utilize
transferrin-bound iron.
d) Transposon mutant, CVD755--a control strain which carries
the transposon, but is fully encapsulated.
1. Serum resistance experiments. When V. vulnificus cells were
incubated for one hour in 65% pooled normal human serum, the
encapsulated (opaque) strain showed a minimal drop in counts. The
translucent form showed nearly a 4 log decrease in counts, and
isolates recovered from serum were opaque, suggesting that there is
selection for the encapsulated form. The CVD752 and CVD737 strains
cannot shift. As a consequence, they are killed immediately by
normal human serum.
2. Mouse virulence experiments. When the iron-loaded mouse model
was used, the LD50 for opaque strain MO6-24/O was less than 102.
With translucent strain MO6-24/T, the LD50 was 3 x 105. But, again,
what were recovered were opaque, encapsulated isolates, suggesting
that there is a selection for the opaque phenotype with animal
passage. For strains CVD752 and CVD737, LD50s were in the order of
106 to 107. The difference between these two strains lies in their
abilities to utilize transferrin-bound iron, and the similar LD50
results suggest that, while playing some role in host
susceptibility and virulence, utilization of transferrin-bound iron
is not that critical a factor in virulence (13).
The capsule appears to be a protective antigen. In
collaboration with Dr. Devi at FDA's Bureau of Biologics (2), the
capsular polysaccharide of strain MO6-24/O was conjugated with
tetanus toxoid and then used to immunize mice, both actively and
passively. This immunization afforded about 80% protection.
In collaboration with Dr. Arnold Kreger, the capsular
polysaccharide also was conjugated to purified cytolysin and
purified elastase. The immunogenicity of the cytolysin alone, the
elastase alone, and the capsular polysaccharide alone showed little
protective effect, about the level of the negative control. Some
degree of protection with the conjugates of the cytolysin and
elastase was evident, but much better protection was seen with the
tetanus toxoid.
3. Capsular typing. In collaboration with Dr. Alan Bush at UMBC
using HPLC, V. vulnificus capsular polysaccharide was isolated and
purified. The structural components of the capsule were then
identified. From Strain MO624, a capsule designated Type I
capsular polysaccharide was isolated and the complete structure
identified. The sugar components were isolated for 19 V.
vulnificus strains, from both clinical and environmental sources
(4), and the complete structures of four of these have been
identified (9, 10). The various sugar components which go into the
capsule polysaccharide comprise the capsular type. The results
suggest that certain capsular types are more common among clinical
isolates. For example, strains with Type I and Type II capsules
have been isolated exclusively from humans. However, a number of
other capsular types have been identified in clinical isolates,
ranging over a wide spectrum of different structures and sugar
contents (4).
Similarly, environmental strains range very widely in
structure and sugar content. Two environmental strains with the
same capsular type have been detected, both isolated at
approximately the same time from Chesapeake oysters harvested at
the same site, and are designated as Type 15 (4). This raises the
possibility that certain capsular types are present in the
environment at certain times. In summary, a great deal more typing
work will be required to delineate possible correlations between
capsule type and the ability to cause clinical illness.
Stimulation of TNF Response
With Drs. Wright and Jan Powell (8,12), the ability of V.
vulnificus strains to stimulate a tumor necrosis factor (TNF)
response has been examined. The TNF may be present in elevated
levels in the sera of patients with meningococcemia or other highly
virulent bacteria. Fulminant septicemia caused by V. vulnificus is
similar clinically to meningococcemia, with rapidly progressive
illness, leading to hypotension and death. It seems logical that
high TNF levels may be present in both diseases. Serum samples
from V. vulnificus victims have not been sufficient to determine
this possibility; however, a series of mouse studies have been
conducted using the encapsulated strain, MO624. Animals inoculated
with this organism showed a high level bacteremia which persisted
for 12 hours, followed by death within 24 hours. TNF levels
paralleled the levels of bacteremia. In contrast, CVD752 initially
caused bacteremia, but the levels of bacteremia fell off rapidly,
the animals survived, and their TNF levels remained low.
Factors that may trigger the TNF response also are being
investigated. The V. vulnificus lipopolysaccharide (LPS) appears
to play a decreased role in eliciting a TNF response. Other
cellular factors appear to contribute more substantively, and this
fosters intriguing speculation that the capsule itself may
contribute to the TNF response. If results can provide an
understanding of why the TNF response occurs in response to V.
vulnificus infection and what critical cellular factors stimulate
the TNF response, they may identify critical bacterial virulence
factors.
Relationship of Pathogenesis to Disease Occurrence
V. vulnificus is ubiquitous in estuaries and oysters during
warmer months. Still, overt disease due to V. vulnificus is
relatively rare. This may be due to a) the need for a high
infectious dose, b) differences in host susceptibility, and/or c)
differences in virulence among strains.
Infectious Dose
Epidemiologic data suggest that with higher levels of the
organism, there is a greater risk of infection. While human
infectious dose experiments are clearly impossible, there appear to
be correlations between gut levels of the organism in animals, as
determined by direct culture from gut contents, and the risk of
dissemination. Within an hour after oral challenge, if fewer than
106 V. vulnificus are present in the small intestine, the organism
will not disseminate. In contrast, if counts rise above this
level, the organism is likely to spread to internal organs and
blood. An approximate DD50 (the 50% disseminated dose) can be
discerned at various times after challenge. Our data suggest that,
at least in animals, V. vulnificus is not necessarily an organism
which rapidly disseminates from the gut; knowing the dose (amount
consumed) may make it possible to predict levels of risk.
Obviously, this must be taken in context with the immune
susceptibility of the host, as noted below.
Host Susceptibility
Susceptibility is clearly important in determining whether
disease will occur, and undoubtedly influences infectious dose.
Predisposing factors related to host susceptibility have been well
covered by the epidemiological data already presented, and include
patients with high levels of transferrin-iron saturation.
Why are alcoholics at increased risk? It might relate to
their various cytokine responses. IL6 is produced by the liver,
and alcoholics may not have a good IL6 response. There may also be
differences in TNF and other cytokine responses. However, all of
this is still very speculative; how alcohol specifically
contributes to susceptibility is still uncertain. Animal models
have not been that helpful in this regard. An alcoholic mouse
model has been developed; the mice are very, very alcoholic, and
very happy. They have prominent fatty livers, yet they show no
change in their susceptibility to V. vulnificus.
Serologic surveys of patient populations in the Chesapeake Bay
area have also provided interesting data. It can be reasonably
assumed that a population of shellfish workers incurs greater
occupational exposure to V. vulnificus and consumes more oysters
than does a population of Seventh-Day Adventists, who do not eat
shellfish, which is religiously proscribed in their diet. Antibody
levels to V. vulnificus differed significantly between these
groups, with higher antibody levels occurring among the shellfish
workers (5). This supports a role for preexisting immunity as
another element of host susceptibility. Quite possibly, people who
have consumed a lot of oysters or who have high exposure to oysters
may have acquired a degree of immunity. Clearly, anticapsular
antibodies can be protective antibodies. It is reasonable to
postulate that immunity can be a factor in the degree of host
susceptibility (3).
Virulence of the Organism
Finally, the virulence of the bacteria must be included. Many
who have worked in the field of pathogenesis have sought to
discover or fortuitously uncover the magic element, the single most
significant thing differentiating virulent strains of V. vulnificus
from those that are avirulent. The fact that even among high-risk,
oyster-eating, liver-diseased Floridians, more than 99.9% do not
get sick, suggests that there is some critical virulence factor
which is still unrecognized. Capsular type? Factors which
stimulate TNF release? At the moment, the answer is not known.
Table 1. Vibrio vulnificus: Role of capsule in virulence
Morph- Serum LD50
Strain ologya resistanceb (morphologyc)
MO6-24/O O 1.13 <102 (O)
MO6-24/T O 3.87 3 x 105 (O)
CVD752 T >6 3 x 106 (NRd)
CVD737 T >6 1 x 107 (T)
CVD755 T 0.95 <102 (O)
aOriginal strain morphology: O, opaque, encapsulated;
T, translucent.
bLog decrease in counts after incubation for one hour in
65% pooled normal human serum.
cMorphology of strains recovered from moribund animals.
dNone recovered.
References
1. Brennt, E.C., A.C. Wright, S.K. Dutta, and J.G. Morris. 1991.
Growth of Vibrio vulnificus in serum from alcoholics:
Association with high transferrin iron saturation. J. Infect.
Dis. 164:1030-1032.
2. Devi, S., U. Hayat, A. Kreger, C.E. Frasch, and J.G. Morris.
1994. E-40. Protection conferred by capsular conjugate
vaccines of Vibrio vulnificus in a murine model, p.150. In:
Abstr. 94th Gen. Meet. Am. Soc. Microbiol. American Society
for Microbiology, Washington, DC.
3. Fiore, A., U. Hayat, A.C. Wright, S.S. Wasserman, and J.G.
Morris. 1992. Infection with Vibrio vulnificus species
elicits an antibody response to capsular polysaccharide.
Clin. Res. 40:428A.
4. Hayat, U., G.P. Reddy, C.A. Bush, J.A. Johnson, A.C. Wright,
and J.G. Morris. 1993. Capsular types of Vibrio vulnificus:
An analysis of strains from clinical and environmental
sources. J. Infect. Dis. 168:758-762.
5. Lefkowitz, A., G.S. Fout, G. Losonsky, S.S. Wasserman, E.
Israel, and J.G. Morris. 1992. A serosurvey of pathogens
associated with shellfish: Prevalence of antibodies to Vibrio
species and Norwalk agent in the Chesapeake Bay area. Am. J.
Epidemiol. 135:369-380.
6. Morris, J.G. 1988. Vibrio vulnificus--A new monster of the
deep? Ann. Intern. Med. 109:261-262.
7. Morris, J.G., A.C. Wright, L.M. Simpson, P.K. Wood, D.E.
Johnson, and J.D. Oliver. 1987. Virulence of Vibrio
vulnificus: Association with utilization of transferrin-bound
iron, and lack of correlation with levels of cytotoxin or
protease production. FEMS Microbiol. Lett. 40:55-59.
8. Powell, J.L, A.C. Wright, A.M. Harris, D.M. Hone, and J.G.
Morris. 1994. B-279. The role of Vibrio vulnificus virulence
factors in the in vitro stimulation of TNF`, p.78. In: Abstr.
94th Gen. Meet. Am. Soc. Microbiol. American Society for
Microbiology, Washington, DC.
9. Reddy, G.P., U. Hayat, C. Abeygunawardana, C. Fox, A.C.
Wright, D.R. Maneval, C.A. Bush, and J.G. Morris. 1992.
Purification and structure determination of Vibrio vulnificus
capsular polysaccharide. J. Bacteriol. 174:2620-2630.
10. Reddy, G.P., U. Hayat, C.A. Bush, and J.G. Morris. 1993.
Capsular polysaccharide structure of a clinical isolate of
Vibrio vulnificus strain BO62316 determined by heteronuclear
NMR spectroscopy and high-performance anion-exchange
chromatography. Anal. Biochem. 214:106-115.
11. Wright, A.C., and J.G. Morris. 1991. The extracellular
cytolysin of Vibrio vulnificus: Inactivation and relationship
to virulence in mice. Infect. Immun. 59:192-197.
12. Wright, A.C., J.L. Powell, and J.G. Morris. 1994. The
relationship of TNF` response during experimental infections
in mice to the virulence of Vibrio vulnificus, p.78, B-280.
In: Abstr. 94th Gen. Meet. Am. Soc. Microbiol. American
Society for Microbiology, Washington, DC.
13. Wright, A.C., L.M. Simpson, J.D. Oliver, and J.G. Morris.
1990. Phenotypic evaluation of acapsular transposon mutants
of Vibrio vulnificus. Infect. Immun. 58:1769-1173.
Questions and Answers
Q. Dr. Gary Hlady, Florida Department of Health: Dr. Morris, you
said of the encapsulated and unencapsulated strains, you had
some laboratory strains that could not switch from one form to
the other. Do wild strains have the ability to convert from
one form to the other?
A. Dr. Morris: No. We have cloned out the genes that are
involved in the switching mechanism. It appears that there is
a percentage of isolates, clearly less than 50%, that lack the
necessary genes to let them switch back and forth. So there
is a subpopulation that probably cannot switch. The other
side of it is that when you look at isolates from the
environment, the majority of them appear to be in the
encapsulated form. So looking for the presence or the absence
of the capsule gives you a subset. Using the probes we now
have for the capsule genes, you can identify a subset of
organisms that would appear not ever to be pathogenic. But I
would have loved it if that constituted 99% of environmental
strains. Unfortunately, as I said, that is a minority of
environmental strains. Most of the environmental strains
appear to have everything they need to cause disease, at least
as far as we are able to identify at the present time. There
are certain capsular types that may cause a preferential
increase in clinical cases. We are a long way from
identifying exactly which types they are or understanding the
underlying pathogenicity.
Q. Mr. Chuck Kaysner, SPRC, ORA, FDA: Glenn, you were going
through the capsular structure and looking at 19 strains, and
only a few of those were well characterized. How many
different types of capsules did you see?
A. Dr. Morris: We found 15 capsular types among the 19 strains
that we examined. So, the bottom line is that there is a
plethora of capsular types. We have done further studies
since that time; what we seem to be seeing is that there are
a hell of a lot of capsular types out there.
Q. Mr. Kaysner: Do you have monoclonal antibodies to all of
those types or just a few?
A. Dr. Morris: We have monoclonal antibodies to one of those
types, the Type I that has been associated with clinical
disease. We are thinking about getting monoclonals to others,
if we can find the money.
Q. Dr. Ben Tall, CFSAN, FDA: Glenn, could you comment on the
skin lesions that are seen with the primary septicemia
patients? Is it a toxin tropism? Is it an organism tropism?
And could you relate it to other diseases?
A. Dr. Morris: I would say that skin lesions are seen in 70 to
80% of patients. So they are clearly a characteristic
component of that. In studies that I did at CDC and
subsequently, we found that in some instances, the skin
lesions contain the organism. In other instances, they do not
contain the organism, which suggests that the skin lesions may
be associated with a toxic-mediated process. If Arnie Kreger
was here, he would jump up and down and insist, absolutely,
that the cytolysin is the cause of the skin lesions. And I
think his data are fairly convincing. He has worked with
purified cytolysin preparations and it would appear that the
cytolysin is a major contributor to the occurrence of the skin
lesions. However, in the studies that we have done with
cytolysin-negative deletion mutants, we still see skin
lesions. So I can't completely answer your question. I think
some of the extracellular factors contribute to the occurrence
of the skin lesions. I think it is probably toxic-mediated in
some way. Cytolysin may play a role. The studies of the
deletion mutants make you back off a little bit from making
definitive statements.
Q. Dr. Sean Altekruse, FDA: Dr. Morris, is there any evidence of
human illness associated with consumption of raw shellfish
from the Chesapeake Bay?
A. Dr. Morris: We get cases in the Chesapeake Bay area. They
are not widely publicized. In an epidemiologic study that we
did out of North Arundel County Hospital, we found that the
incidence of V. vulnificus infections was approximately 0.5
per 100,000 population, which is comparable to that in coastal
areas of Florida and Louisiana.
Q. Dr. Altekruse: But is there any evidence that it is traced
back to Chesapeake Bay shellfish?
A. Dr. Morris: These are hard questions because, as you well
know, tracing shellfish can be extremely difficult. Anyone
who lives in the Chesapeake Bay area swears that they never
eat anything but Chesapeake Bay shellfish. The distributors
admit that probably the majority of what is being eaten comes
from Louisiana and Florida. We have not done the careful
backtracking with oyster lot tags in order to be certain. We
can put the same V. vulnificus that we get out of our Bay
oysters into animals and get a nasty disease. My instinctive
feeling is that yes, indeed, Chesapeake Bay oysters can cause
the same syndrome.
Q. Dr. Ron Siebeling, LSU: Did you or Arnie ever entertain the
possibility that the cytolysin might be a superantigen?
A. Dr. Morris: Yes, this thought has gone through our minds
several times. It might well be. We have been playing with
the cytolysin. We have tried cytolysin conjugates with our
capsular polysaccharide and, again, those weren't very
impressive. But I think the case is still open on that.
METHODOLOGICAL CONSIDERATIONS
Rapid Diagnostics and Environmental Screening
Dr. Anita Wright
Center for Vaccine Development
University of Maryland at Baltimore
10 South Pine Street
Baltimore, Maryland 21201
Vibrio vulnificus is a common inhabitant of the estuarine
environment, possibly existing as a symbiont of oysters. In order
to understand the ecology of this or any other bacterial species,
it is important and advantageous to have a method that will
directly count bacteria without enrichment or selection.
Enrichment methodologies pose problems with overgrowth of
irrelevant strains, and plating to selective media often inhibits
the target isolates, particularly if there is stress due to
starvation or suboptimal growth conditions.
Our method of choice for enumeration and reliable, rapid
identification involves direct plating to nonselective media and
then direct detection with a DNA probe for the cytotoxin-hemolysin
gene (1,3). This procedure involves colony lifts with filters from
spread plates to determine the total bacterial counts. The filters
are hybridized and tested with a nonradioactive, alkaline
phosphatase-labeled probe. Results provide both total bacterial
counts and culturable V. vulnificus counts from the same plate. V.
vulnificus, which are purple or dark brown, are readily detectable;
negative isolates are yellow. This procedure also allows us to go
back and recover target isolates from plates for further
evaluation. This oligonucleotide probe, developed in collaboration
with Ms. Gerri Miceli and Dr. William Watkins (FDA), is based on
the sequence for the structural gene of the hemolysin, and has 100%
specificity and sensitivity (3). The entire procedure takes less
than one day and offers several advantages, especially in terms of
sensitivity. Ten CFU/ml can be readily detected. Additionally,
filtration can be used to concentrate water samples and greatly
increase the sensitivity.
In oysters, where bacterial load is high, sensitivity is of
less concern, but selectivity (preventing growth of background
organisms) becomes more of an issue. The ability to detect V.
vulnificus in oysters can be enhanced by using more selective
media. Comparisons of various plating media available indicate
that thiosulfate-citrate-bile salts-sucrose (TCBS) agar may not be
the medium of choice. Compared to a nonselective medium, Luria (L)
agar, V. vulnificus enumerated from environmental samples were
three- to 30-fold lower on TCBS. On some occasions, though,
detection occurred on TCBS but not on L agar, presumably as a
consequence of decreasing the background to locate the "needles in
the haystack."
This procedure is extremely facile. One person can readily
process more than 50 filters in a day. The technique can also be
used on a large scale in areas that are essentially field-condition
laboratories. Studies using similar methods to detect V. cholerae
are ongoing in Peru in collaboration with a Dr. Guierra and Mr.
Augusto Franco. The study is evaluating 12 sites on a monthly
basis. Preliminary results from Peru show the advantage of using
probe techniques for V. cholerae as well.
Results from studies on Chesapeake Bay, in collaboration with
Dr. Rita Colwell and Russell Hill, show that the levels of V.
vulnificus in seawater and oysters are comparable to those reported
in the Gulf of Mexico during summer months. Densities of this
species in sea water decrease with increasing salinities
progressing down the bay, confirming the estuarine nature of this
organism. Also notable, in oysters harvested from the Chesapeake
Bay during August, V. vulnificus represents >10% of the total
heterotrophic load.
Preliminary data obtained in collaboration with Dr. Marilyn
Kilgen and Ms. Angie Corbin compared V. vulnificus enumerated by
the probe technique and the alkaline peptone water MPN methodology
using ELISA (2). For oysters harvested during summer months in
Louisiana, both methods gave comparable results. However, when
oysters and their bacterial flora are stressed during cold shock,
such as conditions encountered by maintaining samples on ice, the
probe methodology may afford greater sensitivity than techniques
using enrichment or selective media.
To summarize, using the DNA probe detection method, V.
vulnificus are found at concentrations of about 103/g of oyster and
102/ml of sea water in both the Chesapeake Bay and the Gulf of
Mexico. Our data indicate that methodology can significantly
influence environmental results, and DNA probe detection of
colonies from direct plating is a specific and sensitive method for
enumeration of V. vulnificus.
References
1. Morris, J.G., Jr., A.C. Wright, D.M. Roberts, P.K. Wood, L.M.
Simpson, and J.D. Oliver. 1987. Identification of
environmental Vibrio vulnificus isolates with a DNA probe for
the cytotoxin-hemolysin gene. Appl. Environ. Microbiol.
53:193-195.
2. Tamplin, M.L., A.L. Martin, A.D. Ruple, D.W. Cook, and C.W.
Kaspar. 1991. Enzyme immunoassay for identification of
Vibrio vulnificus in seawater, sediment, and oysters. Appl.
Environ. Microbiol. 57:1235-1240.
3. Wright, A.C., G.A. Miceli, W.L. Landry, J.B. Christy, W.D.
Watkins, and J.G. Morris, Jr. 1993. Rapid identification of
Vibrio vulnificus on nonselective media with an alkaline
phosphatase-labeled oligonucleotide probe. Appl. Environ.
Microbiol. 59:541-546.
ANIMAL MODELS
Dr. James D. Oliver
Professor of Biology, Department of Biology
University of North Carolina at Charlotte
Charlotte, North Carolina 28223
Animal models have been used to investigate Vibrio vulnificus
for about 17 years. Early on it was found that when injected,
whether subcutaneously, intraperitoneally, or intravenously, this
organism was lethal for most animals commonly used in laboratories,
such as mice, rats, hamsters, guinea pigs, and rabbits (7). It was
also reported in the early studies that ingestion as a route of
inoculation was not lethal. That is, just feeding animals the
bacterium did not produce death. One interesting result of using
animal models is their ability to differentiate the lethal
pathogenicity of several Vibrio species according to the route of
transmission. V. parahaemolyticus and most other Vibrio species,
for example, do not kill subcutaneously, whereas V. vulnificus
exhibits equivalent LD50s (about 106 in an adult mouse), whether it
is introduced subcutaneously or intraperitoneally (Table 1).
Epidemiological information related to this pathogen pointed
to the role of iron and liver cirrhosis. Drs. Anita Wright and
Linda Simpson made a remarkable observation on this point using the
mouse model, and produced the first experimental data on the role
of iron in V. vulnificus infection (12). Sets of five mice were
injected with various doses of viable cells in conjunction with
either saline or ferrous ammonium citrate. They observed the number
of mice killed to steadily decrease as the number of cells injected
along with saline decreased (Table 2). However, following iron
injection, all five mice died with as few as 102 cells injected,
and four of five were killed with only one cell injected (Table 2).
This represents a million-fold reduction in the LD50, an amazing
effect of iron.
Dr. Glenn Morris and associates subsequently reported this
same finding. It was soon determined that iron-overloading of the
mice enhanced the lethality of virulent, but not avirulent,
isolates of V. vulnificus by several orders of magnitude (6,10).
This same effect was found for some other vibrios, e.g. V. cholerae
and V. mimicus, but not for the other nonpathogenic Vibrio species
tested (6). Another iron-loaded mouse model was obtained by using
carbon tetrachloride to induce liver damage, which increases iron
serum levels due to the release of iron from the liver tissue (12).
This model provides the same type of results as the ferrous
ammonium citrate-injected mice. Further, lethality in these iron
models was observed after the mice ingested V. vulnificus (Table
3). Results comparable to those of the iron-overload mouse were
yielded by the suckling mouse model, which Reyes et al. used to
test the virulence of V. vulnificus strains (8,10). Thus, the
normal mouse, suckling mouse, and iron overload mouse models each
have proved useful in V. vulnificus studies on pathogenicity.
Dr. Gerald Stelma and associates reported results using an
iron-overloaded and immunocompromised mouse model.
Immunocompromised mice were produced using cyclophosphamide, an
agent which inhibits macrophage activity and phagocytosis. The
combination of iron-overloaded and immunocompromised mice revealed
a further increase in the lethality of virulent V. vulnificus
strains (9,10).
Another model which is not easy to use but shows some
interesting reactions is the ligated ileal loop model. This has
been a standard method for testing the pathogenicity of V. cholerae
for some time (2). Briefly, rabbits are anesthetized, the
abdominal cavity surgically opened, and the intestinal tract
ligated into segments of about 10 cm. Test organisms are injected
into the lumen of the intestine, which is then replaced; the
incision is closed, and the animals are allowed to recover for
about 18 hours. The rabbits are then sacrificed, and the ileum is
examined. A typical reaction for V. cholerae is seen as a large
swelling of the ileal loops caused by accumulation of fluid which
cannot pass out of the body due to the ligations. We found that
with V. vulnificus the rabbits did not survive for 18 hours, nor
did rats (Table 4) or other animals examined (7). The model also
has been used by Dr. Stelma and associates (11) to test enterotoxin
production by V. vulnificus strains. Examination of the blood of
dead animals showed that a significant bacteremia had occurred (7)
with V. vulnificus (Table 5). Examination of the intestinal
tissues of these animals using electron microscopy revealed that
tremendous destruction of tissues occurred within eight to ten
hours after injection of V. vulnificus (3). The normal appearance
of the villi, the microvilli, and the lamina propria, was
dramatically altered, and the numerous collagen bundles normally
present in the submucosa were severely affected. The tremendous
destruction of collagen was due to the powerful collagenase
produced by this bacterium. V. vulnificus cells penetrated down to
the intestinal wall and were indeed disseminated throughout the
tissue. Significantly, in such ligated ileal loops, there is no
longer any tissue-blood barrier. This mimics what occurs in human
infections, i.e., tremendous tissue destruction, tissue invasion,
and a greatly increased vascular permeability resulting in large
amounts of edema fluid (1).
Thus, animal models are an aid in determining the pathogenesis
and virulence of this organism. Factors such as hemolysins,
cytolysins, proteases, capsule, and lipopolysaccharide (LPS, a
component of the cell envelope of these bacteria, also termed
endotoxin) can also be investigated in these models for their
possible role in tissue invasion and death of victims.
LPS may be a critical virulence factor for V. vulnificus.
Although mice are generally not a reliable model for discerning the
effects of LPS, rats provide a means of investigating the effects
of this cell envelope component. Ordinarily, LPS is extracted,
purified, and injected intravenously. Results obtained in
collaboration with Dr. John Watts at the University of North
Carolina, Charlotte, showed that injection of V. vulnificus LPS (1
mg/kg body weight) into rats caused both heart rate and arterial
blood pressure to drop dramatically, symptoms typical of classic
endotoxic shock in patients afflicted with V. vulnificus (5).
Within about one hour, virtually 100% of the animals challenged
with purified LPS died. The same response occurs with LPS from
either opaque (encapsulated) cells or translucent (nonencapsulated)
cells. LPS causes the release of nitric oxide which, in turn,
causes tissue and endothelial damage. The characteristic drop in
blood pressure and heart rate can be reversed by intravenously
injecting an analogue of LPS, L-NMMA (20 mg/kg body weight) 10
minutes after V. vulnificus injection, which prevents the action of
LPS on cells (4).
Are these models of value in predicting human infectivity?
Some appear to mimic quite well the symptoms seen in human victims.
Are all V. vulnificus virulent? Using these models, most strains
of V. vulnificus appear virulent. But perhaps the situation is
analogous to that found for V. parahaemolyticus with the Kanagawa
phenomenon. With that bacterium, 99.9% of isolates encountered in
the environment are avirulent, even though animal models may
indicate otherwise. With this analogy, perhaps the 0.1% of V.
vulnificus strains are virulent, deriving their disease-causing
ability through a particular LPS type, as Dr. Glenn Morris of the
University of Maryland and Dr. Ron Siebeling of Louisiana State
University have been studying in V. vulnificus.
