I.
Brief Summary of Rationale for Protocol
The
investigators seek to understand the cause and/or effect relationship between
the mutation in the gene for CFTR and the production of a dysfunctional
chloride ion transport protein (i.e., the genetic basis for cystic fibrosis
(CF)), hypothesized abnormalities in airway surface liquid (ASL) (i.e., mucin
peptides and mucus biochemistry), and chronic infection and inflammation. This
understanding requires the sampling of ASL from infants with cystic fibrosis
prior to and after the onset of infection and inflammation, using
bronchoalveolar lavage (BAL) during bronchoscopy.� There are no appropriate cell culture or animal models in which
to answer this question.
Infants
diagnosed with cystic fibrosis in the neonatal period will be enrolled.� These include infants with meconium ileus,
infants who are diagnosed early in life through genetic testing, and siblings
of known CF patients who are diagnosed prior to or just after birth. The
proposed sample size is 8-10 infants in order to obtain at least 4 to 5 infants
without infection prior to the first bronchoscopy.� The BAL will be performed three times in the first year of life,
starting at less than 6 weeks of age, at 6 months and then at 12 months of
age.� Appropriate controls will be
selected from age-matched infants without CF undergoing clinically-indicated
bronchoscopy and BAL.
II.
Procedures included in the Protocol (along with a
discussion of risks)
A. Flexible
Fiberoptic Bronchoscopy (with bronchoalveolar lavage (BAL))
The
procedure described in the research protocol includes being NPO for 4 hours,
placement of an intravenous line, and the use of 2% lidocaine for local
analgesia.
The
following modifications to the procedure as described in the protocol should be
made in order to minimize the risks of pediatric bronchoscopy.
1. The
protocol description lists �clinically indicated bronchoscopy� as an inclusion
criteria, suggesting that the BAL will only be performed if it offers the
infant the prospect of direct (clinical) benefit.� However, the procedural description on the same page states that
the bronchoscopy will be performed in a similar fashion to clinically-indicated
bronchoscopy (suggesting that the bronchoscopy to be performed in this study is
not clinically indicated).� From the IRB
discussion, it is clear that the bronchoscopy will not be limited to
those performed for clinical indications.�
This should be stated clearly in the protocol.
2. The
proposal does not address when the investigators would stop the bronchoscopy,
such as in response to apnea, oxygen desaturation, bradycardia, and so
forth.� Since one or more of the
bronchoscopy procedures may not offer the prospect of direct benefit (i.e., not
be clinically indicated), the procedure should be stopped sooner than would be
usual when performing a clinically indicated procedure of direct benefit to the
infant.� The investigators need to
clearly specify the criteria for withdrawal of an infant from the study during
a �research only� bronchoscopy.
3. The
maximum dose of lidocaine should be specified on a per kilogram basis (i.e.,
less than 7 milligram per kilogram), independent of the concentration to be
used.
4. There
are no inclusion or exclusion criteria listed for the second and third
bronchoscopy procedures.� For example,
if the infant will undergo a clinically indicated procedure, can the BAL be
performed at that time?� Also, if the
infant is in the midst of an acute respiratory infection, and a BAL is not
clinically indicated, should the BAL be deferred based on an increased
risk?� The investigator indicates in his
response to the OHRP inquiry that infants with signs of active infection or
respiratory symptoms will be excluded.� These
contraindications to bronchoscopy and BAL need to be clearly defined. In
addition, it is left ambiguous whether this exclusion also applies to
subsequent bronchoscopy to be performed in an infant already enrolled in the
study.� The BAL should be combined
with any clinically indicated procedures, and the research-only BAL should not
be performed during an acute illness that increases the risks of BAL.
5. The
investigators provided data about the complication rates from the UNCH
bronchoscopy suite from November 1998 to the present.� During these fours years, there were 2,171 procedures with 432 of
these procedures being in infants less than 12 months of age.� This information alone indicates that the
investigators are experienced in these procedures, averaging one bronchoscopy
every three or four days in the study�s target age population.� The incidence of adverse events is quite low
(less than 1 %), with only a few considered potentially serious. In experienced
hands, these isolated adverse events can be readily handled to prevent any
serious morbidity or mortality from the procedure.
6. The
bronchoscopy and BAL will only be performed by one of three experienced
pediatric pulmonologists, as listed on the protocol.� Trainees will not be involved in these procedures.� This should be explicitly mentioned in
both the protocol and the permission document.
7. The
grant proposal includes a number of other sites for which data concerning the
safety of pediatric bronchoscopy have not been provided.� In conversation with the investigator, it
was clarified that these sites will not be performing pediatric bronchoscopy
as part of this protocol.� This
clarification is important, for the data that supports the risk classification
of pediatric bronchoscopy should be site-specific.
8. The
language contained in the letter from the Data Monitoring Committee of the CF
Foundation that death is �virtually unheard of� from bronchoscopy requires
clarification.� The chances of death
are extremely remote, and usually reflect a failure to stop the procedure in
the face of difficulties and/or complications.�
For example, the death of a healthy adult after a research-only
bronchoscopy was due to a lidocaine overdose.�
The protocol used in that case lacked information about the proper
dosing of lidocaine, and the investigators failed to withdraw the subject when
difficulties arose in performing the bronchoscopy.
B. Procedural
Sedation
The
procedural sedation described in the research protocol includes presedation
with chloral hydrate, followed by intravenous fentanyl and midazolam.�
The
following modifications to the procedure as described in the protocol should be
made in order to minimize the risks of procedural sedation.
1. The
presence of an anesthesiologist is not mentioned in the protocol.� The guidelines for sedation and analgesia
provided with the review materials is for non-anesthesiologists.� Provided that the end-point of the
procedural sedation, and the medications to be used, are clearly specified in
the protocol, the presence of an anesthesiologist may not be necessary to
minimize risks. The guidelines for sedation and analgesia only mention the
use of ketamine, which is not appropriate for use with this protocol.� Rather, the research protocol indicates that
fentanyl and midazolam will be used for the procedural sedation.� A maximum dose of both agents should be
specified (such as up to 5 micrograms per kilogram of fentanyl, and 0.2
milligrams per kilogram of midazolam), along with the dosing frequency and
rate of administration (such as administering fentanyl in 1 microgram per
kilogram increments over no less than 2 minutes).� In addition, the endpoint of the procedural sedation should be