The studies described here have been supported, in part, by
grants from National Marine Fisheries Service (NA36FDQ271), the
National Institutes of Health (AI31216), the North Carolina Sea
Grant Program (R/MRD-24), and the U.S. Department of Agriculture
(91-37201-6877).
Table 1. Experimental mortality
Animal mortalityb
Bacterial Mice Rats Hamsters
challengea IP SC IV Ingestion IP IP
V. vulnificus 62/62c 34/34 8/8 0/12 10/13 9/9
(100%) (100%) (100%) (0%) (77%) (100%)
V. parahae- 19/24 0/24 6/6 0/12 NTd NT
molyticus (79%) (0%) (100%) (0%)
aInocula consisted of 0.1 ml salne with about 2 x 108 culturable
cells.
bMice were challenged by injection intraperitoneally (IP),
subcutaneously (SC), intravenously (IV), and by forced ingestion
with V. vulnificus and V. parahaemolyticus; rats and hamsters were
injected IP with V. vulnificus only.
cNumber of deaths/total number of animals challenged (% mortality).
dNT, not tested.
**********
Table 2. Effect of iron on V. vulnificus LD50
Log10 of Number of fatalities (n=5 mice)
inoculum PBSa Feb
8 5 -
7 4 -
6 1 -
5 0 5
4 0 5
3 0 5
1 -c 5
2 - 4
0 - 4
-1 - 0
aPhosphate buffered saline injected concurrent
with V. vulnificus challenge; LD50 = 6 X 105.
bFerrous ammonium citrate; LD50 = 1 X 100.
c -, not tested.
Table 3. Effect of CCl4 treatment on mortality in mice
V. vulnificus Challenge Mice mortalitya
inoculum route Untreated CCl4 treatedc
2.1 x 105 IPb 4/4 4/4
2.1 x 109 IP 2/8 9/10
2.1 x 103 IP 0/6 9/11
2.1 x 109 Oro-gastric 4/7 9/9
2.1 x 103 Oro-gastric 0/8 10/10
aNumber of mice dead/number of mice challenged.
bIP, intraperitoneal injection.
cA 0.1-ml volume of CCl4 (20% in olive oil) was administered
IP 24 h before challenge.
**********
Table 4. Mortality in ligated ileal loop studies
Mortalitya
Ileal loop inoculum Rats Rabbits
V. vulnificus 12/12 5/6
V. parahaemolyticus 0/6 1/4
Saline 0/3 0/2
aNumber of mortalities before 18 h/total number
tested.
**********
Table 5. Blood concentration of V. vulnificus in rabbits
receiving ileal loop injections
Rabbit Death timea (h) V. vulnificus/ml bloodb
1 9 3 x 106
2 13 6 x 104
3 -c 5 x 103
aTime elapsed after five ileal loops each were injected
with 0.2 ml saline containing 2 x 108 V. vulnificus cells.
bNumber of V. vulnificus determined from plate counts
performed within 15 minutes of death.
cAnimal sacrificed at 18 hours.
References
1. Bowdre, J.H., M.D. Poole, and J.D. Oliver. 1991. Edema and
hemoconcentration in mice experimentally infected with Vibrio
vulnificus. Infect. Immun. 32:1193-1199.
2. De, S.N., and D.N. Chatterjee. 1953. An experimental study
on the mechanism of action of Vibrio cholerae on the
intestinal mucous membrane. J. Pathol. Bacteriol. 65:559-562.
3. Dellinger, J.B., and J.D. Oliver. 1982. Enteropathogenicity
of Vibrio vulnificus in the rabbit ligated ileum. J. Elisha
Mitch. Sci. Soc. 98:105-118.
4. Elmore, S.P., J.A. Watts, L.M. Simpson, and J.D. Oliver.
1992. Reversal of hypotension induced by Vibrio vulnificus
lipopolysaccharide in the rat by inhibition of nitric oxide
synthase. Microbial Pathogen. 13:391-397.
5. McPherson, V.M., J.A. Watts, L.M. Simpson, and J.D. Oliver.
1991. Physiological effects of the lipopolysaccharide of
Vibrio vulnificus in mice. Microbios 67:141-149.
6. Oliver, J.D., R.A. Warner, and D.R. Cleland. 1983.
Distribution of Vibrio vulnificus and other lactose-fermenting
vibrios in the marine environment. Appl. Environ. Microbiol.
45:985-998.
7. Poole, M.D., and J.D. Oliver. 1978. Experimental
pathogenicity and mortality in ligated ileal loop studies of
the newly reported halophilic lactose-positive Vibrio species.
Infect. Immun. 20:126-129.
8. Reyes, A.L., C.H. Johnson, P.L. Spaulding, and G.N. Stelma,
Jr. 1987. Oral infectivity of Vibrio vulnificus in suckling
mice. J. Food Prot. 50:1013-1016.
9. Stelma, G.N., Jr., and A.L. Reyes. 1986. Relationship of
specific virulence factors to pathogenicity in Vibrio
vulnificus. Abstr. Annu. Meet. Am. Soc. Microbiol.
10. Stelma, G.N., Jr., A.L. Reyes, J.T. Peeler, C.H. Johnson, and
P.L. Spaulding. 1992. Virulence characteristics of clinical
and environmental isolates of Vibrio vulnificus. Appl.
Environ. Microbiol. 58:2776-2782.
11. Stelma, G.N., Jr., P.C. Spaulding, A.L. Reyes, and C.H.
Johnson. 1988. Production of enterotoxin by Vibrio
vulnificus. J. Food Prot. 51:192-196.
12. Wright, A.C., L.M. Simpson, and J.D. Oliver. 1981. Role of
iron in the pathogenesis of Vibrio vulnificus infections.
Infect. Immun. 34:503-507.
THE VIABLE BUT NONCULTURABLE STATE
Dr. James D. Oliver
Professor of Biology, Department of Biology
University of North Carolina at Charlotte
Charlotte, North Carolina 28223
Most Vibrio vulnificus infections occur in the summer months.
In fact, it is very difficult to isolate V. vulnificus from water
or oysters in the winter. Indeed, it appears to be absent during
this period (11). Why does this bacterium appear to die off or
disappear in the winter, and why does it reappear in the summer
months? This cyclic phenomenon likely involves the viable but
nonculturable (VBNC) state of V. vulnificus. Nonculturability and
resuscitation are not phenomena unique to this species. Many
species of bacteria have been shown to enter this nonculturable
state (12). The list includes species from the genera Aeromonas,
Campylobacter, Enterobacter, Escherichia coli, Klebsiella,
Legionella, Salmonella, Pseudomonas, Shigella, and a number of
Vibrio species. The VBNC state is induced by a variety of factors
depending on the species, including light, high and low
temperature, salinity changes, and nutrient deprivation. For V.
vulnificus, the VBNC state is induced by cold stress (5, 17). This
may, in part, explain why the organism seems to disappear in the
winter, and why infections in the winter are relatively rare. It
is known that the bacteria are still there in the winter, but they
simply are in a nonculturable state. To understand the ecology of
this species, then, it is imperative that we understand more fully
what happens to bacteria in the VBNC state. We know more about the
VBNC state of V. vulnificus, however, than any other bacterium, and
thus V. vulnificus is an excellent model for studying this state
(12).
Placed in an artificial sea water microcosm and held at
temperatures between 10oC and 30xC, the culturable cell numbers, as
determined by plate counts, show that V. vulnificus levels
gradually decline over about a two-month period (17). At these
temperatures the cells appear able to remain culturable for long
periods of time. However, at 5xC, a dramatic decrease occurs, and
after 40 days plate counts suggest that no viable cells remain. In
fact, the classic microbiologist would assume that these bacteria
are dead. Yet direct viable counts (by both the INT method and
that of Kogure et al. (4)) indicate that very large numbers of the
bacterial cells are still alive, albeit morphologically changed,
with little decrease in the total number of viable cells evident.
Typically, VBNC V. vulnificus cells become very small, decreasing
from about 2 to 3 fm long during log phase to cells on the order of
0.6 fm long, very much rounded up.
That the cells are indeed alive can be shown by a method
developed by Kogure et al. (4) which differentiates viable from
nonviable cells without the need to develop into colonies on a
plate. Simply the addition of small amounts of yeast extract as a
nutrient source, and the antibiotic nalidixic acid, enables this
distinction. Nonculturable cells, if alive, will begin to grow;
however, the nalidixic acid prevents them from dividing, and this
results in very elongated cells which are readily seen upon
microscopic examination.
These earlier studies were conducted using stationary phase
cells; later results were obtained using log phase cells. The age
of the cells used was influential on how rapidly the VBNC event
occurred (15). Whereas stationary phase cells require a month or
more to become nonculturable, log phase cells enter the VBNC state
in about one week.
Concomitant with all the morphological changes are some very
dramatic changes in the macromolecular composition of these cells.
Within 15 minutes of shifting a culture to 5xC, even though the
temperature decreases only a few degrees within that time, the
ability of cells to detect this change is exquisite, and dramatic
decreases in protein, RNA, and DNA syntheses commence (9). Within
only a couple of hours, macromolecular syntheses have reached a low
but stable maintenance level. There are also significant changes
in the major fatty acids of the cell membrane (5).
We have been examining the specific changes in the proteins
that V. vulnificus cells produce as they enter the VBNC state. Two-
dimensional electrophoresis showed that hundreds of proteins were
synthesized by log phase cells growing at room temperature. One
hour after a shift to 5xC, a number of prominent proteins were no
longer present. Even more interesting, some cold-shock proteins
began to be synthesized that were not present in the cells growing
at room temperature (7). These were synthesized in specific
response to the temperature downshift. We can take advantage of
these cold-shock proteins in the production of monoclonal
antibodies (described below).
In addition, using fluorescently labeled antibodies against
the capsule of V. vulnificus, VBNC cells were clearly observed to
retain their capsular material, even after 15 days at 5xC. That is
important because the presence of capsules contributes to the
virulence of this organism by aiding its resistance to human immune
defenses. Potentially then, such cells are still virulent. In
fact, our studies indicate that they do retain virulence.
Resuscitation
The revival process by which microorganisms recover from this
essentially dormant state into the metabolically active and
culturable state may be termed "resuscitation." Resuscitation is
a reversal of those metabolic, physiological, and genetic processes
that induced the VBNC state. Experimentally, VBNC cells of V.
vulnificus can readily reverse the process, becoming culturable
again after one to two days at room temperature (10). Thus, for V.
vulnificus, a temperature reversal is enough to allow VBNC cells to
become culturable again. Simply by shifting temperature from 20xC
to 5xC and back again, the cells can repeatedly go through this
cyclic conversion. The role this entry and exit into the VBNC
state plays in the ecology of this organism is intuitively obvious,
and this cyclic conversion of states by the cells can likely occur
over and over again. This is probably a survival mechanism in
response to environmental temperature changes and is a "stress"
response. In essence, VBNC cells of V. vulnificus resemble a Gram-
negative spore, essentially inert and waiting for the proper
conditions which allow them to become active.
We have shown that V. vulnificus can become nonculturable in
the laboratory, and that it can be resuscitated in the laboratory,
but is this a real phenomenon in the natural environment?
Experiments using the membrane chambers developed by Dr. Gordon
McFeters at Montana State University confirm that the entire
process does indeed occur with V. vulnificus.
Suspended in an estuary from floats, the chambers and their
bacterial contents become exposed to all the temperature, nutrient,
and salinity changes occurring naturally in the environment. To be
certain that the results will be based on the initial cells used,
a TnphoA insertion mutant (marked) strain, courtesy of Dr. Glenn
Morris, is used. This accomplishes two valuable things. It
imparts kanamycin resistance to the V. vulnificus cells, and this
antibiotic very effectively prevents the growth of many other
estuarine bacteria, including those found in oysters. If kanamycin
is incorporated into the culture medium, background growth is
significantly reduced, enhancing the selectivity of the method.
The transposon also enables these cells to produce alkaline
phosphatase, an enzyme which reacts with a specific colorimetric
reagent incorporated into the medium, resulting in brilliant blue
colonies. Cells recovered from the chamber and cultivated on this
medium can instantly be ascertained to be the same cells placed
into the chamber.
Such chamber experiments confirmed that the VBNC state occurs
in the estuarine environment (3). At temperatures below 13xC, very
little change in the total number of V. vulnificus cells occurred
(remaining at about 107 per ml). However, the cell population
became totally nonculturable by 20 days, but remained viable as
shown by the direct viable count method. The reverse experiment,
conducted in spring temperature conditions, showed that
resuscitation also occurs. Culturable cells, whether they are
opaque or translucent, were made nonculturable in the laboratory at
5xC. When placed into the chambers and suspended in estuarine
water at permissively warm temperatures, the cells recovered to the
culturable state within 24 hours. So, this is a real phenomenon
and it can occur very rapidly.
Two critical questions remain. Are VBNC cells capable of in
vivo resuscitation, and will disease result? That is, can a person
ingest an oyster containing only VBNC cells and develop the
disease? Using VBNC cells in phosphate buffered saline, at about
105 to 106 per ml, the virulence of these organisms was tested
using the iron-overloaded mouse model (13). Virulence was
determined throughout the temperature stress period which induced
the VBNC state, and during resuscitation. Dilutions were made and
tested in groups of three mice to detect the minimal lethal doses
and LD50s.
Of course, LD50 values are normally based on plate counts.
Eventually, with VBNC cells, there are no plate counts as the
organisms are no longer culturable. The resulting LD50
calculations thus yield data indicating that 0.02 culturable cells
are able to cause death (Table 1). Clearly, 0.02 cells cannot
cause animal death, yet the LD50 calculations based on culturable
cells were such that values generally fell well below one cell.
The reason these results were obtained is that the mice were
challenged with no culturable cells, but large numbers of
nonculturable cells, which resuscitated in vivo and produced
disease. The challenge initially contained only about seven or
eight VBNC cells, but this was sufficient to kill the mice. Thus,
VBNC cells retain virulence and are able to cause death. This then
suggests the question: Why do we not see a greater number of cases
of human disease among oyster consumers during the winter? Further
results from our experiments provide a clue to this critical
question.
Challenging mice with VBNC cells reveals a significant
increase in the calculated LD50 values. The values begin to change
after two days, and ultimately LD50s rise from about seven to eight
cells to 105 and greater by seven days. So, these cells still are
virulent, but they exhibit a greatly reduced virulence. The longer
they remain nonculturable at 5xC, the less virulent they become.
At ten days, however, when the cells were totally nonculturable,
challenged mice still died within 24 hours. Autopsy revealed that
cells of the marked strain could be recovered from the peritoneal
cavity and blood taken from the heart. These were cells derived
from the same TnphoA-bearing strain initially made VBNC by
refrigeration. This was presumptively confirmed by the blue color
of the colonies which developed, and absolutely confirmed by
conducting PCR on about 15 of the recovered strains.
There is a need to be able to detect V. vulnificus cells in
environmental samples whether in the culturable or nonculturable
state. The cytotoxin-hemolysin gene probe developed in Dr. Glenn
Morris' laboratory is 100% specific for identifying this species
(8), and using this probe to verify the identity of colonies, we
examined the specificity of CPC agar to isolate V. vulnificus from
oysters. CPC is very selective for V. vulnificus (6,14) and
colonies typical of this species are easily identified. From 15
oyster samples over a one year period, the flat, darker center
yellow colonies isolated on CPC were examined. Of the 1,041
colonies tested, 81.5% were confirmed as V. vulnificus (16) by the
gene probe (Table 2). Thus, the specificity of typical colonies on
CPC is very high. Of course, this method is only effective for
cells in the culturable state. Nonculturable cells do not grow on
any medium. We are currently developing the following detection
methods with the assistance of S-K funds.
We are examining the use of monoclonal antibodies to detect V.
vulnificus cells in the nonculturable state. However, there is no
reason to assume that monoclonals prepared against culturable cells
grown in the lab can be expected to reliably detect nonculturable
cells in the environment. Using one of the cold-shock proteins
seen on the two-dimensional gels of our VBNC cell extracts, 32
different monoclonal antibody preparations were developed. Of
these, six react specifically with nonculturable cells. They do
not react with culturable cells or starved cells, only with VBNC
cells (interestingly, one of the nonculturable monoclonals produced
was found to react with culturable cells). We are using these
monoclonal antibodies to detect V. vulnificus cells in oysters.
We are also using the polymerase chain reaction (PCR)
technique (1) to directly detect both viable and VBNC cells of V.
vulnificus. To successfully use this technique, however, assay
conditions must first be optimized (2). An optimizing kit
containing 16 different buffers is available from Invitrogen
Corporation for this purpose. Determining which buffers are
optimal is important, as buffers vary in their composition, pH, and
magnesium concentration, and the PCR reaction will only work under
certain of these conditions. The conditions appropriate for opaque
(encapsulated) and translucent (nonencapsulated) cells, culturable
or nonculturable, were tested in our lab. We have observed, for
example, that translucent, culturable cells yield suitable
amplification of the DNA when some, but not all, of the buffers,
are used. The same is true for opaque cells, but with differences.
Using nonculturable cells with the various buffers, amplification
of some translucent and some opaque cells occurs, but the patterns
on gels are not always the same. Thus, there are differences
between opaque and translucent cells, and differences between
culturable and nonculturable cells, and these can be discerned
using PCR with different buffers.
With respect to the suitability of buffers, we found that the
best buffer for differentiating culturable and VBNC cells was
Buffer K (Table 3). This buffer detects culturable but not
nonculturable cells. Thus, the possibility of detecting both
culturable and nonculturable cells in the environment seems very
promising. In fact, we recently used PCR to detect V. vulnificus
in unamended oysters. Through ongoing research we are also seeking
to verify the changes which appear to be occurring in the DNA of
cells becoming nonculturable.
Conclusions
The VBNC state is a widespread phenomenon among bacteria,
including V. vulnificus, which is induced by low temperature. Many
changes are involved, including morphology, metabolism, physiology
and, possibly, genetics. The capsule is retained during the VBNC
state, which has important implications related to virulence. VBNC
cells can resuscitate in the natural environment and they appear to
retain their virulence, although the virulence seems to be greatly
reduced. Finally, both monoclonal antibodies and PCR techniques
are likely to provide means of detection of VBNC cells in both
water and oysters. Whether the various animal models that have
been used to date are truly valid for determining virulence of the
various V. vulnificus isolates awaits further understanding of the
critical virulence factors exhibited by this pathogen.
The studies described here have been supported, in part, by
grants from National Marine Fisheries Service (NA36FDQ271), the
National Institutes of Health (AI31216), the North Carolina Sea
Grant Program (R/MRD-24), and the U.S. Department of Agriculture
(91-37201-6877).
Table 1. LD50s calculated for V. vulnificus cells as they enter
the viable but nonculturable state
Calculated LD50 valuesb
Plate Viable counts Total counts
Time countsa DAPIc AODCd Xe CTCf DVCg Xc
0 h 3.6 7.5x101 3.9x101 5.7x101 4.3 0.6 2.5
8 h 1.3 2.4x102 2.3x102 2.4x102 3.3 3.0 3.2
24 h 0.2 5.4x101 4.7x101 5.1x101 5.1 1.1 3.1
2 d 0.02 6.5x101 7.0x101 6.8x101 6.0 -h 6.0
4 d 0.5 1.3x104 5.0x103 8.1x103 2.4x103 1.8x102
6.6x102
7 d 0.039 7.4x105 - 7.4x105 2.7x105 - 2.7x105
aDetermined using iron-overloaded mice.
bGeometric mean of triplicate plate counts.
cDirect counts using DAPI.
dDirect counts using acridine orange staining.
eGeometric means of triplicate determinations.
fDirect viable counts using the CTC reduction method.
gDirect viable counts using the method of Kogure et al. (4).
hNo data.
Table 2. Use of CPC agar and an hemolysin gene probe to identify
V. vulnificus in oysters (Crassostrea virginica)
No. presumptive No. colonies
Oyster Sample Probe V. vulnificus positive with
sample date procedurea colonies on CPCb gene probec
1 8-92 S 8 7 (87.5)
2 10-92 S 15 12 (80.0)
3 10-92 S 1 1 (100)
4 10-92 S 22 18 (81.8)
5 10-92 D 9 9 (100)
6 10-92 D 42 40 (95.2)
7 6-93 S 6 4 (66.7)
9 7-93 D 62 50 (80.6)
10 7-93 D 64 56 (91.8)
11 7-93 D 104 66 (63.5)
12 7-93 D 51 37 (72.5)
13 7-93 S 51 31 (60.8)
15 7-93 D 606 518 (86.6)
Total 1041 849 (81.5)
aS, colonies on CPC agar tested after transfer of presumptive
positive colony to heart infusion agar; D, colonies on CPC agar
tested directly.
bFlat, cellobiose positive (yellow) colonies.
cGene probe for V. vulnificus hemolysin gene provided by Morris
and Wright (UMd); only strong signals counted as gene probe
positive (% positive).
**********
Table 3. Effect of different PCR buffers on amplification of
V. vulnificus DNAa
Culturable cells Nonculturable cells
Bufferb Opaquec Translucentd Opaquec Translucentd
A,I,M,O,P -e - - -
B-H,J,L,N +f + + +
K + + - -
aDNA amplification and detection of V. vulnificus hemolysin gene.
bBuffers as designated by Invitrogen Co.
cOpaque (encapsulated) strain of V. vulnificus.
dTranslucent (nonencapsulated) strain of V. vulnificus.
e-, no positive signal obtained from amplification.
f+, positive signal obtained from amplification.
References
1. Brauns, L., M. Hudson, and J.D. Oliver. 1991. Use of the
polymerase chain reaction in the detection of culturable and
nonculturable cells of Vibrio vulnificus. Appl. Environ.
Microbiol. 57:2651-2655.
2. Coleman, S.S. and J.D. Oliver. 1994. Optimization of
conditions for the polymerase chain reaction amplification of
DNA from culturable and nonculturable cells of Vibrio
vulnificus. Submitted for publication.
3. Hite, F., D.McDougald, N. Andon, and J. Oliver. 1994. Entry
into, and resuscitation from, the viable but nonculturable
state by Vibrio vulnificus in the natural environment. Abstr.
Annu. Meet. Am. Soc. Microbiol.
4. Kogure, K., U. Simidu, and N. Taga. 1979. A tentative direct
microscopic method for counting living marine bacteria. Can.
J. Microbiol. 25:415-420.
5. Linder, K., and J.D. Oliver. 1989. Membrane fatty acid and
virulence changes in the viable but nonculturable state of
Vibrio vulnificus. Appl. Environ. Microbiol. 55:2837-2842.
6. Massad, G. and J.D. oliver. 1987. New selective and
differential plating medium for Vibrio vulnificus and Vibrio
cholerae. Appl. Environ. Microbiol. 53:2262-2264.
7. McGovern, V.P. and J.D. Oliver. 1994. Induction of cold
responsive proteins in Vibrio vulnificus. Submitted for
publication.
8. Morris, J.G, Jr., A.C. Wright, D.M. Roberts, P.K. Wood, L.M.
Simpson, and J.D. Oliver. 1986. Identification of
environmental Vibrio vulnificus isolates with a DNA probe for
the cytotoxin-hemolysin gene. Appl. Environ. Microbiol.
53:193-195.
9. Morton, D., and J.D. Oliver. 1994. Induction of carbon
starvation proteins in Vibrio vulnificus. Appl. Environ.
Microbiol. 64:3653-3659.
10. Nilsson, L., J.D. Oliver, and S. Kjelleberg. 1991.
Resuscitation of Vibrio vulnificus from the viable but
nonculturable state. J. Bacteriol. 173:5054-5059.
11. Oliver, J.D. 1989. Vibrio vulnificus, pp 569-600. In:
Foodborne Bacterial Pathogens. Marcel-Dekker, New York.
12. Oliver, J.D. 1993. Formation of viable but nonculturable
cells, p 239-272. In: S. Kjelleberg (ed.), Starvation in
Bacteria. Plenum Press, New York.
13. Oliver, J.D., and R. Bockian. 1994. Virulence of Vibrio
vulnificus cells in the viable but nonculturable state.
Abstr. Annu. Meet. Am. Soc. Microbiol.
14. Oliver, J.D, K. Guthrie, J. Preyer, A. Wright, L.M. Simpson,
R. Siebeling, and J.G. Morris, Jr. 1992. Use of coistin-
polymyxin B-cellobiose agar in the isolation of Vibrio
vulnificus from the environment. Appl. Environ. Microbiol.
58:737-739.
15. Oliver, J.D., L. Nilsson, and S. Kjellerberg. 1991. The
formation of nonculturable cells of Vibrio vulnificus and its
relationship to the starvation state. Appl. Environ.
Microbiol. 57:2640-2644.
16. Sun, Y., and J.D. Oliver. 1994. Effects of GRAS compounds on
natural Vibrio vulnificus populations in oysters. J. Food
Prot. 5:549-558.
17. Wolf, P., and J.D. Oliver. 1992. Temperature effects on the
viable but nonculturable state of Vibrio vulnificus. FEMS
Microbiol. Ecol. 101:33-39.
Questions and Answers
Q. Dr. Steve Jones, University of New Hampshire: Jim, are there
any generalities about the buffers of PCR that would detect
viable nonculturable versus culturable? Is it the magnesium
or the pH or other?
A. Dr. Oliver: As I remember, J and K do differ either in the pH
or the magnesium.
Q. Dr. Chuck Kaspar, University of Wisconsin: Did you try
assessing culturability in nonselective broth versus
nonselective plates?
A. Dr. Oliver: Yes. For years, we and others have looked at
different ionic concentrations and diluted media and
everything imaginable. The obvious question is why do these
things not grow on a plate or in a broth. I talked with
people in Denmark who watched this in a time lapse and said
that they divide two or so times and then stop. But, as far
as I know, no one has ever gotten these cells, under any
condition over any length of time, to grow in broth or on
plates if they are fully nonculturable. One of the big
problems is you have to be careful that there is not a
culturable cell present in there. If you throw 99 percent
nonculturable cells and one culturable cell into the broth,
you have culturable cells that grow and you think you have had
resuscitation. You have to preclude that possibility. But,
no, we have never seen it.
Q. Dr. Kaspar: Did you happen to notice a longer length of
culturability when assessing viability using nonselective
broth?
A. Dr. Oliver: We don't use an MPN. But I talked to somebody at
the ASM meeting who said they used broth and thought they
could detect it a few days longer. And I wouldn't doubt that
a bit.
Q. Dr. Marilyn Kilgen, Nicholls State Univ.: Have you done any
studies resuscitating your nonculturables in the oyster
itself? I know you did studies in the little containers in
the water, but have you done this with the oysters and looked
at the populations of Vibrio and whether or not they would go
back and forth between the two forms in the oyster at the
different temperatures?
A. Dr. Oliver: We haven't looked at oysters so much, but we have
done some work with clams with Gary Rodrick in Florida and
have seen what appears to be both nonculturable and
resuscitation in clams. Mark Tamplin has done some work with
oysters which have no culturable cells, and after a day or two
at room temperature, they were pumping out large numbers of
Vibrio vulnificus cells. So that kind of thing has been done.
The problem with that kind of study is that once you sample
your oysters to see if there is any vulnificus there, you
can't do the study anymore. You have destroyed your sample.
So if you sample one oyster and say there is no vulnificus
there, and then sample another oyster, it can have a thousand
vulnificus. It is really difficult to look at
nonculturables in the oyster as far as resuscitation is
concerned.
Q. Mr. Bob Collette, SINA: I am trying to get a clarification on
one of your conclusions, that the nonculturable state may be
capable of causing infection. In showing that slide with
respect to your animal models, you did show that there was
some death at some point early on. But were those deaths or
infections due to the culturable cells that remain, or was it
truly the nonculturable portion of the inocula?
A. Dr. Oliver: The animals were dying with inocula that were
diluted so much that there could not have been more than
0.00005 culturable cells present. Statistically, it would be
impossible for any culturable cells to be present. Yet, there
were on the order of 104 nonculturable cells. So these animals
were dying strictly with nonculturable cells. But we
recovered culturable cells from them.
THE ECOLOGY OF VIBRIO VULNIFICUS
Dr. Mark Tamplin
3031 McCarty Hall, Home Economics Program
Institute of Food and Agricultural Sciences
University of Florida
Gainesville, Florida 32611-0310
Introduction
Understanding the ecology of microorganisms requires much
information produced over long time intervals. For Vibrio
vulnificus, data addressing its environmental occurrence and
levels, niches, survival, seasonality, influence of detection
methods, interactions with oyster tissues, environmental levels
related to human cases, and strain diversity, are important for
developing controls to protect at-risk consumers from disease.
Since 1987, much information has been gained about the ecology of
V. vulnificus.
Vibrio spp. are ubiquitous in estuarine and marine
environments. They can be isolated from many entities, depending
on the particular species. Most are not associated with fecal
pollution, unless epidemics occur, as with V. cholerae.
Vibrio spp. are metabolically diverse. They serve important
chemical, aquatic, physical, symbiotic, and commensal roles in
environments. They all produce chitinase, and attach to and
biofoul surfaces. Nutritionally, they play an important role in
the food web, being fed upon by marine ciliates and protozoa. Some
species are bioluminescent; others have been linked to various
predator-parasite interactions. As a consequence of evolution in
these environments, many enzymes (such as protease and
phospholipase) have evolved for sequestering nutrients from tissues
of oysters, fish, plankton, and other entities, but which can also
constitute virulence in humans. Thus, most of their interactions
with humans are purely accidental. Molecular evolutionary studies
show that some genes from Vibrio spp. have been passed to
terrestrial bacteria such as Escherichia coli and Salmonella.
The concentrations of Vibrio spp. generally, and V. vulnificus
specifically, in sea water are enhanced by elevated temperatures of
20 to 30xC, high pH, salinity, and dissolved organics. However,
some species such as V. cholerae, V. mimicus, and V. vulnificus
survive well in salinities lower than 10 ppt.
The typical seasonality of V. vulnificus in seawater,
sediment, and shellfish shows highest concentrations in the warm
summer months (14,16,17,21,26). Usually, V. vulnificus levels in
sea water reach about 1,000 per ml during summer months in the Gulf
of Mexico. In sediment and in oysters, levels can be as high as 105
to 106 per gram.
Salinity also has a strong influence on V. vulnificus
survival. Kelly (14) showed that in Galveston bay, V. vulnificus
lived in low salinities (7 to 16 ppt) and at temperatures >20xC.
More recent studies using improved isolation methods, show that
lower salinities tend to favor survival of this organism (22).
Isolations and descriptions of V. vulnificus have been made
from coastal environments all over the globe since the early 1980s.
In the United States, Kaysner and coworkers (13) reported a survey
along the West Coast and indicated that this species was present in
samples at a frequency of 5.9%. Many West Coast strains had low
LD50 values, much like those found for strains collected from the
Gulf of Mexico. O'Neill and coworkers (17) described the organism
in Maine and New Hampshire estuaries, finding relatively high
levels in the summer. DePaola (5) reported very high levels of V.
vulnificus in fish intestines. Previous work focused on seawater,
sediment, and oysters, but there is more to the estuarine
environment by far than just these niches. The intestines of fish
have high metabolic activity and a higher temperature than that of
ambient seawater. Since V. vulnificus ecology is to a large extent
controlled by temperature, the high concentrations of this species
in fish intestines may be important in reintroducing it into
estuaries when seawater temperature rises. Thus, it is not
necessary to explain the winter form of V. vulnificus as only
viable but nonculturable (VBNC).
It would seem that anywhere temperature and salinity are
permissive, V. vulnificus is likely to occur. Certainly this
includes the Gulf of Mexico, where the primary vector of human
disease, Crassostrea virginica, grows. Kaspar and Tamplin (12)
reported that at temperatures between 13 and 22xC, this species
thrives best in sterile sea water microcosms with salinities
between 20 and 25 ppt. However, a rifampicin-resistant strain
added to normal sea water containing natural biota showed a quite
different survival pattern (12). Here, V. vulnificus died off much
more rapidly than in sterile environments. Thus, extrapolations of
pure culture studies to the environment must be viewed with
caution. Interaction with other organisms, bacteriophage, and
other predators or competitors, likely plays a large role in V.
vulnificus survival.
The inability to detect culturable V. vulnificus during cold
winter months can be explained by low numbers of cold stressed
cells, or by cells transformed into a nonculturable state. Since
V. vulnificus can be thermally resuscitated from some cold water
oysters, this could simply be the resultant growth of very few
culturable cells present below culture-detectable limits.
Investigators have reported that winter oysters, containing no
culturable V. vulnificus, could be placed in 25xC seawater, and
within 24 hours, V. vulnificus could be detected at high levels
(21,23,24). In further experiments, Tamplin and coworkers initially
showed that nondetectable V. vulnificus could be found in winter
oysters from the Gulf of Mexico after storage at 25xC under both
wet and dry storage conditions. Again, does this represent
resuscitation of large numbers of VBNC cells, or is this the result
of growth by very few cells, beginning at levels below the
detectable limits of assays? The latter appears to be more likely
since Weichart et al. (27) reported that resuscitation of V.
vulnificus is not due to reculturability of the total viable V.
vulnificus population, but to growth of a very small number of
culturable cells in a seawater microcosm. This finding was also
reported for V. cholerae (18).
Microbiologists using enrichment broths and other techniques
many times assume that methods will recover and detect as few as a
single cell. However, this is often not the case. For example, a
single V. vulnificus cell in an oyster homogenate may not grow to
detectable levels due to inhibition by components in the homogenate
(e.g., other bacteria, enzymes) or the culture medium itself.
Further evidence indicates that the VBNC form of V. vulnificus
is not present in oysters exposed to cold temperature for long time
intervals. An experiment using oysters harvested from cold
Connecticut waters showed that V. vulnificus could not be
resuscitated when exposed to 25xC seawater for 48 hours, indicating
that V. vulnificus dies after prolonged exposure (22). In
contrast, V. vulnificus can be thermally resuscitated from Gulf of
Mexico oysters when exposed to low temperature for just a few
weeks, showing that it is present at low levels nondetectable by
current methodologies.
Infectious dose
In recent collaboration with Dr. Gary Hlady and Mr. David Heil
of the State of Florida, Tamplin and coworkers conducted a study
designed to determine environmental levels of V. vulnificus in
oysters associated with human infections. Beginning in 1991,
temperature, salinity, and levels of V. vulnificus in oysters,
seawater, and sediment were regularly determined at a specific
oyster harvest area in Apalachicola Bay, Florida. Later, V.
vulnificus infections were traced to oysters from Apalachicola Bay,
and environmental levels of V. vulnificus were noted. In general,
sampling had occurred within one to two days of the reported
harvest date of the implicated oysters. The V. vulnificus levels
were higher in oysters in 1992, when nine V. vulnificus mortalities
were recorded in Florida, compared to 1991 when four occurred.
Strikingly, deaths were associated with oysters when V. vulnificus
levels showed annual peaks. V. vulnificus levels in oysters
corresponding to fatalities were about 1,000/g. It is tempting to
think that these levels in the environment signify hazardous V.
vulnificus levels for at-risk individuals.
In May of 1994, oysters from Apalachicola Bay were implicated
in a V. vulnificus-associated fatality. Tamplin and coworkers
obtained oysters from the implicated lot at the restaurant, and
analysis showed that the V. vulnificus level in the oysters was
900/g, close to the 1,000/g level earlier mentioned. Using a
formula to estimate environmental V. vulnificus levels in the Gulf
Coast region, the calculated level of V. vulnificus was about
200/g. The victim consumed about one or two dozen oysters. The
oysters had only been out of the water one day before the fatal
meal, and air temperatures were not very high. This combination of
factors indicated that thermal abuse was less a factor than the
warmer months of the year. A tube of blood obtained from the
patient at admission, and before antibiotic therapy, showed more
than 300 million V. vulnificus/ml blood. The victim expired only
two hours after admission.
In other human cases of V. vulnificus disease, environmental
levels in oysters were approximately 200-300/g. The actual V.
vulnificus levels after oysters are harvested, transported, and
stored is difficult to ascertain. However, in some cases, the time
intervals between harvest and consumption were short, indicating
that temperature abuse and growth of V. vulnificus may be less of
a contributing factor than suspected, and that lethal doses may be
far lower than postulated in earlier years.
Obtaining oysters from restaurants implicated in cases has
been very difficult, as mentioned by Chuck Kaysner earlier in this
workshop. However, even if it were easier, calculating doses is
more complex than originally expected.
Dr. Chuck Kaspar and his coworkers conducted a study to
determine how many different strains of V. vulnificus may be
present in a single oyster. The use of the pulsed field gel
electrophoresis (PFGE) technique provided definitive DNA
fingerprints for V. vulnificus strains isolated from three oysters
(2). Results showed that one oyster contained >150 different V.
vulnificus strains, a second oyster had >100 strains, and a third
contained >50 strains. Thus, there is tremendous diversity within
the V. vulnificus population, and probably in every oyster. If
virulence varies significantly among such strains, application of
a required "safe" dose may have serious limitations. However, if
most strains have similar virulence, then hazardous V. vulnificus
levels may be predicted.
The relatedness of these isolates was investigated using
principle component analysis of PFGE patterns (2). It was found
that the strains did not cluster and that only a few strains were
similar or identical. This indicates that consumption of a dozen
oysters could result in the ingestion of >1,000 different V.
vulnificus strains.
How many of these strains are involved in septicemic
infections? Typically, clinical laboratories save only a single
isolate, which is not sufficient to indicate how many different
types may have entered the victim. One or perhaps hundreds of
different strains may be involved in infections. The answer to
this question is clearly important; it will require networking with
hospital laboratories, the first people to isolate the causative
bacteria, and subculturing not just one colony, but as many as
possible. Since many people in the population have underlying
liver disorders which place them at risk, and since so few cases
arise annually, it seems possible that pathogenesis may involve
very specific predisposing conditions, and multiple strains of V.
vulnificus.
Distribution and Occurrence: Effect of Salinity and Temperature
Recently, Tamplin and coworkers completed a study which
determined the frequency and abundance of V. vulnificus along the
U.S. coast (22). Approximately 20 different sites were surveyed
monthly by 15 collaborators. V. vulnificus was measured in oysters,
sediment, and seawater.
Data collected in Hawaii provided interesting insights about
V. vulnificus. In Hawaii, V. vulnificus was found in clams grown
at an aquaculture farm. Uniquely, Hawaii has a zero tolerance for
this opportunistic pathogen in any served food. Consequently,
aquaculture industries desired information about mechanisms to
control the presence of V. vulnificus.
On the west coast of the big island of Hawaii, the land is
virtually all lava. Here surface water enters lava tubes and flows
under the surface to the ocean. In some places near the coast,
ponds occur at the lava surface and are affected by ocean tides.
These are found uniquely on Fuji and Hawaii. The temperature of
the pond water is about 23xC year round, as is most coastal surface
water around Hawaii. In fact, because the temperature is constant,
it is possible to study the effect of salinity. V. vulnificus
could not be detected in this low turbidity water, but was present
in pond sediments.
Freshwater upwellings also exist along the island coastline,
and occur as water travels down through lava tubes and seeps
through sand at the shoreline. V. vulnificus was not detected in
these outpourings of fresh water or in the coastal waters near
shore where salinity is 35 ppt. However, in the sand near the
freshwater upwelling where low salinities were measured, V.
vulnificus was abundant at levels consistent with those found in
the Gulf of Mexico. Thus, salinity is a very controlling factor of
V. vulnificus ecology.
This United States survey produced a database of >2,000
records (22). The relationship between salinity, temperature, and
V. vulnificus levels was examined to determine if an algorithm
could be produced to predict V. vulnificus levels. The database
was separated into two geographical regions to obviate different
influences in United States coastal regions. Results provided
predictive curves based on different salinity and temperature, from
which densities of V. vulnificus were determined. Curves generated
for two regions, Northeastern and the Gulf of Mexico and South
Atlantic, predicted V. vulnificus densities in sediments, oysters,
and sea water. The predictive model assumes that salinity and
temperature are primary controlling factors. The p value for this
predictive index in the Gulf of Mexico is 0.0001. These models do
not apply to temperature abuse situations in a processing plant or
restaurant, but only to oysters present in the environment.
Another part of this multi-state study included examining
seasonal levels of translucent (low virulence) and opaque (high
virulence) morphotypes. In every estuary, opaque isolates were
usually found at ten-fold higher concentrations than translucent
types. Opaque isolates persisted into winter longer than
translucent forms.
Distribution in Oysters
In other studies, V. vulnificus levels in 20 individual
oysters from the Gulf of Mexico were determined (21). It was found
that if the temperature and salinity were favorable, V. vulnificus
was present in all specimens. In the summer, at least 75% of V.
vulnificus was associated with the intestinal tract of an oyster
(3,4). However, it was also found at significant levels in the
hemolymph, gills, and mantle. Finding the organism in the
hemolymph indicates it may be invasive to the oyster, although no
reports show disease caused by this species in adult oysters.
Capers and coworkers reported that in cold winter months, the
distribution of V. vulnificus in oysters changes (3,4). This
probably reflects V. vulnificus levels in sea water, since oysters
are filter feeders. As V. vulnificus levels in sea water decrease,
a higher proportion of organisms is found associated with adductor
muscle and mantle tissue. During winter, when analytical results
indicated that V. vulnificus was no longer culturable from
homogenates of whole oysters, analysis showed that low levels were
present in individual oyster tissues.
Many V. vulnificus found in the intestinal tract of oysters
are phagocytized by intestinal oyster hemocytes. These hemocytes
then pass through the epithelial wall and move into other tissues.
That may explain why low levels of this organism are found in the
hemolymph. Electron microscopy shows that V. vulnificus cells are
digested by oyster hemocytes (1); however, opaque strains are
phagocytized at a lower rate by oyster hemocytes than translucent
strains (7,8). About 50% of the translucent cells are phagocytized,
whereas only 10% of hemocytes contain opaque cells. Once
phagocytized, opaque and translucent cells are degraded.
Associations with Plankton and Other Marine Life
V. vulnificus, like many other types of marine bacteria, is
associated with plankton (16,21). Plankton are the most abundant
plants and animals in sea water, and many of them contain chitin.
V. vulnificus and other vibrios produce chitinase, an enzyme which
probably contributes to the degeneration of chitinaceous detritus
and the ecological balance in estuarine environments. Electron
micrographs clearly demonstrate the association of V. vulnificus
with exoskeletons shed by zooplankton. In general, V. vulnificus
is associated with more benthic species of plankton, arthropods,
crab larvae, and other organisms that interact with oysters (21).
As reported by DePaola (5), this species is harbored in the
intestinal tracts of fish, and higher levels occur in more benthic
fish species.
Methodology
Much work on improving V. vulnificus methodology has occurred
since 1988. The newly developed enumeration and detection
techniques and media greatly facilitate studies needed to pinpoint
factors involved in V. vulnificus ecology and human disease.
Highly specific methods for V. vulnificus include the DNA
probe of Morris and coworkers and antiflagellar antisera produced
by Siebeling and Simonson. The selective quality of the CPC agar
developed by Massad and Oliver is an excellent means to
presumptively identify V. vulnificus and eliminate background
organisms (15). Tamplin and coworkers described an enzyme
immunoassay which is widely used to confirm V. vulnificus (25).
Miceli and coworkers developed a direct plating medium to enumerate
V. vulnificus in oysters. A PCR method was used by Hill and
coworkers when sufficient numbers of cells (107) were attained by
enrichment (10).
Of broth culture techniques evaluated to date, alkaline
peptone water (APW) is the most satisfactory. Used in combination
with mCPC agar, this procedure has provided very reliable results.
Comparison of many of these media (VVA, VVE, CPC, and the MPN
procedure), using spiked oysters containing no other vibrio flora,
showed comparable recovery with the exception of VVA. Most of
these media can achieve adequate degrees of sensitivity (limits of
detection). Because it was originally developed for V. cholerae,
APW could be improved for its selective and recovery abilities.
Its poor selectivity may be due to the rapid drop in pH from 8.4 to
6.8 in just two hours. Many new and more rapid tests such as EIA
and PCR rely on APW to increase V. vulnificus to detectable levels.
At present, samples must be enriched 12 to 16 hours for optimum
recovery of V. vulnificus.
Improvements to APW would make many other applications more
feasible. The following modification of APW can optimize V.
vulnificus growth: increasing the peptone concentration from 1 to
5%; a pH of 8.0; incubation temperature of 35xC; 1% NaCl; and 0.08%
cellobiose (11). This modification results in higher V. vulnificus
growth within seven hours. For incubations longer than eight
hours, adding colistin to broth at 4 U/ml increases V. vulnificus
recovery more than 1 log10 compared to standard methods and reduces
the growth of other bacteria.
Sensitivity of detection methods must be considered for
detecting V. vulnificus. ELISA assays usually need approximately
r2,000 cells in a well. Co-agglutination requires >106/ml. Direct
plating and colony blots require 100 to 1000 CFU/ml. The PCR
reaction usually requires >104 culturable cells/ml for a positive
signal. Many labs have shown that the oyster matrix itself
inhibits polymerase enzymes. A rapid, dipstick-like method may be
a reasonable future goal. ELISA or immunoblot techniques may also
be appropriate.
Historically, very little research has been targeted at
differentiating V. vulnificus strains, except for two biotypes and
five taxonomic clusters (16). Virulence is recognized at the level
of presence of capsule, but nearly all V. vulnificus strains are
encapsulated. The laboratories of Siebeling and Morris have
investigated the composition of V. vulnificus capsule strains, and
have identified up to 12 strains (9,19,20). Tamplin and coworkers
have investigated the utility of differentiating strains based on
restrictions fragment polymorphism, i.e., digesting DNA and
"fingerprinting" strains using ribosomal RNA probes or PFGE (2).
In these applications, particular characteristics associated with
clinical versus environmental strains were shown. PFGE showed that
all 41 strains were unique; no two were identical. Thus, although
this technique is very sensitive, it is too sensitive for
successfully grouping strains by virulence, geography, or other
characteristics. PGFE can definitively show whether a clinical
specimen is derived from a suspect bag of oysters.
The BiologTM format simultaneously determines utilization of
95 carbon substrates, but not practical grouping by biotype (6).
BiologTM profiles showed seven clusters among the 41 strains, but
these groups were not related to clinical or environmental sources.
Some relationship to geography was discerned, however. For
example, one cluster was totally composed of isolates from Cedar
Key, Florida.
Restriction fragment polymorphism patterns by PFGE provided
interesting results for some strains (6). Two strains showed very
high (97%) homology. Each was isolated from two separate patients
admitted at different times to a hospital in Houma, Louisiana,
during the same year. Two Mississippi isolates were also very
similar, indicating some correlation with geography.
The utility of ribotyping, a technique using probes to 16S and
23S ribosomal RNA, has been used in investigations of other food-
borne illness outbreaks with considerable success, and appears to
be useful with V. vulnificus. Certain strains exhibited high
degrees of similarity. A large percentage of isolates from
Louisiana belonged to one ribotype termed RT2. Of the clinical
isolates, this covers 50 to 60% at a 90% similarity level. Some
new strains from Florida also belong to RT2. In collaboration with
Ron Siebeling at Louisiana State University, there appears to be a
correlation between RT2 and particular combination(s) of LPS and
capsular serotype. Using ten different polyclonal antibodies to
capsular antigens, Siebeling showed that 41% of all clinical
isolates were agglutinated, but only 4.5% of environmental isolates
reacted. Using whole-cell ELISA assays to LPS, 63% of clinical
isolates were reactive. Serovar 1 accounted for 25% of clinical
isolates, but detected <1% of environmental isolates.
The utility of these combined tests is very promising. A far
more extensive data set than currently available is needed, but
some positive associations appear to exist between ribotype, LPS,
and capsule polysaccharide. The Saltonstall-Kennedy cooperative
grant program has been very instrumental in the recent advances
described above. Continued federal and state support is critical
to controlling risks of V. vulnificus disease.
References
1. Aldrich, H.C., M. Martinez, E. Anderson, and M. Tamplin.
1992. Immunocytochemical and morphological studies of Gulf
oysters infected with Vibrio vulnificus and Vibrio cholerae.
Abstr. Southeastern Branch Am. Soc. Microbiol., p. 45.
2. Buchresier, C., V.V. Gangar, R.L. Murphree, M.L. Tamplin, and
C.W. Kaspar. 1994. Multiple Vibrio vulnificus clones in
oysters as demonstrated by clamped homogenous electric fields
(CHEF) gel electrophoresis. Appl. Environ. Microbiol.
61:1163-1168.
3. Capers, G.M., and M.L. Tamplin. 1991. Distribution of Vibrio
vulnificus in tissues of the eastern oyster, Crassostrea
virginica. J. Shellfish Res. 10:293-294.
4. Capers, G.M., M.L. Tamplin, A.L. Martin, and L.H. Hopkins.
1990. Distribution and retention of Vibrio vulnificus in
tissues of the eastern oyster, Crassostrea virginica. Abstr.
Annu. Meet. Am. Soc. Microbiol., p. 305.
5. DePaola, A., Capers, G.M., and D. Alexander. 1994. Densities
of Vibrio vulnificus in the intestines of fish from the U.S.
Gulf Coast. Appl. Environ. Microbiol. 60:984-988.
6. Gangar, V.V., R.L. Murphree, C. Buchrieser, C.W. Kaspar, V.
Garrido, J. Simonson, R.J. Siebeling, and M.L. Tamplin. 1993.
Biochemical, serological, and molecular characteristics of
environmental and clinical Vibrio vulnificus strains. Proc.
18th Annu. Conf. Trop. Subtrop. Fish. Technol., Williamsburg,
VA, p. 424.
7. Harris-Young., L., M.L. Tamplin, W.S. Fisher, and J.W. Mason.
1993. Effects of physicochemical factors and bacterial colony
morphotype on association of Vibrio vulnificus with hemocytes
of Crassostrea virginica. Appl. Environ. Microbiol. 59:1012-
1017.
8. Harris-Young, L., M.L. Tamplin, J.W. Mason, H.C. Aldrich, and
J.K. Jackson. 1995. Viability of Vibrio vulnificus in
association with hemocytes of the American oyster (Crassostrea
virginica). Appl. Environ. Microbiol. 61:52-57.
9. Hayat, U., Reddy, G.P., Bush, C.A., Johnson, J.A., Wright,
A.C., and J.G. Morris. 1993. Capsular types of Vibrio
vulnificus: an analysis of strains from clinical and
environmental sources. J. Infect. Dis 168:758-762.
10. Hill, W.E., S.P. Keasler, M.W. Trucksess, P. Feng, C.A.
Kaysner, and K.A. Lampel. 1991. Polymerase chain reaction
identification of Vibrio vulnificus in artificially
contaminated oysters. Appl. Environ. Microbiol. 57:707-711.
11. Hsu, W.Y., C. Wei, R.L. Murphree, and M.L. Tamplin. 1993.
Improved enrichment broth for recovery of Vibrio vulnificus.
Proc. 18th Annu. Conf. Trop. Subtrop. Fish. Technol.,
Williamsburg, VA, p. 423.
12. Kaspar, C.W., and M.L. Tamplin. 1993. The effects of
temperature and salinity on the survival of Vibrio vulnificus
in seawater and shellfish. Appl. Environ. Microbiol. 59:2425-
2429.
13. Kaysner, C.A., C. Abeyta, Jr., M.M. Wekell, A. DePaola, Jr.,
R.F. Stott, and J.M. Leitch. 1987. Incidence of Vibrio
cholerae from estuaries of the United States west coast.
Appl. Environ. Microbiol. 53:1344-1348.
14. Kelly, M.T. 1982. Effect of temperature and salinity on
Vibrio (Beneckea) vulnificus occurrence in a Gulf Coast
environment. Appl. Environ. Microbiol. 44:820-824.
15. Massad, G., and J.D. Oliver. 1987. New selective and
differential medium for Vibrio cholerae and Vibrio vulnificus.
Appl. Environ. Microbiol. 53:2262-2264.
16. Oliver, J.D., R.A. Warner, and D.R. Cleland. 1983.
Distribution of Vibrio vulnificus and other lactose fermenting
vibrios in the marine environment. App. Environ. Microbiol.
45:985-998.
17. O'Neill, K.R., Jones, S.H., and D.J. Grimes. 1992. Seasonal
incidence of Vibrio vulnificus in the Great Bay estuary of New
Hampshire and Maine. Appl. Environ. Microbiol. 58:3257-3262.
18. Ravel, J., I.T. Knight, C.E. Monahan, R.T. Hill, and R.R.
Colwell. 1995. Temperature-induced recovery of Vibrio
cholerae from the viable but nonculturable state: growth or
resuscitation? Microbiology 141:377-383.
19. Reddy, G.P., U. Hayat, C. Abeygunawardana, C. Fox., A.C.
Wright, D.R. Maneval, C.A. Bush, and J.G. Morris. 1992.
Purification and determination of the structure of capsular
polysaccharide of Vibrio vulnificus M06-24. J. Bacteriol.
174:2620-2630.
20. Simonson, J.G., and R.J. Siebeling. 1993. Immunogenicity of
Vibrio vulfnicus capsular poilysaccharides and polysaccharide-
protein conjugates. Infect. Immun. 61:2053-2058.
21. Tamplin, M.L. 1990. The ecology of Vibrio vulnificus in
Crassostrea virginica. J. Shellfish Res. 9:254.
22. Tamplin, M.L. 1992. The seasonal occurrence of Vibrio
vulnificus in shellfish, seawater, and sediment in United
States coastal waters. Final report to Salstonstall-Kennedy
Grant Program. U.S. Department of Commerce. Project No.
NA27FDO117-01.
23. Tamplin, M.L., and G.M. Capers. 1991. Persistence of Vibrio
vulnificus in tissues of Gulf Coast oysters, Crassostrea
virginica, expoosed to seawater disinfected with UV light.
Appl. Environ. Microbiol. 58:1506-1510.
24. Tamplin, M.L., and A.L. Martin. 1989. Characterization of
interactions between Vibrio vulnificus and tissues of the
American oyster, Crassostrea virginica. Abstr. Annu. Meet.
Am. Soc. Microbiol., p. 334.
25. Tamplin, M.L., A.L. Martin, A.D. Ruple, D.W. Cook, and C.W.
Kaspar. 1991. Enzyme immunoassay for detection of Vibrio
vulnificus in oysters, sewater, and sediment. Appl. Environ.
Microbiol. 57:1235-1240.
26. Tamplin, M., G.E. Rodrick, H.J. Blake, and T. Cuba. 1982.
Isolation and characterization of Vibrio vulnificus from two
Florida estuaries. Appl. Environ. Microbiol. 44:1466-1470.
27. Weichart, D., J.D. Oliver, and S. Kjelleberg. 1992. Low
temperature induced nonculturability and killing of Vibrio
vulnificus. FEMS Microbiol. Lett. 100:205-210.
TIME-TEMPERATURE FACTORS
Vibrio vulnificus in Oysters:
Changes After Harvest and During Processing
Dr. David W. Cook
Gulf Coast Seafood Laboratory (HFH-400),
Office of Seafood
U.S. Food and Drug Administration
One Iberville Road, P.O. Box 158,
Dauphin Island, Alabama 36528-0158
Introduction
In the summary of the 1988 workshop, the Time and Temperature
Work Group reported four areas where additional information was
needed: 1) growth rates of V. vulnificus in oysters while on boats
before refrigeration; 2) time/temperature curves for V. vulnificus
multiplication in shellstock; 3) efficacy of depuration; and 4)
effects of commercial shucking practices on V. vulnificus numbers
in shellfish. Except for depuration, these issues deal with
changes that occur in post-harvest shellfish. This update is
intended to summarize research progress in this area since the 1988
workshop. Published information is referenced, and additional
unpublished findings are presented.
Multiplication of V. vulnificus in Shellstock
The first evidence that V. vulnificus may multiply in post-
harvest oysters came from trace-lot studies. Oysters were sampled
at harvest in Louisiana and when they reached processing plants in
Mississippi, 18 to 20 hours later. Results revealed that
shellstock taken at processing plants had higher numbers of V.
vulnificus than shellstock taken at harvest (2). Regarding the
effect of temperature, oysters held at 22xC showed an increase in
numbers of V. vulnificus, but densities in oysters held at 10xC
generally remained constant or even decrease with time.
The methodology used in the early studies for recovering and
identifying V. vulnificus was not highly refined. The 1988
Workshop provided better methodology and most of the subsequent
research has relied on those methods or modifications of them.
Methodology used in our studies was an MPN technique, using
enrichment in alkaline-peptone water, isolation on modified CPC
agar (7,12), identification with an EIA procedure (12), and
confirmation with API biochemical tests on some of the isolates.
In our recent studies, results for each time point are an average
of V. vulnificus counts on two 12-oyster composites.
Table 1 provides additional evidence that V. vulnificus
multiplies in shellstock oysters between the time of harvest and
the time at which they reach processing plants. These data are
from samples collected over several years and does not represent a
trace-lot study. The data show that shellstock reaching processing
plants during warm weather months, when temperature abuse could
occur, had much higher V. vulnificus counts than did oysters at
harvest.
Murphy and Oliver (8) questioned the ability of V. vulnificus
to multiply in oysters at warm temperatures. They used a
transposon-containing strain of V. vulnificus inoculated into
oysters by feeding, and then placed the oysters at temperatures of
0.5, 5.0, 10.0, 17.0, and 22.0xC. At all temperatures, continual
decreases in the numbers of V. vulnificus were reported.
Kaspar and Tamplin (5), commenting on the work of Murphy and
Oliver (8), suggested that the failure of the organism to multiply
at 22xC may be due to the inability of their transposon-containing
strain to compete with autochthonous microflora or colonize the
oysters. Additional information suggesting that the strain may
react differently from indigenous strains in oysters was recently
published (4). That study found the transposon-containing V.
vulnificus strain was rapidly depurated from oysters when the
naturally occurring strains were not eliminated.
Temperature Control of V. vulnificus in Oyster Shellstock
Before 1993, the National Shellfish Sanitation Program (NSSP)
required that shellstock be held under refrigeration at 10xC or
lower. In 1993, the temperature requirement was changed to 7.2xC
(13).
Kaysner et al. (6) provided data to support the effectiveness
of refrigeration to control the growth of V. vulnificus. In their
study both Crassostrea virginica and C. gigas were injected with V.
vulnificus and held for up to eight days at 0.5 and 10xC. During
this period, V. vulnificus numbers in the oysters remained
relatively constant or declined. Experiments with naturally
occurring V. vulnificus in C. virginica showed similar trends
during cold storage.
Table 2 provides results from six separate experiments in our
laboratory in which Gulf Coast oysters were held under mechanical
refrigeration at 10xC. Generally, there was a reduction in numbers
with time, although V. vulnificus can survive in shellstock for
more than 15 days of storage at 10xC. Studies with shellstock
stored at lower temperatures suggest that the lower the temperature
of storage, the greater the decline in numbers of V. vulnificus
(3).
Kaspar and Tamplin (5) provided further evidence that low
temperatures could control the multiplication of V. vulnificus in
shellstock oysters. In oysters held at 0-4xC, the numbers of the
bacterium decreased with time. Oysters held at 30xC showed a sharp
increase in V. vulnificus, but when the temperature-abused oysters
were then placed under refrigeration at 4xC, the numbers of V.
vulnificus began to decrease.
Minimal Growth Temperature for V. vulnificus
We conducted experiments to determine the minimal growth
temperatures of environmental strains of V. vulnificus in tryptic
soy broth by recording changes in the numbers of colony forming
units over a 6-day incubation period. Typical results from 12
environmental strains tested were as follows: after one day of
incubation at 8, 10, and 13xC, reductions in the numbers of V.
vulnificus were found at all temperatures. At 8 and 10xC, V.
vulnificus continued to decrease throughout the 6-day storage
period. However, after two days at 13xC, an increase in the number
of cells was found, and by day 6 the broth showed turbidity. From
these results, it appeared that the minimum growth temperature is
about 13xC. Research by Kaspar and Tamplin (5) supports this
finding.
Recently completed studies at the Gulf Coast Seafood
Laboratory evaluated the multiplication of V. vulnificus in oyster
shellstock held under controlled temperatures and at outside
ambient temperature conditions (1). Each lot of shellstock was
divided into four portions and placed under temperature control or
ambient storage within two hours of harvest. Storage temperatures
were controlled at 10, 13 and 18xC. Ambient temperatures ranged
from 22.8 to 31.1xC. Oysters were sampled after 12 and 30 hours of
storage. Figure 1 represents a summary of data from five separate
experiments completed during the summer of 1993, and shows the
average net log10 change in numbers of V. vulnificus. Numbers
decreased slightly in shellstock held at 10 and 13xC throughout the
30-hour storage period. At 18xC storage, numbers remained
relatively unchanged during the first 12 hours, but increased by
almost 1 log at 30 hours. Numbers in oysters held at the ambient
outside temperatures increased more than 1 log in 12 hours,
although this level seemed to decrease slightly by 30 hours.
Statistical analyses show that the numbers of V. vulnificus in
oysters held at 18xC for 30 hours and at ambient outside
temperature for 12 and 30 hours were both significantly different
from the V. vulnificus count at the start of the study. Changes at
other temperatures and time periods were not significantly
different.
Effectiveness of NSSP Harvest-to-Refrigeration Time Restriction in
Controlling Post-Harvest Multiplication of V. vulnificus
Concern about the multiplication of V. vulnificus in oysters
prompted the 1993 Interstate Shellfish Sanitation Conference (ISSC)
to recommend that limits be placed on the time shellstock oysters
can remain without refrigeration after harvest. This resulted in
the following addition to the NSSP Manual of Operations, Part II,
Section B.1:
"e. Shellstock, not intended for wet storage or depuration,
shall be placed under temperature control within 20 hours of
harvest from April through November and within 36 hours from
December through March. Once placed under temperature
control, shellstock shall be iced or the shellstock storage
area or conveyance shall be continuously maintained at 7.2xC
(45xF) or below until final sale..." (13).
Our laboratory has carried out studies to determine the
efficacy of this requirement as it relates to controlling V.
vulnificus in post-harvest Gulf Coast oysters. Where possible,
commercial oystermen were employed to harvest and transport the
oysters to landing. At that point, oysters were stored in burlap
sacks at outside ambient temperatures in the shade and sampled at
prescribed times. As shown in Table 3, no multiplication of V.
vulnificus was evident through 36 hours in winter-harvested oysters
even though the oysters had initial low numbers of V. vulnificus
and were exposed to temperatures that support the multiplication of
this bacterium.
During the summer harvest period (Table 3), numbers of V.
vulnificus in oysters increased in as little as 12 hours after
harvest and these continued to increase with additional storage
time. These result suggest a need to reduce the time oysters may
remain outside temperature control during the summer.
Effect of Commercial Processing on V. vulnificus in Oysters
It is notable that few V. vulnificus infections have been
linked to the consumption of oysters shucked in a processing plant.
Commercial processing of raw oysters on the Gulf Coast is usually
limited to the removal of the shell, which we call "shucking," and
washing the oysters to remove detritus before draining and packing.
Sometimes, washing is accomplished by placing the oysters in a tank
of water and agitating them with forced air. This process is
called "blowing."
Ruple and Cook (11) studied the effect of shucking and washing
on the numbers of V. vulnificus in oyster meats to determine if V.
vulnificus were reduced by standard practices. They found that
shell liquor and drained meats contained about equal numbers of V.
vulnificus. Further, commercial shucking and washing of oyster
meats did not significantly alter the numbers of V. vulnificus.
The effect of washing by blowing on V. vulnificus in oysters
was also studied (Ruple and Cook, unpublished data). Table 4
presents results from several studies using a laboratory scale
blower. Blowing in tap water, ice water, and water containing
chlorine were tested separately. Although results were not
absolutely definitive, none of the blowing procedures appeared to
produce a consistent change in the numbers of V. vulnificus. This
may have been expected since the V. vulnificus are internal to the
oyster.
Effect of Heat on Environmental Strains of V. vulnificus
The effectiveness of heat on V. vulnificus has been assessed
(3). Studies with 18 environmental strains isolated from oysters
provided a mean D-value of 39.8 q 12.2 seconds at 50xC. Clearly,
the organism can be easily destroyed by heat.
The ability of mild heat to kill naturally occurring V.
vulnificus in oysters was tested over the range of 45 to 50xC.
Heating was achieved by placing oysters in water at the specified
temperature and agitating them for the specified period.
Reductions were rapid at temperatures approaching 50xC and all
viable V. vulnificus were eliminated from oysters held at 50xC for
ten minutes. Studies also have shown that this heating process
does not destroy the raw characteristic of the oyster (9).
Cooking was found effective in destroying V. vulnificus (10).
Following published recipes for cooking oysters and also using one-
half the recommended cooking times, effects on numbers of V.
vulnificus were determined. Generally, one-half the recommended
time was effective in eliminating V. vulnificus from fried oysters
and broiled oysters. In recipes for baking and stewing oysters,
the full cooking time was required to eliminate the V. vulnificus.
Effect of Cold Storage and Freezing on V. vulnificus in Shucked
Oysters
Oysters that have been shucked and packed in processing plants
and stored on crushed ice generally exhibit a reduction in V.
vulnificus counts with time (11). In some instances, reductions of
up to 3 logs in 7 days are seen. Shucked meats held at 4xC show
similar reduction during storage.
Freezing has been examined as a means of reducing V.
vulnificus (3). Three different processes for freezing were
tested. An IQF CO2 freezer was used to freeze oysters at -30xC; a
blast-freezing process froze oysters in cold air at -23xC; and a
"home" freezer was used to freeze and hold oysters at -20xC. V.
vulnificus were enumerated in samples held at -20xC at different
intervals over a 12-week period. All freezing processes caused a
large reduction in V. vulnificus counts, up to 3 logs. However, V.
vulnificus was not totally eliminated and persisted for 12 weeks.
Unpublished data provided by Dr. Marilyn Kilgen, Nicholls
State University, show the effects of combining blowing and
freezing on V. vulnificus. Oysters were harvested in Louisiana and
shellstock shipped to Virginia to be shucked and frozen. Numbers
of V. vulnificus in the oysters at harvest were 24,000 per gram.
At the Virginia plant oysters were shucked and blown, and the
numbers were reduced to 93 per gram. Blowing appears to have had
an effect, but the length of time between harvest and processing
(12 days) may have contributed to this reduction. The oysters were
divided, with half frozen in a blast-freezer and the other half
cryogenically frozen using CO2 at -60xF. Both freezing techniques
further reduced the numbers of V. vulnificus. Frozen oysters were
stored and examined periodically, and reductions in numbers of V.
vulnificus continued. However, V. vulnificus remained viable for
about 40 days after freezing.
Summary
Since the 1988 Workshop, it has been established that the
NSSP-recommended holding temperatures are effective in preventing
the growth of V. vulnificus in both shellstock and shucked oysters.
V. vulnificus can, however, multiply in oyster shellstock held
above 13xC. The amount of growth is time- and temperature-
dependent. Rapid multiplication can occur in Gulf Coast oysters at
summer temperatures resulting in more than 1 log increases in V.
vulnificus in shellstock oysters in as little as 12 hours.
Processing, including shucking, washing, and blowing, has only
a minimal effect on the numbers of V. vulnificus in the oysters,
presumably because the bacteria are internal to the oyster meats.
However, when oyster meats were stored on ice, the numbers of V.
vulnificus decreased with time of storage. Freezing brings about
a large reduction in numbers of V. vulnificus, but does not
eliminate the organism from oyster meats. V. vulnificus is not
heat resistant and is easily killed by cooking.
Additional Research Needs
Although there are some unanswered question relating to the
time/temperature topic, certain other research areas related to V.
vulnificus are more pressing.
1. Definition of strain pathogenicity/virulence. It is extremely
important to know if all strains of V. vulnificus are capable
of causing disease and if certain strains have increased
virulence. With this information, public health strategies to
better control illness and mortality caused by the organism
can be targeted.
2. Methods to easily differentiate strains of concern. Research
requires that reliable methodology be developed with which to
separate/distinguish strains.
3. Definition of infectious dose and how it may differ for
individuals in different at-risk groups. This information is
needed for risk assessment.
4. Development of rapid methods (24 hours) to detect and/or
enumerate V. vulnificus strains of concern. If for regulatory
reasons tolerance standards on V. vulnificus levels in
shellfish are imposed, rapid qualitative and/or quantitative
methods will be needed.
Table 1. Comparison of Vibrio vulnificus levels (MPN/g) in oyster
shellstock at harvest with levels in shellstock
taken at processing plants
Oysters at harvesta Oysters at processing plantb
Median Number of Median Number of
Month V.v. MPN/g Samples V.v. MPN/g samples
Jan. <0.3 11 2 16
Feb. <0.3 12 <0.3 10
Mar. <0.3 8 <0.3 10
Apr. 2.6 8 20,000 8
May 930 11 >110,000 10
Jun. 4,300 8 29,000 6
Jul. 2,200 6 >110,000 8
Aug. 930 9 >110,000 6
Sep. 4,300 7 >110,000 6
Oct. 490 8 31,000 6
Nov. 18 6 1,100 8
Dec. 0.6 6 33 8
aOysters (Crassostrea virginica) harvested from Mississippi and
Alabama waters, chilled immediately, and examined for V.
vulnificus within eight hours.
bDetails on harvesting and shipping described by Ruple and Cook
(11).
**********
Table 2. Vibrio vulnificus in oyster shellstock stored at 10xC
V. vulnificus MPN/ga in oystersb
Trial Day 0 Days 6-7 Days 13-15
1 490 240 150
2 930 590 190
3 12,000 6,800 9,300
4 14,000 930 970
5 240,000 16,000 19,000
6 260,000 230,000 32,000
aAverage MPN/g based on results from two 12-oyster composites.
bCrassostrea virginica.
Table 3. Evaluation of post-harvest multiplication of Vibrio
vulnificus in shellstock oysters before temperature
control is provided
V. vulnificus MPN/ga in oystersb
Water Storage Stored for:
Month Temp(xC)c Temp(xC)d 0 h 12 h 20 h 36 h
Winter harvest samples
Feb. 12 12-18 <0.3 -e - <0.3
Mar. 18 10-13 33 - 58 68
Mar. 18 10-17 59 - 29 57
Nov. 20 20-25 26 - 64 36
Dec. 16 9-22 <3 - <3 <3
Summer harvest samples
Apr. 21 16-24 2,400 - 63,000 58,000
Apr. 23 21-24 1,400 14,000 36,000 58,000
Apr. 23 21-27 190 12,000 93,000 -
May 27 24-26 4,300 240,000 290,000 -
Jun. 28 25-27 1,600 130,000 140,000 -
Oct. 22 18-23 680 4,300 6,600 14,000
aAverage MPN/g based on results from two 12-oyster composites.
bCrassostrea virginica.
cWater temperatures in the harvest areas at times of harvest.
dRange of temperatures inside the sack of oysters during storage
at outside ambient temperatures.
eMPN not determined at this interval.
**********
Table 4. Effect of washing by blowing on Vibrio vulnificus
levels in shucked oyster meats
Conditions of V. vulnificus MPN/g in Oystersa
blowingb Before After
Tap water, 15 min >110,000 >110,000
Ice water, 15 min 65,000 110,000
Ice water, 15 min 12,000 9,300
100 ppt Cl2b, 15 min 33,000 >110,000
200 ppt Cl2b, 15 min 29,000 2,300
200 ppt Cl2b, 15 min >110,000 >110,000
aCrassostrea virginica.
bLiquid used in blower, and agitation time.
cChlorine is not approved for this process.
Figure 1. Changes in the levels of V. vulnificus in oysters (C.
virginica) held 12 and 30 hours at various temperatures. Bar
heights represent mean (n = 5) log10 changes in V. vulnificus MPN/g
determined subsequent to harvest; error bars indicate standard
deviations (5 trials); AAT - outside ambient air temperature (1).
References
1. Cook, D.W. 1994. Effect of time and temperature on
multiplication of Vibrio vulnificus in postharvest Gulf coast
shellstock oysters. Appl. Environ. Microbiol. 60:3483-3484.
2. Cook, D.W., and A.D. Ruple. 1989. Indicator bacteria and
Vibrionaceae multiplication in post-harvest shellstock
oysters. J. Food Prot. 52:343-349.
3. Cook, D.W., and A.D. Ruple. 1992. Cold storage and mild heat
treatment as processing aids to reduce the numbers of Vibrio
vulnificus in raw oysters. J. Food Prot. 55:985-989.
4. Groubeth, T.N., and J.D. Oliver. 1994. Interaction of Vibrio
vulnificus and the Eastern oyster, Crassostrea virginica. J.
Food Prot. 57:224-228.
5. Kaspar, C.W., and M.L. Tamplin. 1993. Effects of temperature
and salinity on the survival of Vibrio vulnificus in seawater
and shellfish. Appl. Environ. Microbiol. 59:2425-2429.
6. Kaysner, C.A., M.L. Tamplin, M.M. Wekell, R.F. Stott, and K.G.
Colburn. 1989. Survival of Vibrio vulnificus in shellstock
and shucked oysters (Crassostrea gigas and Crassostrea
virginica) and effects of isolation medium on recovery. Appl.
Environ. Microbiol. 55:3072-3079.
7. Massad, G., and J.D. Oliver. 1987. New selective and
differential medium for Vibrio cholerae and Vibrio vulnificus.
Appl. Environ. Microbiol. 53:2262-2264.
8. Murphy, S.K., and J.D. Oliver. 1992. Effects of temperature
abuse on survival of Vibrio vulnificus in oysters. Appl.
Environ. Microbiol. 58:2771-2775.
9. Ruple, A.D. 1993. Mild heat treatment as a processing aid to
eliminate Vibrio vulnificus from raw oysters. Final Project
Report, Grant Number NA27FD0061-01, National Marine Fisheries
Service, Washington, DC.
10. Ruple, A.D., and D.W. Cook. 1992. The effect of cooking
method and cooking time on Vibrio vulnificus in oysters.
Abstract, J. Mississippi Acad. Sci. 37:52.
11. Ruple, A.D., and D.W. Cook. 1992. Vibrio vulnificus and
indicator bacteria in shellstock and commercially processed
oysters from the Gulf Coast. J. Food Prot. 55:667-671.
12. Tamplin, M.L., A.L. Martin, A.D. Ruple, D.W. Cook, and C.W.
Kaspar. 1991. Enzyme immunoassay for identification of
Vibrio vulnificus in seawater, sediment, and oysters. Appl.
Environ. Microbiol. 57:1235-1240.
13. U.S. Department of Health and Human Services. 1993 Revision.
National Shellfish Sanitation Program Manual of Operations.
Part II. Sanitation of the Harvesting, Processing and
Distribution of Shellfish. Shellfish Sanitation Branch, U.S.
Food and Drug Administration, Washington, DC.
Questions and Answers
Q. Mr. Thompson: My question deals with the freezing issue. As
you are aware, we have a CO2 plant in Texas that goes into the
IQF process. I have some interest in that as their Regulator.
What is your opinion from the data that you've shown?
Obviously, there is a reduction in the IQF. There is also a
reduction in the other freezing processes. But it appeared,
at least from the graph that you showed, that long term, there
was perhaps a better reduction, at least not a back growth, of
the vulnificus in the IQF product. Can you comment generally
about IQF process and what you might think about that?
A. Dr. Cook: In the studies that we've completed -- and this was
done a couple of years ago when I was at GCRL -- we did see
this very rapid reduction, but the organism stayed around for
a long time. Now, whether or not the blast-freezing or the
IQF-freezing is going to make any difference in how that tail
end of that curve goes off, I really can't say. I just don't
have that information, and we don't know what freezing does to
the different strains if there are some that are pathogenic
and others that are not. We're just looking at total numbers.
Q. Mr. Thompson: Basically, what you did was sort of on the
beginning edge of this, and you're not sure just what it all
means as far as IQF goes?
A. Dr. Cook: Correct.
Q. Dr. Walter Canzonier, Bivalve Packing and Marshall Oyster
Culture Foundation: Most of this work seems to have been done
with Gulf Coast oysters under Gulf Coast conditions.
A. Dr. Cook: That's correct.
Q. Dr. Canzonier: I think there is a little bit of a risk
involved in making general assumptions about this because we
do know that animals, invertebrates in particular, are not all
created equal, and they don't all behave equally, both because
of physiological status and seasonal variations. I think we
have to do a little bit more in looking at those factors in
terms of levels of both the suspected contaminant and the way
the organism, or the way the host, if you want to use that
term, or the way the oyster, handles these contaminants under
these various conditions. This has not been done, and I think
we have to do this. I agree that there is certainly a lot of
room for improvement. In particular, the so-called infectious
dose is something that is very, very bothersome to the
industry because a lot of people are talking about cutoff
points, and we don't know anything about the appropriate
cutoff points even for the compromised individual, to say
nothing of the normal population.
A. Dr. Cook: I agree with what you say. We focused on the Gulf
Coast because that is where the problem is.
Q. Dr. Hlady: I believe that the low heat treatment results that
you showed were with shucked product.
A. Dr. Cook: Correct.
Q. Dr. Hlady: Which, as we know, has not been associated with
illness. Any experience with low heat treatment of
shellstock?
A. Dr. Cook: This would be a sufficient amount of heat to cause
the shellstock to gape open; it would not be an acceptable
process for the shellstock product.
OYSTER PURIFICATION AND INTERVENTION MEASURES
Dr. Mark Tamplin
3031 McCarty Hall, Home Economics Program
Institute of Food and Agricultural Sciences
University of Florida
Gainesville, Florida 32611-0310
Depuration
Depuration has been successfully used to reduce levels of
fecal coliforms and Escherichia coli and thereby recover shellfish
resources (10) that would otherwise not be approved for harvest and
human consumption. This commercial practice has been investigated
to determine whether the numbers of Vibrio vulnificus associated
with oysters could be reduced, as occurs for coliform bacteria.
In one of the earliest depuration studies of vibrios, Greenberg (4)
found that V. parahaemolyticus, V. harveyi, and perhaps a few other
vibrios were not eliminated as readily as E. coli. Investigators
in Australia made similar observations (3).
In the early 1980s, Kelly (6) reported on retention of V.
vulnificus in oysters. Oysters were placed in seawater in a
bucket, and then inoculated with V. vulnificus. The water was
recirculated, but the system did not use UV light disinfection. V.
vulnificus was detected at 7, 14, and 21 days. However, the
results did not apply to commercial interests where UV light and a
48 hour period are used.
Richards (10) provided an excellent review of depuration.
Rodrick and coworkers have frequently reported their findings on
depuration of V. vulnificus in clams. A review by Canzonier (1)
provides information on shellfish depuration as well. Jones and
collaborators (5) in New Hampshire published reports on depuration
of E. coli and vibrios, showing that although fecal coliforms are
eliminated fairly readily, vibrios are not.
In detailed studies of V. vulnificus retention in oysters,
Tamplin and coworkers (11) showed that V. vulnificus replicates and
is retained in oysters. During depuration, V. vulnificus
distribution in tissues shifts from the digestive tract to other
tissues. In addition, a significant number of V. vulnificus are
shed into sea water from the surface of an oyster shell.
Comparative results produced by Murphree and Tamplin (8)
showed that at 19xC oysters eliminated E. coli readily in 48 hours;
Salmonella typhimurium decreased, but at a lower rate than E. coli;
V. cholerae was still present at elevated levels after 48 hours;
and V. vulnificus consistently showed an increase in numbers at
temperatures >15xC.
Because the flora of an oyster consists not only of organisms
found in the digestive tract, but in and on all other tissues as
well, including the shell, a considerable number of bacteria can be
eliminated or shed by oysters in a given period of time.
Experiments showed that a single oyster releases over 100,000 V.
vulnificus per hour into seawater (11). This finding is relevant
to depuration systems and also to environmental persistence. The
data signify that replication of V. vulnificus occurs on or in the
oyster during depuration treatment. Oyster depuration trials have
also been conducted by Tamplin and coworkers (11) using sea water
at 23 and 15xC, two temperatures recommended by the National
Shellfish Sanitation Program. At 15xC, E. coli were readily
eliminated, while a relatively low level of V. vulnificus was
maintained. This work was performed in November, when oysters
contained ten V. vulnificus cells per gram. At higher
temperatures, V. vulnificus levels exceeded 106 within 48 hours.
Thus, while E. coli and Salmonella are being eliminated, the
depuration process may unwittingly be increasing the risk of V.
vulnificus disease.
Other experiments have shown that depuration experiments using
laboratory V. vulnificus strains do not provide realistic data when
compared to natural flora (9). At lower depuration temperatures,
these endogenous strains look similar to natural strains in their
profile. However, at the elevated temperature, the endogenous
flora increased, but the laboratory added strain decreased in
number.
In the absence of significant rates of elimination for V.
vulnificus through depuration, the effect of various substances on
their abilities to promote the elimination of V. vulnificus was
investigated. Logan et al. (7) tested the use of certain GRAS
substances (i.e., generally regarded as safe by FDA) to determine
if any would kill V. vulnificus in the laboratory, with the intent
of testing promising substances in depuration water. Some
compounds showed inactivation properties, while others did not.
The molar concentrations used and found effective in the
laboratory, and the corresponding times of exposure indicated that
sodium erythorbate and tannic acid might be of some use. At four
hours, V. vulnificus was undetectable using tannic acid and sodium
erythorbate. However, in depuration trials, no positive reductions
were obtained. The oysters were sensitive to the presence of these
compounds and ceased siphoning activity. Consequently, elimination
did not occur. Limited attempts to "mask" this effect were
unsuccessful.
Comparison of ozone disinfection of seawater to that afforded
by UV in depuration systems has shown no difference between these
methods relative to levels of V. vulnificus in oysters. In the
absence of successful efforts to reduce V. vulnificus levels in
oysters by depuration, other investigators are examining the
possibility of adding parasites of V. vulnificus to sea water
systems (DePaola, personal communication). Luftig and coworkers at
LSU are examining bacteriophages specific for V. vulnificus. Some
work with bdellovibrios, which are parasites of vibrios, is being
pursued by Williams and coworkers at the University of Maryland.
It is reasonable to assume that V. vulnificus proliferates in
oysters in depuration systems at permissive temperatures.
Temperature control may effectively restrict growth, but decreases
in levels of the organism are too gradual to accomplish meaningful
reductions within a commercial time frame.
Irradiation
A number of investigators have explored the effectiveness and
practicality of using radiation to reduce V. vulnificus levels in
oysters. Reports from Dixon and Rodrick (2), and others
demonstrate the effectiveness of irradiation. Dixon and Rodrick
(2) report a trade-off when gamma radiation is used. Radiation may
be effective against V. vulnificus but usually damages the oyster,
which leads directly to a shortened shelf life. For example, at 1
kGy, a 1 to 2 log decrease in V. vulnificus can be expected;
however, within five days, there is a 25% loss in viable product.
At 2 kGy, which is very effective against the pathogen, there is a
40% decrease in viable product within five days.
References
1. Canzonier, W.J. 1982. Depuration of bivalve mollusks--what
it can and cannot accomplish and some practical aspects of
plant design and operation. Proc. Int. Seminar on Management
of Shellfish Resources. Irish Mar. Farm. Assoc., Tralee,
Ireland.
2. Dixon, D.W., and G.E. Rodrick. 1990. Comparative effects of
ionizing radiation and high energy electron beams on molluscan
shellfish. Proc. 15th Annu. Conf. Trop. Subtrop. Fish.,
Gainesville, FL, pp. 73-75.
3. Eyles, M.J., and G.R. Davey. 1984. Microbiology of
commercial depuration of the Sydney rock oyster, Crassostrea
commercialis. J. Food Prot. 47:703-706.
4. Greenberg, E.P., M. Dubois, and B. Palhof. 1982. The
survival of marine vibros in Merceneria merceneria, the
hardshell clam. J. Food Safety 4:113-123.
5. Jones, S.H., T.L. Howell, and K.R. O'Neill. 1991.
Differential elimination of indicator bacgteria and pathogenic
Vibrio spp. from eastern oysters (Crassostrea virginica
Gmelin, 1791) in a commercial controlled purification facility
in Maine. J. Shellfish Res. 10:105-112.
6. Kelly, M.T., and A. Dinuzzo. 1985. Uptake and clearance of
Vibrio vulnificus from Gulf Coast oysters (Crassostrea
virginica). Appl. Environ. Microbiol. 50:1548-1549.
7. Logan, M., J. Bemiss, J. Sample, and S. Price. 1992. Vibrio
vulnificus inactivation by selected substances in seawater.
Proc. Aquaculture '92 Conference, Orlando, FL, p. 148.
8. Murphree, R.L., and M.L. Tamplin. 1994. Uptake and retention
of Vibrio cholerae O1 in the American eastern oyster,
Crassostrea virginica. American Society for Microbiology, p.
375.
9. Murphy, S.K., and J.D. Oliver. 1992. Effects of temperature
abuse on survival of Vibrio vulnificus in oysters. Appl.
Environ. Microbiol. 58:2771-2775.
10. Richards, G.P. 1988. Microbial purification of shellfish: a
review of depuration and relaying. J. Food Prot. 51:218-251.
11. Tamplin, M.L., and G.M. Capers. 1991. Persistence of Vibrio
vulnificus in tissues of Gulf Coast oysters, Crassostrea
virginica, expoosed to seawater disinfected with UV light.
Appl. Environ. Microbiol. 58:1506-1510.
RELAYING
Oyster Relay and Depuration Experience in New Hampshire
Dr. Stephen H. Jones
Jackson Estuarine Laboratory
University of New Hampshire
Durham, New Hampshire 03824
Research at the Jackson Estuarine Laboratory at the University
of New Hampshire has focused on developing strategies for removing
bacterial pathogens from harvested shellfish. A particular
emphasis has been directed to the removal of Vibrio vulnificus from
eastern oysters. The Jackson Estuarine Laboratory is located in
the middle of the Great Bay estuary in New Hampshire. The estuary
is bounded by Maine to the north and is the dominant estuary of the
coastline of New Hampshire and its inland waters. Seven rivers
flow into the estuary and the larger bodies of water, Great Bay and
Little Bay.
In accordance with current NSSP criteria, the shellfish
growing waters in Maine are classified as restricted, a portion of
the New Hampshire waters are classified as approved, and the rest
of New Hampshire waters are essentially classified as prohibited.
Commercial shellfishing occurs in the restricted Maine waters of
the Salmon Falls and Piscataqua Rivers, while only recreational
shellfishing is allowed in the approved New Hampshire waters. The
approved area in New Hampshire is largely located in Great Bay, 60%
of which can be exposed at low tide; at high tide the tidal flats
are covered by eight to ten feet of coastal sea water. The waters
in this relatively shallow area and in the tributaries can warm
considerably in the summer. The incidence of V. vulnificus is
variable in these different areas of the estuary (5,6).
Spinney Creek is a salt pond located in Eliot, Maine, that is
classified as approved and is separated from the main trunk of the
Piscataqua River by a tidal gate and causeway. A commercial
depuration and container relay facility, Spinney Creek Shellfish,
Inc. (SCS), is located on this salt pond and it is from here that
some interesting results have been obtained.
Studies since 1989 show a consistent trend for the temporal
incidence of V. vulnificus in the estuary (Table 1) (5,6). The
species appears in late June or early July and disappears in
October. The temperature of the water in this estuary ranges from
below 0xC in winter, when it can freeze over, to about 25xC in some
places during summer. Densities of V. vulnificus increase as the
temperature increases during the summer. A spatial heterogeneity
of the organism is also evident within the estuary that appears to
be related to salinity and fecal contamination levels (Table 1)
(6). V. vulnificus is detected rarely in the approved waters of
Great Bay and frequently in the tributary rivers that
characteristically have lower salinities and include commercial
harvest areas used by SCS; to date it has not been detected in the
approved waters of Spinney Creek.
Depuration of oysters for 48 hours at SCS was successful in
removing fecal coliforms and Escherichia coli, but not V.
vulnificus (4). However, it seemed possible to take advantage of
the spatial heterogeneity observed by means of relaying. This is
not relaying in the traditional sense, where shellfish are taken to
approved areas. By moving oysters from waters like the commercial
harvest sites in the Salmon Falls River with a relatively high
incidence of the organism and placing them in waters at SCS, and at
Adams Point in Great Bay, it seemed possible that exposure to those
waters could reduce levels of the organism in the shellfish.
The container relay system at SCS consists of lagoons that
receive water pumped up from the approved waters of Spinney Creek.
Oysters are exposed to these waters for seven days by virtue of
relay verification studies and routine testing for fecal coliforms,
after which the shellfish are depurated.
Initial studies, based on samples examined each week, showed
that V. vulnificus levels in oysters held in Spinney Creek water
were significantly lower than those detected previously in the
freshly harvested shellfish (Table 2) (2). This result was
obtained in each instance when batches of relayed oysters were
examined.
Follow-up studies in 1991 and 1992 confirmed these findings.
Consistently during the summer, freshly harvested oysters contained
high levels of V. vulnificus and, following seven days of exposure,
the levels of this species in relayed oysters were substantially
lower. Studies in 1993 found the same consistent removal of the
organism in relayed oysters (1).
Parallel with these findings, oysters were relayed from New
Hampshire waters near the commercial harvest site in the Salmon
Falls River to Adams Point, which is an approved site in Great Bay
with a consistently low incidence of V. vulnificus. The results of
this study showed a significant decrease in levels of the organism
after four weeks exposure to these waters (1,2).
These results suggest that relaying oysters containing
relatively high levels of V. vulnificus to waters with low or no
detectable V. vulnificus can result in the removal of the organism
from the shellfish. The relay waters used in the described studies
characteristically have relatively high salinities (about 25 ppt
even at low tide) and low levels of fecal coliforms, dissolved
inorganic and organic nutrients, and suspended solids during the
summer (3). This is in contrast to the harvest site in the Salmon
Falls River where salinities were variable and relatively low at
low tide because of fresh water input. Levels of dissolved
inorganic and organic nutrients and fecal coliforms from non-point
source pollution are relatively high in Salmon Falls River and all
of the other tributaries where the frequency of V. vulnificus
detection is high during the summer period. Although either seven
or 28 days are required, the salinity difference, and possibly
other water quality differences experienced by relayed oysters,
appears to be a means to control and reduce the levels of V.
vulnificus in oysters.
Table 1. Environmental and microbiological characteristics of
water at harvest and relay sites in Great Bay Estuary
Site Temp. Salinity Fecal coliform Incidence of
(xC) (ppt) per 100 ml V. vulnificusa
Salmon Falls River
Nov-May 0-12 1-5 412b 0/9c
June-Oct 11-26 1-17 100 11/20
Spinney Creek
Nov-May 0-12 18-29 52 0/4
June-Oct 11-26 18-29 6 0/20
Great Bay
Nov-May 0-18 10-30 27 0/6
June-Oct 9-24 13-30 17 3/27
aSummary of incidence from water at 3 sites.
bGeometric mean.
cNumber of positive samples/total number of samples.
**********
Table 2. Vibrio vulnificus in oysters relayed to SCS relay lagoons
V. vulnificus/g oyster meat
Date freshly harvested relayed V. vulnificus/100 ml water
8/21 930 24 <3
8/28 4600 4.3 <3
9/4 1500 <3 <3
9/19 150 <3 <3
10/3 4.3 <3 <3
10/16 <3 <3 <3
11/7 <3 <3 <3
References
1. Jones, S.H., T.L. Howell, R. Langan, and K.R. O'Neill.
Relaying to eliminate Vibrio vulnificus and fecal-borne
bacteria from oysters in northern New England. Submitted for
publication.
2. Jones, S.H., T.L. Howell, K.R. O'Neill, and R. Langan. 1995.
Strategies for removal of indicator and pathogenic bacteria
from commercially harvested shellfish. Aquatic Living
Resources. In press.
3. Jones, S.H., R. Langan, W.H. McDowell, and B.W. Summer-Brason.
1994. Influence of nutrient enrichment on microbial
contamination of the Great Bay Estuary. Abstr. Annu. Meet.
Am. Soc. Microbiol.
4. Jones, S.H., K.R. O'Neill, and T.L. Howell. 1991.
Differential elimination of indicator bacteria and pathogenic
Vibrio sp. from Maine oysters (Crassostrea virginica) in a
commercial controlled purification facility. J. Shellfish
Res. 10:105-112.
5. O'Neill, K.R., S.H. Jones, and D.J. Grimes. 1990. Incidence
of Vibrio vulnificus in northern New England water and
shellfish. FEMS Microbiol. Lett. 72:163-168.
6. O'Neill, K.R., S.H. Jones, and D.J. Grimes. 1992. The
seasonal incidence of Vibrio vulnificus in the Great Bay
Estuary of New Hampshire and Maine. Appl. Environ. Microbiol.
58:3257-3262.
OTHER GULF COAST OYSTER RESEARCH
Current Commercial Considerations
Dr. Marilyn Kilgen
Department of Biological Sciences
Nicholls State University
Thibodaux, Louisiana 70310
Winter Effects on Vibrio vulnificus
Gulf Coast oysters harvested during winter months reveal no
detectable (culturable) V. vulnificus upon bacteriological
examination. The question posed was, "What happens to the organism
in the winter season of the oyster?"
During February, 1994, experiments were conducted to provide
some insight. Six dozen oysters obtained from commercial
harvesters were divided randomly into six groups of 12 oysters
each. One group (Group 1, control) was assayed to confirm that
there was no detectable level of V. vulnificus. The recommended
alkaline peptone water enrichment MPN technique was used in
conjunction with mCPC streak plates to detect and isolate
presumptive positives, which were then confirmed by an ELISA.
The other five groups of 12 oysters were treated as follows.
Group 2 was placed in an aerated sea water tank with 10 ppt
salinity artificial sea salt at room temperature of 22xC (wet
storage) for five days. Group 3 was placed in a sea water tank
with 10 ppt salinity artificial sea salt at 30xC (wet storage) for
five days. Group 4 was left in an unsealed plastic sack on the lab
table at room temperature of 22xC (dry storage) for five days.
Group 5 was placed in the walk-in cold room in an unsealed plastic
sack at a temperature of 5xC (dry storage) for five days.
Group 3 oysters held in sea water at 30xC supported a large
increase in detectable V. vulnificus (from <0.3 MPN/g to 2.3 x 105
MPN/g). This was the only experimental group that showed a
significant increase above nondetectable levels.
Oysters from Group 2 in wet storage at room temperature (22xC)
only increased to 2.3 MPN/g from nondetectable original levels.
Oysters from Group 4 in dry storage at room temperature (22xC)
showed no detectable increase in V. vulnificus, but very large
increases in the aerobic plate count (APC)/g (from 5.4 x 103 to
5.05 x 107), indicating spoilage. Oysters in Group 5 in dry
storage at refrigerator temperature (5xC) showed a very slight
increase in V. vulnificus to a detectable level of 4.3 MPN/g.
This experiment will be conducted on a seasonal basis
throughout the year. In this manner, it is hoped that the effects
of parameters including the time of year harvested, the prevailing
water temperatures, and the site of harvest on V. vulnificus
outgrowth, will be elucidated.
Ionizing Radiation
Previous work using gamma radiation processing for live
shellstock, fresh shucked, and frozen shucked oysters was conducted
on a pilot scale. Live oysters were seeded in a salt water tank by
feeding for six hours with high levels (104-106 MPN/g) of
Escherichia coli, Klebsiella pneumoniae (environmental isolate),
Salmonella typhimurium, V. cholerae, and V. parahaemolyticus. The
seeded oysters were processed at the Louisiana State University
(LSU) Nuclear Science Center with sublethal doses of ionizing
irradiation (0.0 [control], 0.5, 1.0, 1.5, and 2.0 KGy). Results
showed that the Vibrio species were reduced to nondetectable levels
at 1.0 KGy. E. coli dropped to nondetectable levels at 1.5 KGy.
Klebsiella, Salmonella, and APCs showed 2 log reductions at 1.5 KGy
(2).
A pilot collaborative study with the California Health
Department, Radiation Sterilizers, Inc. of California; Motivatit
Seafoods, Inc.; and the U.S. Food and Drug Administration was
conducted during the summer of 1990 to evaluate the effect of low
dose gamma radiation on V. vulnificus levels in Louisiana
shellstock oysters. The levels of V. vulnificus in Louisiana
oysters commercially harvested, processed, and shipped to
California were reduced from 4.3 x 105 MPN/g to 0.4 x 100 MPN/g by
commercial irradiation (1.5 KGy dose) at Radiation Sterilizers,
Inc., California.
A pilot study using the first commercial food irradiation
facility in the United States (Vindicator, Inc., Florida) was
conducted in the summer of 1992 and a large scale commercial study
will begin this summer (1994). The pilot study was a preliminary
effort to evaluate results not only on V. vulnificus levels, but
also on live shellstock mortality or shelf life. It was conducted
in collaboration with Motivatit Seafood, Inc., Houma, Louisiana;
and Vindicator, Inc., Mulberry, Florida. Oysters were harvested,
shipped, and processed by normal commercial routes. They were
boxed into five commercial shellstock boxes of 50 oysters each.
These received doses of 0.0 (control), 0.5, 1.5, and 2.0 KGy,
respectively. Twelve oysters from each of the five boxes were
randomly selected for V. vulnificus analysis on days 1, 6, and 14
post irradiation. These were actually days 6, 12, and 20 post
harvest.
Table 1 shows results of the percent cumulative mortality and
V. vulnificus levels for experimental groups of 50 animals/dose
from days 1 through 14 post irradiation (days 6 through 20 post
harvest). The normal mortality of refrigerated live shellstock
post harvest varies with the condition of the oyster at different
times of the year, but is normally 2 to 4%. In this preliminary
study the irradiation level of 1.0 KGy showed good reduction of V.
vulnificus, from 2.3 x 105 MPN/g to 0.7 MPN/g, six days post
harvest (one day post irradiation). The dose of 1.5 KGy showed
reduction to undetectable levels (<0.3 MPN/g) six days post harvest
(one day post irradiation), with only 4% associated mortality. By
12 days post harvest (six days post irradiation), the cumulative
mortality was still only 4%. This is within acceptable ranges, and
the V. vulnificus was reduced to undetectable levels.
This summer (1994), funding from LSU Sea Grant will provide
for an extensive collaborative evaluation of the commercial
irradiation of live shellstock oysters, fresh shucked oysters,
frozen shucked oysters, and fresh crabmeat with Robert Grodner at
LSU; Motivatit Seafoods, Inc.; Vindicator, Inc.; and Cowart and
Morgan Seafoods, Inc. in Virginia. Dose-volume ratios for
commercial packaging and dose-time ratios will be carefully
evaluated to determine the most economically feasible method to
greatly reduce or eliminate V. vulnificus from live, fresh, and
frozen oyster products without adversely affecting the shelf life
of live oysters or the organoleptic qualities of fresh processed
products.
Commercial Processing and Freezing of Oyster Products
In the summer of 1993, Louisiana oysters were commercially
harvested and shipped by refrigerated truck to Virginia for
commercial shucking, 'blowing,' and freezing. For experimental
purposes, the oysters were shucked and 'blown' or rinsed from the
bottom of the tank to an overflow at the top with potable cold
water at 65xF. This process can vary from 3 to 10 minutes
depending on the time of year and the oyster texture. The state of
Virginia requires further rinsing in cold water until the rinse
water is clear. Rinsed oysters were then rapidly precooled to 45xC
by addition of clean ice to the tank. Samples of 24 of these
processed oysters were vacuum-packed in nonpermeable Mylar bags.
One bag of 24 oysters was refrigerated, unfrozen. The remainder
were commercially package-quick-frozen (PQF) by two methods: CO2
cryogenic freezing at -60xF, and blast freezing at -20xF.
Processed samples (bag of 24 oysters) were analyzed each week from
September 26, 1993, through January 3, 1994.
Initial levels of V. vulnificus were 2.4 x 104 MPN/g two days
post harvest in Louisiana. The level in fresh blown oysters was
reduced to 93 MPN/g; this was 12 days post harvest. Oysters were
held at 5xC before shucking and processing. The levels of V.
vulnificus in blast frozen oysters was 9.3 MPN/g four days post
freezing and 12 days post harvest. The levels in CO2 frozen
oysters was 1.5 MPN/g for that sampling date. By 20 days post
harvest and 12 days post freezing, levels of V. vulnificus varied
slightly from 100 to 10-1 to <0.3 (undetectable) MPN/g. The
combined commercial processing technologies of washing in fresh
water, storage at 5xC, and freezing were extremely effective in
reducing V. vulnificus to very low or nondetectable levels.
Shellstock oysters were also processed by removing the top
shell only, and packaging in a round styrofoam container for six
half-shell oysters. Cocktail sauce was placed in the center of the
six half-shell oysters. They were shrink-wrapped and PQF frozen in
liquid CO2 at -60xF. The process of freezing and thawing alone is
effective in greatly reducing V. vulnificus levels. However, these
frozen packages were also irradiated at levels from 0.5 to 5 KGy at
Vindicator, Inc. Irradiation of a frozen product at higher levels
of gamma irradiation provides the best technology for elimination
of all possible Vibrio species, and affects the sensory qualities
least. Overall, individual or combined technologies of cold fresh
water washing and soaking in ice water, freezing, and irradiation
can effectively reduce V. vulnificus in Gulf Coast oysters to
insignificant or undetectable levels.
Table 1. Effect of ionizing radiation on numbers of V. vulnificus
and shelf life of shellstock oysters
Dose (KGy)
0.0 1.0 1.5
Days
post irrad. % mort. MPN/ga % mort. MPN/g % mort. MPN/g
1 0 2.3x105 0 7x10-1 4
<0.3
6 0 1.5x104 4 4x10-1 4 <0.3
14 0 1.5x103 10 <0.3 16 <0.3
aMPN V. vulnificus/g values are presumptive on mCPC agar (1).
References
1. Elliot, E.L., C.A. Kaysner, and M.L. Tamplin. 1992. V.
cholerae, V. vulnificus, and other Vibrio spp., p. 111-140.
In: FDA Bacteriological Analytical Manual, 7th ed. AOAC
INTERNATIONAL, Arlington, VA.
2. Kilgen, M.B., M.T. Cole, and C.R. Hackney. 1988. Shellfish
sanitation studies in Louisiana. J. Shellfish Res. 7:527-530.
ICING AT HARVEST
Dr. Mark Tamplin
3031 McCarty Hall, Home Economics Program
Institute of Food and Agricultural Sciences
University of Florida
Gainesville, Florida 32611-0310
On-board (harvest vessel) icing has long been discussed as a
possible means to control the Vibrio vulnificus problem. Aside
from indications that levels of organisms in oysters taken straight
from the water may be sufficient to afflict at-risk individuals,
recent data from Florida show that icing oysters at the time of
harvest might be a viable means to limit bacterial growth between
harvest and processing.
Dr. Gary Rodrick and coworkers found that oysters iced at
harvest showed no increase in vibrio densities from harvest to
landing. However, in shellstock that was not iced, but simply
maintained under shade or wet burlap sacks for evaporative cooling
only, vibrio levels increased significantly during the four hour
period before reaching the processor. This occurred when water
temperature was warm, usually after April 15. Levels of fecal
coliforms and aerobic plate counts were also determined.
CONSUMER EDUCATION AND HEALTH ADVISORY PROGRAMS
Educational Outreach Efforts on Vibrio vulnificus
Ms. Ruth Welch
FDA Seafood Hotline, Consumer Affairs
Center for Food Safety and Applied Nutrition
U.S. Food and Drug Administration
200 C Street, S.W.
Washington, DC 20204
A question frequently asked is, "What are FDA and some other
groups doing to help educate the consumer on this very important
issue?" In earlier times, like the days of the Neanderthal,
foodborne hazards such as Vibrio vulnificus, Listeria
monocytogenes, Salmonella, and Escherichia coli 0157-H7 were
unknown. There were no warnings or health-related information on
potential risks when consuming rare or raw meats, eggs, or even
oysters. The risks were unknown, and the only risk management in
existence probably was a foreboding glance over the shoulder to
check for saber-toothed tigers.
Science today enables us to perceive foodborne risks where
perception was never possible before and where assessment or
management were possible only after the fact. The marked advances
in scientific knowledge and technology over the last half-century,
even the last five years in the case of V. vulnificus, have
heightened public expectations for a risk-free environment and
quick solutions to public health problems. While government is
expected to protect us from every risk possible, we still demand
every right of the Bill of Rights and sometimes more. This sets
the stage for the raw oyster and the accompanying V. vulnificus.
Undetected by sight, smell, or taste, difficult to detect, and
present in clean waters, it can cause death in a relatively small
population of middle-aged men and other selected at-risk
individuals. Despite all of the findings from investigations and
research on this now infamous bacterium, a small but significant
number of people continue to consume raw shellfish and die as a
result. What are officials in government, federal and state, and
local health professionals, doing to warn these individuals of
potentially mortal risks?
Summations (Tables 1 and 2) of an article published in the
March, 1994, issue of the Journal of the American Dietetic
Association (20) suggest that the delivery of information is as
important to consider as the information itself. The study
described was performed by Dr. Rodrick of the University of
Florida, and involved a group of 164 patients in five Florida
cities. The survey found that most participants got their
information on food safety from multiple sources, but the main
source was television and, predominantly, the show "20/20." This
raises a really significant question for us in the public health
field: "Where are people getting their information?"
Of people in the six high-risk groups surveyed (individuals
with hepatic disease, renal disease, HIV, cancer, diabetes, and the
elderly), only 26% recalled being instructed to avoid eating raw
oysters, and only 20% recalled being told to cook oysters before
consuming them. It is interesting that the survey was made in
Florida, the state reporting the largest number of oyster-borne V.
vulnificus mortalities. The results of this study encouraged
health professionals, especially dieticians and physicians, to
provide community education on consumption of raw shellfish to
their high-risk patients. It also suggested possibly providing
warning labels as an educational activity.
Considering the information on food labels and packaging
today, it is interesting that people even get information on food
consumption from the Bible, which used to serve as such a resource.
So, it is important to understand where people obtain their
information, and this is a factor which merits serious
consideration by this group.
The FDA has been at the forefront of education on V.
vulnificus for the past ten years. FDA Consumer Magazine has a
circulation of about 22,000, and it has been one of the prime
methods of consumer education by the agency. Articles often are
published, reprinted, and made available from FDA Public Affairs
Specialists across the country; these articles are available for
reproduction and unlimited use. FDA Consumer Magazine published
the first article produced by the FDA discussing Vibrio pathogens
in October of 1984 (1), almost ten years ago, when this species was
first recognized as a public health hazard. In this article, V.
vulnificus is described as "The species most likely to kill...",
having about a 40% mortality rate in people with liver disease.
The article also outlines safe handling procedures for shellfish,
principally keeping them chilled and cooking them thoroughly.
In February, 1985, another article entitled "Pollution Narrows
the Shellfish Harvest" (2) also informed those in the high-risk
groups of the public health hazard represented by this species. In
addition, the April, 1985, issue of the FDA Drug Bulletin, a
publication circulated to over 1.1 million physicians and other
health professionals, published an article called, "Vibrio
vulnificus in Patients with Liver Disease" (5). This issue reached
over a million doctors.
Recalling some discussion earlier today on the capability
physicians have to communicate information to patients, it must be
recognized that they continually receive a great deal of
information from government. Whether each piece of information is
read and then passed on is another question.
An article in FDA Consumer (1987) entitled "Fewer Months Are
Safe for Eating Raw Gulf Oysters" (15) characterized victims of V.
vulnificus as most frequently men over 40 who ate raw or
undercooked oysters between April to October, had liver disease or
were heavy consumers of alcohol, had iron or other blood disorders
or immune system disorders.
A 1988 article in FDA Consumer entitled "Weighing the Risks of
the Raw Bar" (3) further described the various health conditions
which put people at risk such as cancer, diabetes, liver disease,
chronic gastrointestinal disease, and immunity deficiencies.
People so afflicted were advised to cook molluscan shellfish before
eating it. Also in 1988, an article in the FDA Drug Bulletin
entitled "Concern Continues About Vibrio vulnificus" (6) was
distributed to over 1.1 million physicians and health
professionals. In 1989, an FDA Consumer article called "Fishing
for Facts and Fish Safety" (17) discussed risks of illness and
mortality from Vibrio organisms for individuals in high-risk
groups.
The FDA Seafood Plan for the fiscal year 1991-1992 addressed
the issue of alerting high-risk individuals in a four-part
approach. Three types of publications were printed and
distributed, an article in FDA Consumer (16), an FDA "backgrounder"
(21) on seafood safety, and a series of four brochures (9-12) for
high-risk individuals. In addition, FDA's Seafood Hotline was
implemented (4). The article published as an FDA "backgrounder"
was disseminated to the press, and the four "high-risk" brochures
were developed and distributed upon Hotline request and were also
sent to targeted groups affiliated with human conditions placing
people at high risk. The brochures on seafood safety related to
liver disease, diabetes, eating disorders, and gastrointestinal
disorders are published in different colors and, to date, FDA has
distributed 160,000 sets of these brochures, including 67,000 to
FDA Public Affairs Specialists in the field for distribution, and
about 3,000 to contacts on the Seafood Hotline. These brochures
are available in black and white slicks, which were sent to the 50
state agencies. States may order multiple copies or reproduce the
slicks as needed.
The FDA Seafood Hotline was activated in October, 1992. As
part of the Agency's outreach to educate consumers, the Hotline to
date has received 42,000 calls, most with questions on seafood
safety, purchasing, handling, storage, and nutrition. The Hotline
is accessible 24 hours a day, and has provided more than 25,000
messages on seafood safety to callers. Callers can speak directly
with a Public Affairs Specialist between noon and 4 PM during the
week, and they can request FDA publications by mail or fax 24 hours
a day.
In addition to the seafood safety brochures, FDA also provides
two other publications specifically concerned with V. vulnificus.
The Hotline has a prerecorded message on consumption of raw
molluscan shellfish for high-risk consumers, and during the first
1.5 years of Hotline activity, over 12,000 prerecorded messages and
publications on eating raw shellfish were delivered to callers.
Because of the amount of publicity on FDA's Hotline, calls are
sometimes received directly from physicians and patients,
especially patients with liver disease.
One Hotline call came from a young lady who had undergone two
liver transplants and whose occupation was shucking shellfish at a
fish market. She had been instructed by her physician, who had
read information on risks from V. vulnificus in the FDA Medical
Bulletin, to contact FDA and inquire. She was referred to Dr. Karl
Klontz of FDA, who advised her that perhaps she should work at the
cash register or do something else with less exposure and risk
(13). FDA has received many calls from physicians requesting
Agency publications and brochures. If specific questions cannot be
readily accommodated, callers are referred to the appropriate
individuals for answers.
FDA issued a Talk Paper on February 8, 1993, entitled "Advice
on Consumption of Raw Molluscan Shellfish" (8). The paper
reaffirmed the FDA's policy on the consumption of raw shellfish,
and this message too is available via the Hotline. In March 1993,
an article entitled "To prevent Vibrio infections, high-risk
patients should avoid eating raw molluscan shellfish" (7) appeared
in the FDA Medical Bulletin. The article identified the high-risk
groups, highlighted the symptoms of Vibrio for doctors who may not
be familiar with them, mentioned some treatment for Vibrio
illnesses, and directed physicians to call or write for the
brochures for their patients. Again, over a million physicians
received this bulletin.
Also early in 1993, the FDA convened a working group to plan
strategies for communicating with the high-risk populations.
Articles were written for professional journals and FDA
publications, and a package of materials was compiled which
included the four "high-risk" advisory brochures, a reprint from
the FDA Medical Bulletin, a camera-ready newsletter article, and
information on the FDA Seafood Hotline. Folders containing these
materials were sent to 1,200 multiplier organizations, including
1,700 high-risk disease associations, 1,200 members of the National
Council on Alcoholism and Drug Dependence, 210 state and local
public health agencies, 81 AIDS organizations, and 25 consumer
activist groups. Distribution through multiplier groups and by
newsletters were selected because they afforded the most efficient
means to inform target audiences. Some of the groups receiving
these materials included Alcoholics Anonymous, American Diabetes
Association, American College of Gastroenterology, American Cancer
Society, the Hemochromatosis Research Foundation, American Medical
Association, American Dietetic Association, and many more. This
outreach effort was projected with the potential to reach two to
three million high-risk individuals. Clearly, FDA believes it is
very important to reach these target audiences.
A comment card was included in the package of materials
distributed for respondents to return by mail upon use of the
packet. From the cards returned, it was learned that the outreach
materials were used in several trade and employee newsletters,
resource centers, and direct mailings to the constituents.
At the same time, part of the Agency's overall outreach effort
provided articles to medical journals aimed at informing medical
professionals not otherwise reached. The articles were published
in Pharmacy Times (14), circulation 98,000; the Journal of the
American Medical Association (18), circulation 350,000; and
American Family Physician (19), circulation 144,000. These four
journals together with the FDA Medical Bulletin reached nearly two
million health professionals last year with important information
on the hazards from Vibrio.
The working group for this outreach program continues to
monitor its efforts and is contacting these groups again for
further assistance and feedback. The FDA Office of Health Affairs
also is working with the U.S. Pharmacopeia's patient education
program for possible patient information sheets on V. vulnificus.
FDA Public Affairs Specialists across the nation have also been
conducting a major seafood education initiative during the past few
years. Highlights include speeches, workshops, exhibits, media
events, and many cooperative efforts in developing consumer
education materials with various groups, such as National Oceanic
and Atmospheric Association, Sea Grant, universities, medical
societies, and consumer education groups.
In addition to those available from the FDA, other resources
related to Vibrio hazards are available to consumers and health
professionals. In July, 1993, the ISSC Education Subcommittee,
under the direction of Dr. Steve Otwell from the University of
Florida, compiled a listing of informative materials available to
consumers and professionals through the various federal and state
agencies. While most of the material comes from the State of
Florida and FDA, some other materials are available through various
trade organizations. Additionally, the Environmental Protection
Agency has a project entitled "The Public Health Action Agenda for
the Gulf of Mexico" (22), which has a component on educational
materials for consumption of raw shellfish.
A brochure on V. vulnificus developed by Florida Sea Grant and
the University of Florida is widely used; more than 10,000 have
been distributed both in and outside the State of Florida, and FDA
uses it on the Hotline.
During the ISSC survey on available materials, Dr. Otwell
found that only nine states had educational materials readily
available for consumers: Florida, Georgia, Louisiana, Mississippi,
Missouri, South Carolina, California, North Carolina, and New York.
Most of these states had not developed their own materials, but
were using materials from the State of Florida and FDA.
The presence of warning labels on raw molluscan shellfish has
been proposed by several consumer groups over the years, and
recently (May, 1994), this was proposed in House hearings.
Currently, warnings are required in Louisiana, Florida, and
California. Data are not yet available on the effectiveness of
these warning labels, but preliminary results from the Food Safety
Survey at FDA's CFSAN indicate that risk-taking individuals may not
heed warnings.
To understand this, it is useful to examine the profile of a
raw seafood consumer. It is similar to that cited for risk takers
generally, and eating raw seafood may be a deliberate risk-taking
behavior for many consumers. Many raw seafood consumers are
informed of the risks but are not deterred. Thus, warning labels
may not deter a significant segment of raw seafood consumers.
These are the conclusions stemming from study results for about
1,150 respondents. This is an important consideration when
discussing the value of scientific and outreach endeavors for
protecting consumers from potential hazards in raw molluscan
shellfish.
The 1993 edition of the FDA Food Code (23) is a reference
guide that advises retail outlets such as restaurants, grocery
stores, and institutions on how to prepare food to prevent
foodborne illness. Section 3-603.11 of this guide is a consumer
advisory on eating raw animal protein. In essence, it states, "If
a raw or undercooked animal food such as beef, eggs, fish, lamb,
milk, pork, poultry, or shellfish is offered in a ready-to-eat form
such as a deli, menu, vended, or other item or as a raw ingredient
in another ready-to-eat food, the permit holder shall inform
consumers by brochures, deli case or menu advisory, label
statements, table tips, or other effective means of the significant
increased risks associated with certain, especially vulnerable,
consumers eating such foods in raw or undercooked form."
This consumer advisory is recommended by FDA, and states can
choose to adopt the Food Code in part, in its entirety, or not at
all. This type of advisory, if adopted by states, could play a
significant role in the outreach education effort aimed at
individuals who consume raw molluscan shellfish. At the moment
there is no standard language for this advisory, but that is being
worked on at FDA. Some of the components of the language
requirements are that the high-risk individuals should be
identified, the types of food should be identified, and the public
health advice regarding these food products needs to be identified.
In March, 1994, the Conference of Food Protection met and
considered a request by FDA on advising how best to implement the
Food Code advisory provision. This matter has been referred to a
committee for response.
The ISSC Education Subcommittee, which met last August and
again this February, discussed the directives of the Executive
Director, including one of the prime directives to examine the
available educational materials and to develop an educational
packet for distribution to the states. The Committee reviewed the
list compiled, determined that sufficient educational materials
already existed, and decided to develop the educational packet from
them. It also discussed a second directive: to develop a position
on the Consumer Advisory in the 1993 Food Code. The Committee
reiterated its recommendation given at the 1993 annual meeting that
there be no specific mandatory warning on raw molluscan shellfish
at the retail level. However, it was felt that if the Consumer
Advisory message became mandatory, the ISSC should support a more
generic statement which would be inclusive for all raw animal
protein foods of concern rather than a statement addressing only
raw molluscan shellfish.
The FDA, the states, and other organizations have been
publishing materials for consumers, health professionals, and
target multiplier groups for the last ten years, and quite
intensely during the last two years. All educational materials
generated within the CFSAN now undergo field testing, and questions
on how to measure the effectiveness of the educational materials
and messages are continually being raised. Is information reaching
the right people, those at the highest risk? Will receiving
information change the behavior of those individuals who are at the
highest risk, considering the profile of the risk taker? Is
information more effectively delivered by a physician, dietician,
pharmacist or other health professional, or when seen in a
newspaper or on television? Should advisory messages be mandated
for a small number of potentially high-risk individuals? Finally,
government must reflect on the cost-benefit of all of these
efforts. Are those people that need most to be reached receiving
the information, and will this change their behaviors?
Table 1. Participant demographics
Mean age No. recalling
High-risk q standard Number of food safety
group deviation individuals men women counselinga
Hepatic disease 40 q 15.6 30 17 13 9
Renal disease 39 q 18.8 18 6 12 4
HIV infection 36 q 9.4 62 53 9 21
Cancer 56 q 14.4 21 7 14 4
Diabetes 55 q 20.0 4 2 2 1b
Elderly 84 q 3.5 29 9 20 4
Total 164 94 70 43
aSources of counseling shown in Table 2; number of individuals
who remembered receiving food safety instructions to avoid
eating raw oysters.
bSmall sample size suggests numbers may not be indicative for this
group.
**********
Table 2. Sources of information for individuals
recalling food safety counseling
Sources of information No. of individuals (%)
More than 1 answer 17 (37.8)
Television 8 (17.8)
Dietitian 4 (8.8)
Physician 3 (6.7)
Magazine 3 (6.7)
Newspaper 3 (6.7)
Nurse 2 (4.4)
Friend 2 (4.4)
Othera 3 (6.7)
Total 43 (100.0)
aRelative, Bible, personal beliefs.
References
1. Ballentine, C. 1984. For oyster and clam lovers, the water
must be clean. FDA Consumer (October).
2. Ballentine, C. 1985. Pollution narrows shellfish harvest.
FDA Consumer (February).
3. Ballentine, C. 1988. Weighing the risks of the raw bar. FDA
Consumer (June).
4. Bradbard, L. 1993. Seafood hotline responds to consumer
needs. FDA Consumer (September).
5. FDA Drug Bulletin. 1985. Vibrio vulnificus in patients with
liver disease (April).
6. FDA Drug Bulletin. 1988. Concern continues about Vibrio
vulnificus (April).
7. FDA Medical Bulletin. 1993. To prevent Vibrio infections,
high-risk patients should avoid eating raw molluscan shellfish
(March).
8. FDA Talk Paper. 1993. Advice on consumption of raw molluscan
shellfish (February 8).
9. Important health information for people with diabetes
mellitus. DHHS Publication No. (FDA) 93-2264, May, 1993.
10. Important health information for people with gastrointestinal
disorders. DHHS Publication No. (FDA) 93-2265, May, 1993.
11. Important health information for people with immune
disorders. DHHS Publication No. (FDA) 93-2267, May, 1993.
12. Important health information for people with liver disease.
DHHS Publication No. (FDA) 93-2266, May, 1993.
13. Klontz, K.C. 1994. Personal communication.
14. McGinnis, T.J. 1993. Advice to high-risk patients about raw
molluscan shellfish. Pharmacy Times 3:100-101.
15. Miller, R. 1988. Fewer 'R' safe for eating raw Gulf oysters.
FDA Consumer (June).
16. Miller, R. 1992. Get hooked on seafood safety. FDA Consumer
(June).
17. Modeland, V. 1989. Fishing for facts on fish safety. FDA
Consumer (February).
18. Nightingale, S.L. 1993. From the Food and Drug
Administration. J. Am. Med. Assoc. 272:882.
19. Rheinstein, P., and K.C. Klontz. 1985. Shellfish-borne
illness. Am. Family Physician 47:1837-1840.
20. Ross, E.E., L. Guyer, J. Varnes, G. Rodrick. 1994. The
necessity of education for high-risk individuals. J. Am.
Diet. Assoc. 94:312-314.
21. Seafood safety. 1991. FDA Backgrounder (February).
22. U.S. Environmental Protection Agency. 1993. Public Health
Action Agenda for the Gulf of Mexico. Office of Water, Gulf
of Mexico Program, Stennis Space Center, MS.
23. U.S. Food and Drug Administration. 1993. Food Code. U.S.
Department of Health and Human Services, Public Health
Service, Washington, DC.
Questions, Answers, and Comments
C. Ms. Smith-DeWaal: Ms. Welsh, it's great. I'm really thrilled
to hear about the consumer efforts that you are taking. Maybe
the problem, the fact that you're not getting your message
across adequately enough to these physicians and these high-
risk groups who somehow aren't aware of the problem, is
because in your attempts to make the message so generic and so
nonalarmist that it's just not getting people's attention.
Maybe I could help. Maybe you need to talk about seafood
roulette because that is, in fact, what FDA is asking people
in these high-risk groups to play. Because if they don't get
your little pamphlet or if they don't get the information from
their doctor, if they don't get this information before they
get that raw oyster, they could be dead. It really concerns
me, particularly, to hear the information you've talked about
on risk takers and the fact that raw shellfish consumers are
somehow risk takers. This is similar to information that was
given out in the early '80's about people who might be at risk
for AIDS.
A. Ms. Welsh: This was a study that was just recently done, and
it is a scientific study.
C. Ms. Smith-DeWaal: That same study, at least the preliminary
results that I saw on that study, said that in the four -- I
believe it was four years -- between the time that the initial
telephone survey was taken and the most recent time it was
taken, that consumer concerns about seafood safety had doubled
from 7% to 15%. Those are people who thought that someone in
their household had gotten ill. I think this stuff about risk
taking and to hear the Agency -- you're not the first person
at this meeting from the Agency to talk about whether
shellfish consumers are risk takers. I think that is
meaningless when you are talking about getting information out
to people so they can protect themselves. Some people aren't
risk takers. One of the people who died was an 80-year-old
man, and his daughter is very active. She wanted to be here
today to try to bring this and the very personal nature of
this issue to this gathering. He was not a risk taker and I
believe many, many other victims aren't risk takers. I think
we need to be very serious when we look at this issue and not
try to lump people into one group and say, "Therefore, we
don't really need to warn them because they wouldn't take our
advice anyway." Let's give them the advice and get the
information out there. I think a warning label is the most
appropriate method to do that. Thank you.
RETAIL FOOD SAFETY
The Retail Food Protection Branch and
Its Role with Regard to Vibrio vulnificus
CDR Lawrence C. Edwards, USPHS
Retail Food Protection Branch
Center for Food Safety and Applied Nutrition
U.S. Food and Drug Administration
200 C Street, S.W., Washington, DC 20204
Seafood in Retail Stores and Restaurants
The Food and Drug Administration (FDA) works through
cooperative mechanisms with state and local regulatory food
agencies in several ways to promote food protection for consumers.
It specifically provides technical assistance with food safety
issues; provides training through various avenues, either from our
office, through the State Training Branch, or through our Regional
Food Specialists; administers a comprehensive standardization
program for inspectors; and performs state food protection program
evaluation.
New Business Format
In the future, the model codes (Food Service, Retail Food
Store, or Vending Codes) will no longer be used as a basis for food
protection activities. Instead, the Food Code 1993 (1) will be
used.
Retail
Retail is the first point of contact for most consumers. It
is appropriately and collectively our job to do everything possible
to control what goes on behind the counter regarding food safety.
Consumers do not have the opportunity to look at what goes on
behind the food display. It was mentioned that in Florida, 85% of
individuals who died from V. vulnificus infection consumed the
oysters in restaurants. One serious concern is that the retailers
really do not know if they are receiving a good product or a bad
product. They are not able to track that product from harvest
through the distribution system to their establishment. They
cannot see what temperatures have been maintained, what protection
measures have been taken, and what handling practices have been
used. Again, we all have a larger role to routinely monitor
operations and prevent mishaps from occurring.
Hazard Analysis Critical Control Point (HACCP) Programs
More industry and regulatory food agencies are training
themselves in the HACCP concept at all levels of food service.
Today we heard of some new critical control points with regard to
oysters going from harvest through distribution. Washing hands, a
provision that has been highlighted in the Food Code (1), is an
example of a HACCP control. FDA is looking for controls that build
barriers against foodborne disease and that work; HACCP can help us
accomplish that.
Seafood at Retail
Since our 1988 Workshop on V. vulnificus, the Retail Food
Protection Branch has been working closely with our shellfish
people. The creation of the Office of Seafood, the Shellfish
Program Implementation Branch, and the Seafood Hotline have all
provided retail with new resources to contact. In 1991, the Joint
Seafood Committee was formed out of our office. With membership
from both of the areas covering food and shellfish, i.e., the
Conference of Food Protection and the Interstate Shellfish
Sanitation Conference, coordination of program design has improved.
With their assistance, FDA successfully conducted two national
pilot programs in retail from 1991 to 1993.
First, FDA went to 22 retail stores in 12 different states,
asked them to develop HACCP plans and to try to comply with those
plans. The focus was on their most critical points of the seafood
operations. The physical application was primarily in the fish
departments, covering fresh and frozen products.
Then, in 1992 and 1993, FDA returned and worked with the
National Restaurant Association, with 12 different restaurants in
11 states, focusing on seafood entrees, but including their entire
operation. This focus was very different from the retail food
store pilot, because when different people work in different areas
with the same product, the risk for mistakes goes up considerably.
FDA was able to convince the restaurants to write HACCP plans for
their seafood products and, by the end of the pilot, to be in
compliance with their plans. The pilots provided a definite
learning curve for everyone.
Food Code 1993
In 1993, FDA printed the Food Code 1993 (1) and made it
available through the National Technical Information Service
through an announcement by DHHS Secretary Donna Shalala. The Food
Code is now out and is being reviewed and considered for adoption
by a number of states and federal agencies. An early copy was
issued so that states that were considering new regulations could
conduct a quick review of those provisions.
In the past, the model codes were vague with regard to seafood
or shellfish and were difficult to enforce. They were not detailed
enough. Clearly, we have improved the detail by going from
approximately a 78-page to a 450-page document. Although the
provisions are contained in only 178 pages, there are other
significant components such as the 15-page index that uses key
words for conducting a search. Seven different annexes go into
specific areas of importance to support the provisions of the Food
Code such as Establishment Inspection. The new Food Code is a much
more state-of-the-art document.
In the Food Code, the National Shellfish Sanitation Program
(NSSP) provisions from the Manual of Operations, Part II, are
reiterated verbatim. Everything from receipt of product to proper
tag storage is contained in the Food Code. Other things, like
temperature requirements, have changed from the former model codes.
The FDA requirement has gone from 45xF to 41xF (5xC). Items like
cooking temperatures are covered. Included under seafood are
steaks and filets and other types of fish and crustaceans, and raw
molluscan shellfish. The Code recommends that they be cooked at
145xF (63xC) for 15 seconds. Instead of just temperature, we have
added cooking time. Established NSSP shipping temperatures are
also covered.
Of all of the issues in the Food Code, the Consumer Advisory
is among the most controversial. The requirement describes what
the Consumer Advisory should provide to the consumer. Clearly, at
a minimum, the consumer should receive the following messages.
First, the hazards associated with consumption of undercooked
product should be given and the more vulnerable populations
identified. A list is given of people who are likely to become
seriously ill from foodborne illness, including those with disease
conditions that are vulnerable to V. vulnificus and other pathogens
associated with raw or undercooked product. Second, it requires
that the types of undercooked or raw food being served in an
establishment need to be listed for the consumer. Third, the
public health advice regarding consuming these types of food
products must be conveyed. An example basically states that
"public health officials and food safety experts advise that
thoroughly cooking or pasteurizing food from animals such as dairy
products, meat, eggs, poultry, seafood, provides an extra level of
safety for all consumers. However, certain consumers are more
likely to become seriously ill from foodborne illness. These
consumers should always order and eat animal foods thoroughly
cooked or pasteurized." The proposed Consumer Advisory format
explains what needs to be in that Consumer Advisory.
Conference for Food Protection
The Conference for Food Protection said, "Wait a minute. That
is an issue that is going to cause a lot of discussion. We are
going to set that issue aside with other issues like double-washing
your hands and nail brushes and things like that for further review
and discussion." The Conference pulled these issues directly off
and formed a group of committees to further deliberate these
issues. We want them to deliberate those issues with all parties
represented, so that a conclusion that will be acceptable to all
parties can be reached before the 1996 Conference.
Industry Perspectives
The Food Marketing Institute membership includes about 70% of
the retail food stores in this country. The Institute, for some
time, has used consumer advisories in their member stores, and
without difficulty. They have different advisories for meats,
poultry, and undercooked products, in order to warn the consumer.
The advisories are provided at the point of sale. Not all of their
membership participates, but a fair number of them do. They are
very comfortable with what is proposed in the Food Code.
However, the National Restaurant Association views it a little
bit differently. They believe in the concept that "The consumer
should be warned. But the warning should be a friendly reminder."
They think that some of the language just provided to you is too
negative. It embodies too much of a bad implication. They believe
that going to restaurants for dinner with your family or friends
should be an enjoyable occasion, not a sit-down to a warning. In
discussing this and other issues with the National Restaurant
Association, FDA is saying, "Let's sit down at the table and talk
about what we can say that will accomplish the goal, and the
language that will be acceptable to you.
The National Academy of Science (NAS) Report
The NAS report covered all of the different segments of the
seafood process. Generally, a few major issues concerned retail.
One issue was cross-contamination, another was that state and local
regulatory food agencies should work together to produce better
food protection, and, finally, the report recommended federal
oversight of the food protection program so that better protection
is afforded the consumer. The Food Code 1993 (1) addresses all of
these recommendations.
Additionally, an annex of the Food Code lists the public
health reasons behind the provisions and why they were written as
they were. Cross-contamination is addressed there.
The Interstate Certified Shellfish Shippers List (ICSSL)
A recent change in the Shellfish Program concerns the
processing plants listed on the ICSSL. The standardization process
with state inspectors has been completed by FDA, and some of the
plants are no longer on the ICSSL. FDA is coordinating the use of
the ICSSL during inspections. In the near future a letter will be
sent to all state food program officials promoting the use of the
ICSSL. By contacting these program officials, FDA is reminding
them that some of the plants have come off the list. It is
important for retailers to take an even closer check to ensure that
the tag on each produce container is from a certified source.
FDA also is developing a protocol on what an acceptable
procedure would be if an uncertified source product is found. A
National Marine Fishery Service study about two years ago stated
that 13 countries were successfully shipping molluscan shellfish
into the United States from uncertified sources. These were not
countries having a Memorandum of Understanding with FDA, which is
required by the NSSP. If a product is, in fact, getting into the
country uncertified, what are we doing about it? A number of state
and local jurisdictions may not be as forceful as they should be
when discovering uncertified product. FDA is working with the
shellfish people across the country to tighten up controls at
retail, to initiate accepting only certified source product in the
distribution chain, and to allow only certified product in
commerce.
Liability
In the Food Code 1993 (1), there are new requirements for a
person in charge (PIC). FDA has never really stated this as
clearly as it has this time. One person must be identified at each
establishment who is responsible during all hours of operation, and
that PIC is responsible for implementing the provisions of the Food
Code. In the Food Code, the PIC is legally liable for meeting the
Food Code requirements.
Infected employees are also identified as liable in that
operation. If infected employees fail to tell the PIC or manager
that they are sick, or if the manager does not take steps to remove
them or restrict them from the operation under specific situations,
these people are liable. This provision of the Food Code will
change things considerably.
Educational Needs
The educational needs associated with the new business format
are immense. FDA is "swamped" trying to meet the informational and
educational needs. The Retail Food Protection Branch and our
retail food specialists around the country are loaded with training
requests on the Food Code. There are so many different provisions
in the Food Code; people are truly fearing the document. It is
quite large, and regulators need training on what the provisions
are and on how to enforce them. The Retail Food Protection Branch
has actually committed to 40 national meetings to present Food Code
information, all before October 1 of this year. At each of these
meetings, we will present a 1-5 hour presentation on the Food Code.
FDA's Regional Food Specialists are doing the same thing. For
all of the states, the Specialists are responsible for giving two,
three, and four different training sessions. It is a tremendous
amount of training, a major transfer of information, and a major
task ahead of FDA. However, it is definitely an exciting challenge
for us all.
Reference
1. Food Code 1993. U.S. Department of Health and Human Services,
Public Health Service, Food and Drug Administration,
Washington, DC.
Questions and Answers
Q. Dr. Hlady: How many states currently adopt the Food Code in
total?
A. Mr. Edwards: To be honest with you, I don't think any state
has adopted it to date. However, at least five states are
moving through legislature right now to propose just that.
The U.S. Air Force adopted it in February. They even added a
few provisions of their own. So, we do have some people who
are really taking charge and taking a good hard look at it.
Some of those controversial issues I mentioned are holding
states up from adopting it in its entirety. They don't like
things like the "Consumer Advisory." They don't know how they
are going to enforce it. The double handwash? I don't know if
you know about that, but we require it when people do certain
things. Going to the bathroom and returning, they not only
have to wash their hands once with a nailbrush, but twice at
20 seconds each. Things like that are holding up some of the
states from adopting it in its entirety.
Q. Dr. Hlady: What kind of enforcement responsibility does FDA
have? Are you talking about this just being government
institutions?
A. Mr. Edwards: Where FDA does have direct regulatory authority.
We have a program called the Interstate Travel Sanitation
(ITS) Program. It's a lot like the CDC Vessel Sanitation
Program, where we govern the interstate conveyances. That is
an area that we are proposing to go into and directly enforce.
The only thing we have to do is get our ITS people trained,
get the industry geared up for receiving that, give some
implementation time, and then begin to use it. But we haven't
been approached by the National Restaurant Association. They
want us to go into contract care nursing homes and they want
to set up a direct voluntary program where we use the Food
Code in the near future; sometime this calendar year, I think.
We've got a couple of others who are asking us to come in and
start regulating that fairly soon.
Q. Ms. Smith-DeWaal: Where did you come up with that list? I
noticed there were things like raw beef, raw pork, and raw
poultry. Where did you come up with that list of raw food
items?
A. Mr. Edwards: We know that any of those raw protein foods can
cause illness. The E. coli situation is one that popped up
and surprised us even though Washington State had an
enforcement temperature of 155xF. This includes consumption
of any of food, not just shellfish. People need to recognize
that whether it be steak tartare or Caesar salad, it can
honestly make someone at risk very ill. Basically, our job is
to cover all foods, and that's why this Consumer Advisory must
not leave anything out. I want to tell you also that we are
working directly with CDC, which has been giving us advice.
CONCLUDING REMARKS
Summary of Information from the Scientific and Technical Update
Dr. William D. Watkins
U.S. Public Health Service
Northeast Seafood Laboratory
FDA, Bldg. S-26, School Street, Davisville
North Kingstown, Rhode Island 02852
A large amount of new scientific and technical information has
been presented during this update session, and it is fairly obvious
that a great deal has been learned over the past six years. How
well have the research and technical needs identified at the 1988
Workshop been addressed? While time does not permit a full review
of this issue at this juncture, it would serve us well to at least
summarize some of the information just presented, in order to focus
on the next workshop task, that of identifying further information
needs. Accordingly, this wrap-up is intended simply to summarize
some of the major scientific findings and conclusions presented and
a few of the prinicpal needs remaining for each of the topics
covered.
Epidemiology
Scientific Update
Oysters from the United States Gulf Coast waters are the
vehicle of concern, especially during the period of April
through October.
High-risk individuals are the problem.
Enhanced surveillance and more complete epidemiological data
are available from the Gulf Coast Surveillance Program.
Perhaps as few as 1,000 V. vulnificus per gram can be lethal.
Information Needs/Activities
Carefully examine infection cases, as well as mortalities, to
gain insights on the hazard and host factors involved.
Investigate and document the likelihood that numbers of cases
for infections are probably woefully inaccurate.
Determine the incidences of infection derived from shellstock
other than Gulf coast oysters.
Considering the worldwide estuarine distribution of V.
vulnificus, find out what other seafoods have transmitted
illness and caused mortality.
Risk assessment
Scientific Update
Based on consumption and other population data, a plausible
estimate of the risks for death, and perhaps infection, can
be made.
Healthy consumers are essentially not at risk for mortality,
and the level of risk for compromised individuals is
extremely low.
Some people do not know their states of health are
compromised, placing them at higher risk.
Information Needs/Activities
What levels of V. vulnificus are hazardous to consume?
What levels of V. vulnificus are safe to consume?
Acquire better and more current consumption data.
Acquire accurate data on infections.
Case studies
Scientific Update
Freshly harvested (one to two days) oysters can be lethal.
Host predisposing factors are critical.
Information Needs/Activities
Require more complete epidemiological information from each
case.
More aggressively seek certain information not being readily
obtained.
Pathogenicity
Scientific Update
Virulence factors causing enzymatic tissue damage aid in
invasiveness, but are not critical to virulence.
Encapsulation (opaque) is a key factor for invasiveness,
providing resistance to phagocytosis.
Ferritin iron saturation levels appear related to host
susceptibility.
V. vulnificus lipopolysaccharide (LPS) appears to contribute
significantly to lethal virulence factor, leading to
symptoms of endotoxic shock.
Animal and human pathogenesis exhibit similarities and
differences.
Information Needs/Activities
Refine and expand investigations to determine key
differences related to strain virulence.
Expand studies on virulent strain relatedness.
Methodology
Scientific Update
Reliable DNA probes and monoclonal antibodies have been
developed with moleular techniques for specifically
identifying isolates of V. vulnificus.
Advances to differentiate and relate isolates of V.
vulnificus have been made using PCR techniques, 2-D pulsed
field gels, serotyping, ribotyping, capsule typing, and
restriction fragment length polymorphism (RFLP).
Information Needs/Activities
Consensus and collaborative testing of specific, direct
enumeration method to foster widespread use.
Further development, application, and collaboration on
molecular techniques to relate and distinguish strains,
particularly with regard to virulence.
Viable But Nonculturable
Scientific Update
VBNC state is triggered by cold stress, and enables survival
of V. vulnificus during winter months.
Specific proteins are produced in response to cold stress.
Resuscitation and outgrowth from the VBNC state occur in sea
water, oysters.
Information Needs/Activities
What threat do VBNC V. vulnificus represent to consumers?
What threat does resuscitation of VBNC V. vulnificus
represent to consumers?
Ecology
Scientific Update
Benthic environments and fauna such as fish and plankton
appear to provide abundant niches for V. vulnificus in the
estuarine environment.
Abundance of V. vulnificus appears related to water
temperatures and salinities.
Temperature-salinity matrices are being developed for
several United States coastal regions and these may
ultimately provide indices for predicting water or oyster
densities of V. vulnificus.
Information Needs/Activities
Additional field results are needed to refine the salinity-
temperature matrices, and verify their reliability for
predicting V. vulnificus abundance.
Time-Temperature
Scientific Update
Current ISSC requirements (refrigerate within 20 hours
during summer) are inadequate for controlling V. vulnificus
growth in harvested oysters from Gulf Coast waters during
warm seasons.
Increases in V. vulnificus densities in summer-harvested
oysters occurs within a few hours and can exceed ten-fold
within 12 hours.
On-board icing is effective for controlling V. vulnificus
growth in harvested oysters.
Information Needs/Activities
Definitive growth data.
Intervening Measures
Scientific Update
Depuration is not effective for lowering V. vulnificus
levels in oysters.
Gamma radiation can effectively lower V. vulnificus levels
in shellstock oysters, but also decreases shelflife.
Freeze-thaw process can effectively lower V. vulnificus
levels in half-shell oysters, but does not eliminate the
pathogen entirely.
Information Needs/Activities
Irradiation data to obtain FDA premarket approval of
petition for irradiation processing.
Investigate potential of temperature and/or salinity
controlled wet storage to reduce V. vulnificus levels in
shellstock oysters.
Educational Outreach
Scientific Update
Disseminated information is not yet effective enough.
Warning labels are not effective enough.
Information Needs/Activities
Specifically warn the high-risk population, and educate all
consumers on the risks from eating raw animal protein.
Develop and implement reliable means to evaluate the
effectiveness of education efforts.
Retail Food
Scientific Update
Food Code provisions comprehensively describe means for
retailers and consumers to address and avoid hazards
introduced after harvest.
Hazard Analysis Critical Control Point (HACCP) programs
place responsibility for controlling shellfish hazards on
harvest and handling of shellfish.
Information Needs/Activities
Overcome lack of industry consensus on specific labeling
language.
Obtain consensus by industry and adoption by states of the
Food Code or equivalent provisions.
Implement and refine HACCP.
Comments on June 16, 1994 Sessions
Tomorrow's sessions will be conducted in a panel format.
Discussions will center on identifying information and research
needs to better address the hazard of V. vulnificus. After panel
members summarize their individual views on further needs, time
will be devoted to audience participation in the discussions. The
approach is for each panel individual, starting with the
moderator, to identify the information needs, efforts, and means
to pursue those needs that should be priorities. Following this,
the moderator can interact with the audience and direct a
vigorous, open discussion.
DISCUSSION
Questions, Answers, and Comments
Q. Mr. Thompson: I had one comment about your earlier list.
One of the things that I saw mentioned in the data that was
not in your group was questions about the host, the human
aspect. I think that definitely needs to be in there. I am
aware of a number of cases, at least two in Texas, that I
have investigated, where the individual routinely consumed
shellfish on a periodic basis and something changed, and the
patient died. There are some host questions here that need
to be answered, and I didn't see that in our list. They
were at risk and had been at risk for some time and had
routinely consumed raw oysters for some time, and then on
one occasion they consumed raw oysters and before they knew
it, the next week they were dead.
A. Dr. William Watkins, FDA: Interestingly, today, we heard
mentioned that consumers perhaps can have a level of
immunity develop. If they routinely consume oysters and
then stop for awhile, maybe the antibodies or whatever is
involved have disappeared and they've lost that level of
immunity.
C. Mr. Thompson: I think that should be definitely included.
I want to make one other comment, if I might, as Richard
Thompson, Chairman of the ISSC and Chairman of the 1988
Workshop. I want to remind everybody, while I am not by any
means belittling any of the efforts that we are trying to
undertake or that we have undertaken, I want to remind you
that one of the things that we left that conference with was
a definite concern that because we were starting to look at
this issue, we would suddenly see a dramatic increase in the
number of cases. When you start turning over every rock and
looking in every cranny for cases, you frequently get a
rapid increase simply because you are reporting process.
Even though we have done that and we now have a reporting
process in place for all of the states that are involved, we
have not seen an increase in cases. There is another factor
here somewhere that is playing into this thing. We don't
have our hands on it yet; we've got to get our hands on it.
I'm not saying we don't need to do anything more, but we did
not see that dramatic increase in cases that we expected as
we were starting to look at it. So, something is happening
out there. At the same time, a number of states instituted
some regulatory and some voluntary controls. Whether those
have had a corollary effect to decrease it, I don't know.
But there are other factors that have not let this issue
play according to all of the rules that we've had with all
of the other illnesses that we've looked at.
C. Dr. Watkins: As we look more and more, we may expect to
find more cases, and, at the same time, we have nothing
but death or infections as a final arbiter. And, what we
really may be seeing is a balance of what we are doing
against what we are finding in our intensified look.
C. Mr. David Dressel, FDA: Actually, Richard made the very
point I was going to raise. At the conclusion of our last
workshop, the real fear that we went home with was that the
floodgates were going to open on reported illnesses and
deaths. As Cynthia Whitman's data showed, that did not
happen. I think it is important tomorrow when we sit down
in our groups to try and look at what has happened since
that workshop and get a feel for where we were then and
where we are now so that we know where we are going to try
to go tomorrow. After that workshop, Rich mentioned we had
active surveillance. That is correct. We had five states.
We try to go out and actively beat the bushes and get
information on illnesses and deaths. We had industry
education that we should not overlook. The states here in
the Gulf area and FDA specialists went to industry and
started specifying time/temperature controls. That was what
we were looking for initially. We had media education not
only for the consumers, but the press, as you recall, had
front page articles, it seemed like, every week. We have
the consumer education through our own health brochures that
Ruth has touched on. So we can say how effective our
message is. That is what we have to look at. I can't say
that they aren't reaching people. They aren't reaching
everybody. But, as Rich said, there has got to be a factor
here that we haven't put our fingers on. Until we do, I
don't know where we are going to go and say what we have to
do tomorrow after this workshop is over. I'm curious to see
how the pieces fit together. I came in and saw a lot of
ideas that we had sketched up on boards kind of falling
apart when we looked at the scientific data to back them up.
That still isn't there. But until we know what has happened
-- I think we must know that to figure out where we are
going to go tomorrow after the workshop is over.
C. Dr. Watkins: At this time, is there anybody here who would
like to bring up issues that were not brought up today,
scientific or technical information, studies, factors, that
haven't been brought forth? I'm aware of a couple that I
ought to mention; one, in particular. There is a group
of people in Louisiana working on Vibrio vulnificus
bacteriophage in the lytic cycle. Apparently, they have a
number of isolates that are very effective. It's Ron
Luftig and another individual working with him. This kind
of bacteriophage or virus of vulnificus conceivably could
prove to be useful if we ever reach a situation where we've
got to intervene and do something with the oysters in the
processing sense. They are working on that. I thought
that was pretty interesting. I know that a lot of groups
are starting ribosomal typing. It was mentioned today.
Any other techniques or strain separations going on? Any
other issues at all that people want to bring up?
C. Dr. David Cook, FDA: My aspects on needs that remain
certainly deal with the pathogenicity or virulence of the
organism. Are we dealing with more than one strain? Are we
dealing with a number of strains? Do they differ?
Infectious dose data is certainly needed before we can set
any type of standards. Methods to detect pathogenic
strains, if there are specific ones, are greatly needed, and
methods need to be rapid.
Part II. INFORMATION NEEDS AND APPROACHES, June 16, 1994
TODAY'S FOCUS
Mr. John Marzilli
Division of Field Science
Office of Regulatory Affairs
U.S. Food and Drug Administration
Rockville, Maryland
Panel members will proceed to describe briefly their views on
remaining information and research needs, approaches and means to
address these needs, and mechanisms by which they might be
accomplished. Following the summations provided by all of the panel
members, moderators will, at their discretion, open the floor for
audience participation. Audience participants may ask questions,
answer questions, raise other needs or issues, or simply comment on
the discussion.
EPIDEMIOLOGY AND RISK FACTORS
Moderator: Dr. Scott Rippey, FDA
Panel Members: Dr. Cynthia Whitman, CDC; Dr. Gary Hlady, Florida
Department of Health; Mr. Bob Creasy, FDA; Mr.
Chuck Kaysner, FDA; Mr. Tom Herrington, FDA
Dr. Rippey:
Retool, and forge a proactive system for obtaining
epidemiological data
Dr. Whitman:
Data on the epidemiology of infection cases
Determine the truly high-risk factors
Conduct thorough tracebacks on current and previous reports
Obtain and differentiate many more clinical isolates
Improve the surveillance system
Create an ICD-9 code for vulnificus infection
Dr. Hlady:
Improve surveillance
Focus on the oysters (the food vehicle) and determine other
Vibrio infections associated with them
Information about consumption patterns
People who become infected, more about what their history of
consumption has been
Identify some more primary prevention strategies, and evaluate
effectiveness of different prevention strategies
Complete and validate the predictive index based on salinity
and temperature
Establish tolerance and action levels
Mr. Creasy:
Examine the links in the chain of consumer protection,
strengthen some
Solve the problems of incomplete surveillance reporting
Strengthen the epidemiological traceback
Get more samples and results from incriminated oysters more
often; build database on dose
Determine the importance of time-temperature abuse after
harvest
Build the database and validate the predictive tool developed
Mr. Herrington:
Include in reporting form: "Did the patient or the family
know that the patient was at risk?"
Change the level of risk, either by warnings or by standard
levels of tolerance
Mr. Kaysner:
Find ways to look at incriminated oysters more often, and see
what levels of Vibrio were being consumed
Obtain reliable consumption data
Complete work on the temperature and salinity predictive
index, and test it
Active approach to investigation
Panel and Audience Discussion:
Periodically reinforce the procedures established for
reporting
Find a mechanism to capture the cases which are occurring
outside the Gulf Coast area
Alert physicians, at least in Gulf coast, of the need for
speedy contact with patients by investigators
Warn people with hemochromatosis
More reliable means to obtain hospital specimens, incriminated
oysters
Obtain many more clinical isolates
Establish a Vibrio strain repository and fund this
sufficiently
Develop a network of the state shellfish control agencies,
focal points for information
Expand to include inland states
Distinguish sublethal septicemia cases in a separate category
Gain support for epidemiological work and traceback efforts by
redefining outbreak as a single case
Change and improve the report form to enhance timeliness
Enhance efforts to advise at-risk individuals through
physicians
Rely on contact with state, hospital labs to gain isolates,
and truer perspective on the incidences of infection
CAUSATIVE FACTORS: PATHOGENICITY, VIRULENCE,
VBNC, LETHALITY
Moderator: Mr. Chuck Kaysner, FDA
Panel Members: Dr. Chuck Kaspar, UWisc; Dr. Glenn Morris, UMd; Dr.
Jim Oliver, UNCC; Dr. Ron Siebeling, LSU; Dr. Linda
Simpson, UNCC; Dr. Mark Tamplin, UFla; Dr. Anita
Wright, UMd
Mr. Kaysner:
Use seasonal, environmental, and market data as well as case
information to begin closing in on hazardous doses
Determine seasonal trends in consumption
Develop fuller database on capsular types
Develop methods which enable dose data to be focused on
encapsulated cell doses
Develop databases derived from ribotyping or the restriction
fragment length polymorphism (RFLP) of various strains to
enable clearer distinction among and between clinical and
environmental strains
Dr. Morris:
Establish recommended, standard methodology based on gene
probe and immunoassays
Obtain many more clinical isolates, overall and from the same
patient
Expand efforts to distinguish strain differences using
molecular epidemiologic techniques, particularly by
carbotyping, essentially identifying the capsular
polysaccharide types, ribotypes, RFLP, to identify certain
groups that have an increased potential for human disease
Establish a centralized sample repository for clinical and
environmental isolates
Support basic research on pathogenesis to understand what
causes the intractable sepsis syndrome
Funding
Dr. Siebeling:
Include immunotyping of capsular types to distinguish strain
differences and identify those environmental isolates which
are truly virulent
Dr. Oliver:
Support efforts to determine the role and significance of the
VBNC state in this problem
Expand work on strain differentiation by capsular and LPS type
Determine what types of strains are present in nature,
oysters, using direct, specific techniques like PCR,
antibodies
Explore other animal models to distinguish human pathogenic
strains
Exploit PCR technology to reliably differentiate between
culturable and nonculturable, opaque and translucent
Establish a strain repository
Dr. Simpson:
Investigate the role of ferric uptake regulator genes or the
control of attachment factors or virulence factors by iron.
Iron in oysters and victims suggest this may be a key.
Support collaboration among labs specialized in particular
areas to enhance progress, obviate duplicity
Dr. Tamplin:
Refine the definitive critical factors causing human high risk
into subsets
Obtain new strains and promote investigation networking to
achieve this
Establish a strain repository
Establish Gulf Coast tagging which allows for facile traceback
of oysters
Investigate available techniques to "fingerprint" oysters for
traceback and correlations to "hot spots"
Develop solid information on other foodborne illnesses, deaths
in order to place the Vibrio problem in proper perspective
Start a pilot program to employ on-board icing as an interim
measure when estuaries are identified as sources of lethal
oysters
Consider other indices of risk, such as total vibrios
Dr. Wright:
Determine whether this is classic endotoxic shock, a response
to LPS, or other components, using perhaps other animal
models, such as the limulus assay
Elucidate the molecular basis of what component is
contributing to this disease
Be wary of the phase variation and genetic instability in lab
work, minimizing passages and maintaining frozen stocks of
strains
Panel and Audience Discussion:
Find a responsive animal model for LPS effects
Develop a common definition for "viable but nonculturable"
Conduct a collaborative study, and disseminate wider use of
the probe technique for examining oyster meat
Conduct detailed field studies of "hot spots" to determine
factors involved and those as yet unrecognized, and establish
sentinel monitoring to relate future mortality to harvested
dose
CAUSATIVE FACTORS AND CONTROLS:
TIME-TEMPERATURE, HARVESTING, HANDLING, PROCESSING
Moderator: Dr. Jim Oliver, UNCC
Panel Members: Dr. David Cook, FDA; Ms. Angela Ruple, NMFS; Mr.
Chuck Kaysner, FDA; Dr. Marilyn Kilgen, NSU; Mr.
Larry Edwards, FDA; and Mr. Tom Herrington, FDA
This session will be conducted a little differently. It will
cover topics, beginning with harvesting practices in individual
states, and state regulations about harvesting. Following this,
some needs will be cited dealing with temperature and time factors.
A discussion will follow on all topics, including treatments that
might be used other than depuration, and on depuration and
relaying.
Harvesting
Mr. Richard Thompson, Texas Department of Health:
To understand the Texas industry in the summertime, I need to
tell you a little bit about the Texas industry in the wintertime.
Our normal open oyster season is from November 1 to November 30.
During that time, we have from 50 to 100 certified dealers.
Probably, 25% of those are shucker-packers; the rest are shellstock
shippers, and, of course, a lot of shucker-packers do a lot of
shellstock shipping out of Texas. We have anywhere from 500 to 700
harvest boats working during that open season. During the
summertime, the harvest season is closed to public harvest; the
only activity going on is private leasing where transplant
activities occur. In some cases, transplanted oysters are allowed
to cleanse and are then harvested. In other cases, reefs have
actually been built on their private leases over the years, and now
natural stock or stock from transplant activities has been there
for a year or more.
In considering the volume of activity, we have found over the
last several years that six, seven, or eight dealers operate
during the summer months and pull anywhere from 75 to 100 boats.
This group will move as many oysters in the summertime as the
entire public fleet will move during the open season. Our volume
or production stays fairly stable year round on a month-to-month
basis.
The summer activity is restricted to Galveston Bay. It is the
only area where we have private leases, so it is a relatively small
concentrated effort. When we began looking at this in 1988, we
made some recommendations for changes to their practices; the
industry voluntarily adopted those in 1988, and we made the
regulations effective in 1989.
It takes an hour or an hour and a half for a boat to run from
the dock to the harvest area. They harvest in the morning, keep
the decks covered with tarp shields over the boat, bring the
product in generally around noon, go back out in the afternoon,
make a second load, and come back that evening.
In addition, we investigated a requirement that shellfish
harvested be placed under mechanical refrigeration at 45xF or less
within two hours of unloading from the boat. What we generally
see, particularly in the summer, since the processors are located
on the water, is that the boat will come in and unload directly
into a cooler so that the product is actually refrigerated within
10 to 15 minutes after unloading.
The two-hour requirement came up primarily during the
wintertime when we found 30, 40, or 50 boats lined up at a dock
trying to unload. Sometimes it takes a bit longer than 20 or 30
minutes to unload one boat. We had boats sitting there sometimes
all night, so we instituted the requirement.
Since we don't have any real shallow water, all oystering is
done by dredge; we don't have any tonging. The boats are probably
50% split between the Louisiana lugger flat bottom type designed
entirely for oystering and converted shrimp boats. The big chunk
of the industry that is not oystering in the summertime goes
shrimping. There are a few of those who remain with the oyster
activities off and on during the summer depending on the shrimping.
You have a combination of boats designed for oystering and boats
designed for shrimping, all of which are using a standard dredge.
In the summer, the dealers will usually put their shellstock in
their cooler at least overnight, then load out of that cooler to
fill a truck up after they get an order.
The industry has decreased significantly in the last several
years to the point that they no longer have enough business to load
out a truck every day. Consequently, many of the oysters get
refrigerated before they go out on a truck. In the past, when the
boats would come in, they would load directly onto a refrigerated
truck, and head straight out of there. That doesn't happen very
much anymore, mostly because of the demand for the product.
You didn't have much up here in the way of illness for Texas;
we are not in your region. I am going to give you something by
memory; don't hold me to this. We started out in 1988-1989, with
about 11 or 12 illnesses and about nine deaths. We then saw a
significant drop. I don't pretend to explain why. In 1990, 1991,
and 1992, we dropped down to five to seven illnesses, with two or
three deaths in the last year. I don't have the most current data,
but it was probably in that same range.
Our epidemiologists are pretty good about notifying us of any
changes, and they have not notified us of anything. So, without
having the actual data, I would say we are looking at a drop in the
number of illnesses. We like to think that it is because of some
of the refrigeration practices and the fact that we get it off the
water quickly and get the oysters refrigerated and keep them
refrigerated.
Our other requirement that might have a bearing on this is the
boxing or repacking act for shellstock. Obviously, for shucking, we
have a lot of rules. But for repacking, since the product goes
into the cooler, we require that once the product has been chilled,
it must remain under refrigeration. It cannot be out of
refrigeration in excess of one hour. If they choose to take sacks
and cull and repack them for the half-shell trade (which a lot of
our dealers do to select the prime half-shell oysters for those
boxes for which they get a higher price) they do that under
refrigeration. They do not take it out and do it on the dock or
out in some unrefrigerated location.
Those are the things that we have done directly or voluntarily
within the state to respond to summertime operation. We're not in
a position to tell you, "Yes, that is what made our numbers drop,"
but we have seen a drop in numbers. I think the drop in business
is considerably greater. In 1988, 1989, and 1990, our industry was
still functioning quite well. In the last two years or so,
industry demand has dropped off dramatically, but the drop in the
number of cases came before the drop in the industry level of
activity.
Reasons for decline: Bad publicity, lack of demand for the
product. A lot of big businesses have stopped selling oysters, and
we have lost a lot of major accounts from that. But overall it is
just a lack of demand. We've got oysters. The price is lower than
it has ever been; it's as low as it has been since I've been in the
business, 19 years, and we can't move the product. It is sitting
in the cooler.
It is a public education effort that is needed. When I say
"public education," that covers a multitude of factors. Some of
you heard me say for the last couple of days that I think we ought
to start having a required food safety course in school and teach
every person who goes to school in this country food safety issues,
and vulnificus would be one of those; consumption of raw shellfish
would be one of those. That would be my response, but it is a long
term response. The short term issue is, should it be controlled?
We have a limit of 150 sacks a day. The industry produces far
below that. It generally produces somewhere between 30 and 35
sacks per boat per day. In the summertime, it may go slightly
above that. That's as a result of the demand, not of any
restrictions. In Texas, they are individually sacked.
Mr. Mark Chatry, Louisiana Oyster Dealers and Growers:
The first thing I can say about the Louisiana industry is that
it produces about 11 to 13 million pounds a year. Jim was talking
about the variables that impact illnesses. That, of course, is
another variable that has to be considered when you talk about
illnesses; that is, the amount of supply. The Gulf, in general,
typically produces maybe 50% of the oysters that are consumed
nationally. While I don't have any specific data, I know that a
large percentage of those oysters are consumed raw. Those are very
important variables that have to be considered along with the
incidence of illness, not to mention the less-than-unbiased method
that has been used for determining illnesses on a nationwide basis,
which we have learned over the past two days.
Temperature controls that Louisiana instituted in either 1988
or 1989 before ISSC implemented more stringent temperature
controls, were that from April 1 to April 30, there can be no
oysters on the deck after midnight. What that essentially means is
same-day harvesting. In essence, since we can't fish at night,
harvest is restricted between daylight (typically, 6:00 a.m.) to
darkness. So, oysters are at the dock, generally speaking, within
12 hours. The oldest oyster on the boat will be no more than 12
hours old. There is a regulation in Louisiana that within three
hours after those oysters reach the dock, they have to be placed
under mechanical refrigeration at 45 degrees and maintained at 45
degrees in all levels of commerce. The only exception is when the
oysters are being transported to a processing plant that is within
one hour of the dock. Those are our temperature regulations.
The state of Louisiana also implemented other regulations.
There must be an awning on the boat to protect it from the sun year
round. I am sure Louisiana intends to go with ISSC's recent change
so that during the winter months, the product cannot be on the boat
more than 36 hours.
The method of harvest is largely, almost exclusively, by large
oyster boats. The product has to be at the dock by midnight
unless, of course, the fisherman has mechanical refrigeration on
board, which, perhaps, 20% of the industry has at this point. They
can keep the product on the boat provided there is mechanical
refrigeration on board. The industry is based on a public resource
and private fishermen. The public resource consists of certain
areas of the state that are designed for public use. All one needs
to have to harvest oysters from those areas is a boat and a
license. The leases that the state provides to leaseholders are
the predominant source of the product in Louisiana, perhaps, 80%.
Those areas are open year round, whereas the public areas are only
open September 1 through the beginning of April. Generally
speaking, the oysters are harvested by large oyster luggers and two
dredges, and they fish from 50 to 300 sacks per day. How many
bushels (or units) can be loaded on one boat in one day? About 300
sacks, each sack about a bushel and a half. These are harvested
off the bottom by dredge. In Louisiana, they are individually
sacked on the boat as they go.
Mr. David Heil, Department of Environmental Protection:
Florida Marine Fisheries Commission is a separate independent
agency that sets seasons, value limits, and size limits in the
State of Florida. The Florida Department of Environmental
Protection classifies the waters and does the processing plant
inspections for the health controls. Essentially, the Marine
Fisheries Commission has set the seasons for oyster harvesting, in
general, from October 1 through June 30.
Our major producing area is Apalachicola Bay. It produces at
least 90%, probably 95%, of the oysters coming from Florida. That
area has a summer season, open from July 1 to September 30. During
the peak winter harvesting season, we have probably 350 to 400
oyster boats working. Requirements for bag limits vary from
fifteen to twenty 60-lb bags per person or boat per day, whichever
is less. Only one trip is allowed per person. Our regulations
require that oysters are delivered to a certified dealer the same
day of harvest. Again, in general, the Apalachicola Bay area is
the major producing area. The bars are located some two to five
miles from shore. All harvest is done by hand tong only on the
public reefs. Mechanical harvesting is permitted by court action
on leased parcels.
In the Apalachicola Bay area, there are nine leases totaling
600 acres. Fewer than 3,000 bushels or bags of oysters are
produced per year off those. The tonging boats are very small, in
the 30-foot range, all powered by outboards. On the harvest day,
you go out as early as you can to beat the summer heat (or even the
winter heat) and winter winds and come back as quickly as you can.
With the small bag limit with the available resource that we have
in Apalachicola Bay, it doesn't take long to harvest product; so
they are out there an average of two, three, four, sometimes five,
hours at a maximum. We do have enforcement problems with people
sometimes going out twice a day to get extra bag limits.
Eighty-five to 95% of the product harvested from Apalachicola
Bay enters the half-shell trade, and, presumably, most of this is
consumed raw. Most of our major shucking facilities are in the
Apalachicola area as well as right on the water. We also receive
product from Louisiana and Texas to shuck and put on the market as
gallons or pints.
About some general temperature ranges: Over a five-year
period, we saw from 9 to 31xC, and ambient air temperatures from 10
to 36xC. As far as Vibrio vulnificus relating to a public health
problem, Florida believes it is. That is why I am attending today
as an official representative of the State of Florida. From an
epidemiological standpoint, it appears that the illnesses are
indeed regional. It also appears that the people getting ill are
fairly well defined individuals in a high-risk category. As such,
education is an appropriate way to deal with this problem.
Florida has taken at least one step forward and has required
the consumer information message to be on all product that is
produced in Florida or all out-of-state product that comes through
one of the certified processor plants in Florida. We are, of
course, in the business of regulation. Therefore, if a standard is
reasonable and adopted by the National Shellfish Sanitation
Program, Florida would certainly request the funds to implement the
program and we would certainly move forward very aggressively on
enforcement.
Dr. David Cook, FDA:
Mississippi and Alabama both have very short coastlines. They
have somewhat different strategies in dealing with their shellfish.
Mississippi has both the tonging industry and a dredge boat
industry. The dredge boats are usually limited to about a 30-sack
limit and the tonging to ten or less, depending on the availability
of the resource. Mississippi also has a closed season from,
roughly, the 1st of April until the 1st of October. This can vary
a little bit. The resource primarily goes into the shucked oyster
trade.
In Alabama, it is strictly a tonging resource. The state can
choose to leave it open year round. Currently, they are harvesting
at this time of year from Mobile Bay. However, they have strict
time limits. During the winter season, boats have to be in by 2
o'clock. They cannot be out overnight. During the summer season,
ongoing now, harvesting can start at light in the morning and boats
have to be in by noon.
Needs
Conduct definitive experiments to determine the effects of
high storage temperatures on densities of V. vulnificus in
oysters throughout the year.
Develop capability to inform, educate inland states, and
obtain epidemiological data from them.
Specifically determine factors and interrelations involved
between naturally occurring V. vulnificus and oysters.
Conduct studies tracing lots of commercially harvested and
marketed oysters to determine effects of varying temperature
regimens experienced by oysters on the densities of V.
vulnificus.
DISCUSSION
Questions, Answers, and Comments
C. Dr. Mary Wang, California Department of Health Services: I do
want to share with you some of our data. California issued a
raw oyster warning regulation in March of 1991. Toward the
end of March, we had one case; a person died. In 1992, two
cases. In 1993, we had a total of 6 cases. Chuck Kaysner
presented the data we published in the California Morbidity
Report.
We traced all these oysters. All the people that died had
previously said they had consumed oysters within that week.
We traced it all to Louisiana oysters. Last year, six cases,
four deaths. Five cases occurred in June and July. One case
occurred in September. We experienced the largest number in
1993. Even before the promulgation of our regulations, we had
only one or two cases per year. So we don't know what went
wrong last year.
Chuck has followed up with the virulence studies. I guess you
have obtained two or three strains. You can share that if you
like.
C. Mr. Wittman: Just some information on the Virginia
experience. The majority of oysters shucked in the State of
Virginia come from outside Virginia. Our commercial harvest
last year, I believe, was about 6,000 bushels because oyster
season was closed for a period of time. The majority of those
oysters were shucked. We have not had the cases of Vibrio
vulnificus infections that you would expect in Virginia from
those or based on the number of oysters that we shucked from
the Gulf.
Certainly, that can be explained by a number of different
reasons. First, the idea of reported diseases comes into
play. Second, some oysters are shipped out of state. We're
not sure, because some of the information may not get reported
back to us, how many cases of Vibrio vulnificus infections are
caused by those oysters. But it does provide an interesting
piece of information concerning the disease itself; that is,
it makes you question what is happening concerning Vibrio
vulnificus infections, especially with the number of Gulf
Coast oysters that are processed in Virginia.
Q. Dr. Kilgen: Bob, we did a project last summer with one of
your commercial processors, looking at the effect of
commercial blowing and freezing. They told me that it was, in
fact, a state law in Virginia that you require this blowing
process. Is that correct?
A. Mr. Wittman: It is not required, but the majority of
processors do blow the oysters.
C. Dr. Kilgen: The process, it was told me, was that they
tumbled the oysters in 65 degree water which was continually
running until the oysters were completely clear, they iced
them down to 45 degrees, and then they packed them. We also
did this using completely commercial conditions. And this,
again, was not in accordance with the data that Dr. Cook had.
This was done under real life commercial conditions where
oysters harvested out of Louisiana went through regular
commercial routes, which, as you know, can take quite a few
days to get where they are going. But, also, we saw a drop in
the levels just with the blowing process of over 90% in
vibrios. And, then, once they were commercially frozen, it
was almost to undetectable levels.
It is certainly no secret that vibrios have tremendous osmotic
sensitivity to freshwater. I have brought this question up
before to the National Advisory Committee regarding the HACCP
guidelines where the contact with ice water is going to be
greatly limited.
That concerns me because you are weighing a safety issue
against an "economic fraud" issue. I would much rather see
oysters soaked in water. Of course, Cajuns don't like that
either. It gets rid of the salt, but it can certainly kill a
lot of vibrios if they soak in ice water long enough. And if
it requires labeling to say that they have been washed in
water to control vibrios and the public knows what they are
getting, to me, it seems that would certainly -- I think that
may be part of why you don't see vibrio infections from the
shucked product.
C. Mr. Wittman: I think that could be part of the explanation.
I think it is probably a combination of a number of factors.
Of course, another question is, what are the consumption
patterns for oysters that are shucked? Are some of those
consumed raw? Are some of those cooked? That question also
comes into play.
But we just feel it is interesting that a large majority of
our oysters do come from the Gulf. The diseases reported from
Virginia oysters as far as Vibrio vulnificus infections, are
much fewer than what we would expect.
Q. Mr. Tom Herrington, FDA: What do you think the reason is?
A. Mr. Wittman: I think it is a combination of factors, I think
consumption patterns. Probably after the oyster is shucked,
with a cooling process, you do get some reduction in
organisms. Possibly, as Dr. Kilgen said, the blowing process
may lower the number of organisms in the oyster.
Possibly, shortfalls in the reporting system, coming back to
Virginia. If the majority of the product that goes out of
Virginia is shucked, it is hard to trace back a shucked
product to where that oyster actually came from. So,
potentially, if the disease is being caused and it is being
caused out of state, maybe the information is not making it
back in order to trace it to a Virginia oyster.
But, again, a number of different possibilities exist to
explain the differences for the lower-than-expected number of
Vibrio vulnificus infections.
Q. Ms. Angela Ruple, Gulf Coast Research Laboratory: Did you get
a count on the oysters before blowing as well as after
blowing?
A. Dr. Kilgen: No, because they weren't able to ship me a sample
back. That is the one thing I need to do. I would expect it
might have dropped a log or two, but not as much as it did,
not from 10,000-24,000 to, say, 100. That's a big drop.
Q. Ms. Ruple: How many days between the time of harvest and --
A. Dr. Kilgen: It took about four or five days to get up there.
C. Dr. Oliver: These are the last two questions here.
C. Mr. Moore: I just wanted to add a couple of comments to the
discussion about transportation and the effects it has on the
levels of vulnificus and other illnesses as well, having
managed the shellfish program for the State of South Carolina
for about ten years up until about a year ago.
In South Carolina, the tourist industry now equals the
agriculture industry in the state. The tourist industry is
focused, probably 90%, on coastal tourism.
South Carolina is a shellfish-importing state. There is a
tremendous raw bar market during the summer. There is not any
harvest of oysters in the summer in South Carolina. Ninety-
nine percent of the shellfish that are consumed in South
Carolina are consumed during the summer; this occurs on the
coast. Seafood is a predominant food.
Over the six-year period from 1987 until 1993, we actually had
five reported cases of vulnificus. Let me change that. There
were six reported cases and five deaths. One death from the
consumption of shellfish, three deaths from wound infection.
So, obviously, something is occurring because 99% of these
shellfish came from the Gulf Coast, and they came during the
summer when they were harvested.
So, again, I find that very interesting. When you ask the
question, why, I think that we are as baffled as this group
is. I don't think we know enough to draw much of a
conclusion.
C. Mr. Thompson: It would be nice if we were dealing with a
problem with one variable so that we could look at that one
variable and find out what is happening. I would point out
for the group's information, in the beginning when we had this
problem and started trying to deal with it, all of the cases
that we found involved interstate transport of oysters. We
were talking about oysters that were 6, 8, 10, 12 days old.
What I am hearing today is that the picture has changed. We
are seeing cases now that are in the state of harvest or close
to the state of harvest. We are not seeing them in other
states, and we are jumping to the conclusion it is because it
is not being reported in other states.
I would point out for your information that coincidental to
all of the fussing that we have being doing about Vibrio
vulnificus in the last seven or eight years, the ISSC has been
doing a lot of fussing about other issues. One of those was
transportation of shellstock. We have instituted requirements
for refrigerated transport of shellstock, for sealed doors,
for temperature recorders on the trucks.
Coincidental with all of this looking at vulnificus over here
and trying to figure out why the numbers are going down in the
other states and saying we are not getting it reported, we are
requiring the product to be transported in refrigerated trucks
where it may not have been in the past.
So we've got another variable added into the equation. We are
shooting at multiple variables here. I don't know if that is
the reason we are not getting them from the other states, but
I think it is something we can't afford to overlook.
Perhaps handling practices not associated with vulnificus, but
associated with quality or safety or other issues, have
resulted in a lowering of vulnificus simply as a result of
lowering all bacterial counts, and, therefore we are not
getting the interstate transportation cases that we used to.
It is a possibility. It is another variable that we have to
consider.
Generally speaking, what I have seen and heard as the Chairman
talking to other states, the refrigeration practices are much
better now than they used to be. We are getting to the point
where we've gotten the truckers' attention and in many cases
we no longer have to put the recorders in. In fact, I believe
there is an issue submitted this year to eliminate the
requirement for a recorder. Where you have trustworthy
transporters, you can take $25 to $50 off a load because you
don't have to have that recorder there. So, if the transport
industry is behaving better than it used to, that may be
another factor to plug in here, making our decision about what
to do all the more difficult.
C. Dr. Whitman: I guess I have some concerns about the idea that
there may be fewer cases inland. I simply think we don't have
enough information to know what the case situation is in the
inland states. We hear about the cases in the Gulf Coast
because that is where they are reportable and that is where we
are looking.
Even when we are looking, we are finding, say, less than 50
cases a year, in fact, on average, fewer than 30 cases a year.
So when you are looking and you are only finding 30 cases a
year, I think to expect to hear of more than a few isolated
reports from areas that are not looking is probably
unrealistic.
I think a couple of things have to happen before a case gets
reported. First, the physician has to recognize it. In areas
away from the seacoast area, it may be that medical personnel
don't recognize it as much, don't request specimens.
Second, if it is not reportable even if it is found, the state
may not know about it. And then it certainly doesn't get to
CDC, except that sometimes specimens are sent or sometimes we
are called.
But my point is that this is a rare illness, and even where
you are looking, you find a steady number of cases over the
past five years. But it is rare enough that to try to make
estimates of what is happening in places that aren't
reporting, I think, is -- I don't really think we have the
information for that at this point.
Q. Mr. Herrington: Do you consider this a public health problem?
A. Dr. Whitman: Absolutely. It is an illness that is
preventable and highly fatal. So, yes, I think it is.
Q. Mr. Herrington: And you do think, then, that some other
measures of control should be placed on the product?
A. Dr. Whitman: I think that it is very important to take
action, again, by control. I would say that certainly we
should target prevention. We should target intervention
activities to prevent illness.
C. Dr. Oliver: We have 15 minutes left to cover one more major
topic; that is depuration, relaying, and other methods that
might be used to try to eliminate or to lower the population
of vulnificus in oysters.
Mark Tamplin spoke yesterday about depuration. Just a couple
of things that I jotted down that I thought were unusual or
worthy of comment. He detected over 100,000 Vibrio vulnificus
cells being released per hour per oyster in one study.
Incredible levels of vulnificus, apparently, growing in these
oysters. Further, oysters that were harvested with a very low
vulnificus load, on the order of 10/g, reached very high
levels when depuration was carried out at room temperature.
The problem with depuration is that studies from this lab,
from Gary Hlady's lab, from Dr. Rodrick's lab, and numerous
others, have shown that depuration clearly does not eliminate
any vibrios. So one question is, where should we go from here
with depuration studies? One idea, of course, is to combine
that with something else, add something in the water, for
example, that might help depuration.
While certainly not commercially feasible, if you added five
very powerful antibiotics to the depuration water, some of
that would go into the oysters and kill the vulnificus, and
numbers would drop about one log.
We heard yesterday that freezing certainly lowers vulnificus
levels a lot. It doesn't eliminate them. Moderate heating
was not effective in shellstock oysters. Radiation is a real
possibility, but we have a lack of data at this point, and,
certainly, there is a lot of consumer concern.
I want to show a couple of real quick slides. We've been
looking at some GRAS compounds, which are compounds generally
recognized as safe by FDA.
[Slide]
I want to point out that levels of both the translucent and
the opaque in the presence of low amounts, in parts per
million, would dramatically decrease in artificial seawater.
[Slide]
In oysters, as you might expect, it is not nearly as
effective. Here is an effect of diacetyl, which is the best
one we found, against vulnificus. We pumped through the
diacetyl for a couple of hours and then took the oysters out
and put them in 5 degrees to mimic transportation, so they
would accumulate the diacetyl and then be transported. There
was about a 3 or 4 log drop, with the levels you would have to
use, being realistic, of about 0.05%. So I am not sure if
that can be of use or not.
[Slide]
I would like to show one last thing just for fun. Some of
you, I am sure, are aware that there was a report from the
Louisiana Medical School that you could put tabasco sauce on
oysters and it would cure the whole problem; the vulnificus
would go away. Having done this sort of stuff for so long and
with no success, we thought it was not possible. So we
decided to look at that.
[Slide]
The control oysters had about 10,000 vulnificus cells per
million. When we added the horseradish-tomato cocktail sauce,
it didn't do anything at all. When we added tabasco sauce, it
wiped it out entirely. But this was on the surface. You just
put tabasco sauce onto an oyster on the half shell for ten
minutes, and it wipes out the vulnificus.
But when you look inside the oyster, there are about ten-fold
more in the control oysters. The cocktail sauce didn't touch
a bit, and with the tabasco sauce, there was not a significant
reduction at all. So, tabasco sauce, unfortunately, is not
the answer, despite what you may have heard.
I would like to open up discussion on depuration, treatment,
or any other ideas anybody has.
C. Dr. Mark Tamplin, University of Florida: I just wanted to say
in defense of the people at LSU, I think they qualified what
they were doing in a test tube and said it might have some
application. But this is an example of what happens when you
talk about a promising technology and the newspapers and CNN
take it forward. Eventually, it came back, and a lot of us
thought they were saying that. Actually, I think the
scientists were being fairly objective, but, as usual, the
media weren't.
C. Dr. Oliver: One problem was that the media picked it up and
said things like, "Possible Cure for Deadly Bacteria," and
made it sound really good. The other thing is a precaution we
should all take home; that is, you don't talk to the media
before a thing is published. This thing was never really
published, but CNN didn't come knocking on their door.
Someone had to go to their people and say, "We have an
interesting thing here," and it got picked up.
I agree that the media may have also been a problem, but we
brought on some of that problem.
Q. Mr. Alpha Diallo, Norfolk Public Health Department: I was
just wondering, has anybody studied the biological control of
Vibrio in oysters? I am thinking of including some
bacteriophage before the oysters get to the market or to the
restaurant. What would be the significance?
A. Dr. DePaola: We and some investigators in New Orleans are
also looking at this. Over the past several months, we have
been able to isolate some phages that will inactivate some
strains of Vibrio vulnificus but, apparently, not other
strains. It may be possible eventually to get a cocktail of
phages that could reduce the levels of Vibrio vulnificus in
oysters, but I don't think it will ever be possible to knock
them down to zero. That's just my personal opinion.
C. Dr. Oliver: In response to that, Gram-negative phages tend to
attach to the LPS. If there are dozens of LPS types, you will
find a real problem using that, too, just because of the kind
of thing we heard yesterday in the LP response.
C. Dr. DePaola: But they are also attached to the capsules in
some cases, and that may offer a way of separating a
particular capsule or it may be a simple way to identify
certain capsular types.
So, the fact that they don't attack all different strains may
be useful in separating virulent from avirulent strains and
might be blended in nicely with some of the ribotyping and
other studies that are being done as a fairly simple assay to
conduct.
We only have a handful of phages at this point, but we plan to
continue to collect phages as the summer goes on and would be
interested in exploring some possibilities to determine their
effectiveness, perhaps as indicators, also.
Q. Dr. Oliver: Any other comments or input from anybody about
depuration or possible treatment methods?
[No response]
If not, let's go to Steve.
C. Dr. Jones: I am going to be on the next panel, so I will go
into this a little more in detail. In this context, I think
we shouldn't discount the potential of relaying. It's not
really clear to me whether vulnificus is absolutely ubiquitous
or if there are cold spots in areas in the Gulf of Mexico
where there may be potential for relaying.
I know that in the Great Bay Estuary of New Hampshire, which
doesn't have any reported incidence of disease, we find these
cold spots. We can relay. We can reduce the levels of
vulnificus in the shellfish perfectly naturally in seven days,
and I think it is a potential strategy to remove these
organisms from the shellfish that we should still consider.
C. Dr. Oliver: If we could, let's start with Tom and each take
a minute to summarize any comments.
C. Mr. Herrington: I agree with Steve. I think there are
some potential things that we haven't looked at. One of them
is relaying offshore, higher salinity areas, perhaps. There
is a possibility of that.
I know we have some illnesses and deaths reported from
commercially shucked product, but it might be anecdotal. A
lot of information is missing there.
If you want to see some other controls, Dave Cook, I think,
was talking about some heat shock within temperatures of 145
to 150 degrees for about three minutes and then perhaps
blowing in really cold water and then packing it up. You may
have some different figures. You may see a decrease.
Shellstock is another story.
C. Dr. Kilgen: I did have one overall comment. I like to get to
the bottom line, and I talked with Dr. Whitman from CDC about
this. The thing that struck me the most out of all of the
really outstanding papers that we heard -- and it is the
bottom line - looking at the percentages of illnesses for
Vibrio vulnificus in high-risk individuals, 85% are from
oysters eaten in restaurants, 11% from retail stores, and
only 4% from wholesale industry.
It strikes me that the job that the ISSC has done, the FDA,
the states working together with the industry, have
implemented a lot of controls and a lot of things in terms of
time, temperature, and handling. FMI (Food Marketing
Institute), which represents the retail stores, also
participates in ISSC. I could be mistaken, but I've never
seen anyone there from the NRA (National Restuarant
Association). Certainly looking at this bottom line data, if
85% of illnesses are in high-risk people, again, I agree with
Rich Thompson, it is education, education, and more education.
Let's bring the NRA into ISSC, wherever they should be, so
that they can participate in this program, too. I don't know
if they would be happy to hear me say that, but that strikes
me as the most outstanding thing that I saw.
C. Ms. Ruple: As Tom was saying, the mild heat treatment doesn't
have a benefit when it comes to shellstock oysters. But in
commercially shucked oysters, it can be of great benefit and
it can also be incorporated into some of the normal processing
techniques.
We've worked with engineers at the University of Southern
Mississippi to develop equipment using existing oyster
processing equipment, for example, using a blower and
modifying it to do a controlled time-temperature treatment in
the blowing process. That may be of some benefit as far as
commercial oysters go.
Another comment concerning time-temperature relationships is
that we need to keep in mind the effect of salinity. Mark
Tamplin pointed out the significance of looking at different
salinities. About the discrepancies in some of the data, we
need to keep in mind that there is more than just temperature
involved in whether or not the organism is present and what
strains are present. When we are looking at time and
temperature, we shouldn't eliminate salinity.
C. Dr. Cook: I think industry has a large responsibility here in
trying to lower their liability. Whenever someone dies of a
vulnificus infection, a lawsuit usually follows and that goes
back to the industry personnel. We know that many industry
practices can help reduce this liability in terms of cooling
product soon after its harvest and maintaining it at lower
temperatures. Some of our studies have shown that even levels
in the shucked product continue to reduce as it is stored on
ice.
So, with industry carrying through with the prescribed
procedures of ISSC and those listed in the National Shellfish
Sanitation Program, they can certainly reduce their
liabilities.
C. Dr. Oliver: I'd like to make a final generic comment if I
may, taking advantage of the chair here. A number of old
problems need to be resolved, like the temperature abuse
question, which is extremely important. I can't believe there
are not any researchers in this room who haven't developed
some ideas that they would love to pursue back in their
laboratories. But there is a real funding problem. To give
you an idea, the S-K program, which funded all our work last
year, is a yearly program, and that is over now. We can't go
to them now. Our Sea Grant Program in North Carolina is on a
two-year cycle, and that is over, so we can't go to them for
two years. The FDA, as far as I know, doesn't fund. So I
can't run back and do a proposal. There is no one to go to.
We can't do research without some funding. So I hope there
will be some consideration of that problem by those of you who
have some ability to do that.
CAUSATIVE FACTORS:
ECOLOGICAL AND ENVIRONMENTAL INFLUENCES
Moderator: Dr. Chuck Kaspar, UWisc
Panel Members: Dr. George Hoskin, FDA; Dr. Mark Tamplin, UFla; Dr.
Steve Jones, UNH; Dr. Jim Oliver, UNCC; Mr. Chuck
Kaysner, FDA, Dr. Linda Simpson, UNCC; Dr. Glenn
Morris, UMd; Dr. Anita Wright, UMd
Dr. Hoskin:
Determine what constitutes a hazard, with regard to HACCP
Determine what controls can be implemented
Determine the critical control points and critical limits
Learn more about the vulnerable population so that we know
exactly who the target population is
Dr. Tamplin:
Validate and employ predictive indices to enable safe
harvesting
Determine specific factors of pathogenicity and virulence
Collaborate studies with local authorities, industry, labs
Dr. Jones:
Determine areas low in V. vulnificus, determine the causative
factors, and use this approach to gain control
Find out what causes virulence among strains
Improve notification and traceback capabilities
Mr. Kaysner:
Examine thoroughly and define the conditions occurring in the
largest productivity areas, especially those implicated
previously
Do likewise for identifiable "hot spots"
Dr. Oliver:
Thoroughly investigate LPS types, both clinical and
environmental strains
Thoroughly investigate encapsulation more definitively
Investigate possibilities of gene rearrangement that occur in
this species
Learn more about the actual ecology of V. vulnificus
Dr. Simpson:
Consider examining the osmoregulation in oysters in
conjunction with the salinity sensitivity of V. vulnificus
Further determine the mechanisms of attachment of V.
vulnificus in oysters
Dr. Morris:
Develop the data needed to effectively apply HACCP to this
hazard
Determine what V. vulnificus is doing in harvest waters, and
its ecology in the water column
Dr. Wright:
Get the gene probe direct method in widespread use
CONCLUDING REMARKS
Summation of the 1994 Vibrio vulnificus Workshop
Dr. William D. Watkins
U.S. Public Health Service
Northeast Seafood Laboratory
U.S. Food and Drug Administration
Building S-26, School Street, Davisville
North Kingstown, Rhode Island 02852
Clearly, we've covered a lot of ground, far more than I can
possibly retrace here. It is worth reviewing some of the take-home
messages which came across during today's proceedings, as well as
a few from the previous day. Forgive me if I oversimplify.
The epidemiology group identified the need for more
comprehensive epidemiological data, in terms of other regions in
the country and other Vibrio diseases. This would be very helpful,
but establishing the resources needed and overcoming the logistical
difficulties may prove costly. Perhaps, if this cannot be achieved
on a grand scale, these kinds of recommendations can be implemented
on a lesser scale or, in some instances, pilot studies.
The need was echoed for greatly improving our abilities to
conduct better, more effective traceback investigations when cases
of illness or mortality occur. Very often, we have a difficult
time tracing shellfish back to the state, let alone the growing
area or the precise harvest site. It is difficult to learn of your
hazards when you are not sure where they are coming from.
We heard that we should probably be thinking more in terms of
all of the naturally occurring pathogens, vibrios or others that
fit that category, in addition to vulnificus. It seems logical
that effective controls for some may also be effective against many
of the rest. If the V. vulnificus problem were handled more
effectively, we would probably deal a large blow to some other
naturally occurring organisms as well, since they are basically
mesophilic heterotrophs and exist in much the same way.
Today, the call was heard for obtaining perhaps less technical
information and devising better prevention strategies. I think
that to get to well-reasoned, scientifically sound prevention
strategies, this kind of technical information is crucial.
The second work group made a unanimous plea for a Vibrio
strain repository to service the needs of the research community
for all aspects of research. Most of us can identify with this
problem. The need here is for annual funding without fear of
evaporation at some point.
The critical need for more clinical isolates was voiced, not
only from all the patients, but numerous strains from the same
victim. In addition to clinical isolates, clinical specimens of
blood or serum are needed so that we can understand the pathogenic
process a little better.
In looking at strain virulence and occurrence in the
environment and trying to establish a strain repository, we find
that funding is a problem, as was mentioned a number of times here.
Some of these studies and investigations are expensive operations,
requiring collaboration and several years to conduct. Relatively
speaking, only a handful of investigators have been involved with
V. vulnificus for quite some time, and they have struggled
desperately to maintain their group activity in this area of
research. That is a separate but very involved issue in terms of
gaining information needed in the future. Most of the funding that
has been provided for this problem has come from National Marine
Fisheries Service, S-K grants, Sea Grant, and a few other sources.
If there are ways to obtain funding for the naturally occurring
problem with seafoods in general, these should be explored.
In the third panel, the influence of temperature and time on
V. vulnificus growth in oysters is an area with some residual needs
and resolution. One controversy concerned whether V. vulnificus
grows in oysters. Work in the Northeast Seafood Laboratory showed
a substantial amount of growth in summer oysters with indigenous V.
vulnificus left to grow at permissive temperatures. That is a
concern. However, we also heard that, on the Gulf Coast in
particular, in almost all of the states, the harvest practices now
used seem rather reasonable in trying to curb the time from harvest
to refrigeration and exposure to temperature. With regard to
transportation of shellstock, the situation appears somewhat
improved. Although we do not have data on this, it seems that in
the past five years, the ISSC's efforts in both of these areas have
had an impact. Icing may be a technique worth examining further.
It also was stated that perhaps a real focus of attention might be
directed to the retail or to the restaurant arena since we see a
preponderance of the cases coming from restaurants, at least from
Florida's study.
The last panel discussed the need to determine what the
critical control points and critical limits will be with regard to
HACCP and V. vulnificus. As pointed out, this organism is not a
contaminant in the sanitary sense. It is naturally occurring, part
of the ecology of the estuaries, and it is hazardous at the same
time, at least to those particularly at risk.
The HACCP regulation for seafood is in the comment period. It
can reasonably be expected to take effect sometime within two
years. This puts a fairly large burden on the dealer, transporter,
processor, who are not out there harvesting but must take in
shellstock, and with the understanding that there are no hazards
associated with it. So, it is a very difficult problem to sort
out.
I would like to bring up one thing from my own perspective.
Is it possible that the high and low levels we see associated with
oysters from the same reef and the same vicinity of the same reef,
and in different shipments of oysters that are perhaps being
consumed raw, might be related to the state of the oyster? Are raw
oysters on a given reef nice and healthy or are there a certain
number of them that are in decline or dying? And could these
harbor large numbers of V. vulnificus that we are simply not
seeing? Has anybody looked at dying or dead oysters for V.
vulnificus levels? Does anybody here know how to gauge the health
of the oyster? I am just pointing out that the raw biological
animals carry microbial flora. Certainly, a dying animal's flora
will change, and the predominant species that are very competitive
may take off very early. That may be V. vulnificus or vibrios in
general.
I would like to thank all of the panels and all of the
individuals who were invited to speak here. I recognize that it
took a lot of effort on your part. Your presentations were
excellent and have helped us understand a great deal more than we
did just two days ago. This workshop has represented an important
increase in knowledge. I will be puzzling over how we are going to
put that together in terms of trying to help the industry,
consumers, and the regulatory agencies. I appreciate your efforts
very much. Thank you for taking the time to be here. Thank you
very much.
Return to FDA Prime Connection Seafood references
Go BACK to the
CFSAN/FDA Food and Consumer Information pages
Hypertext updated by ear 1998-MAR-06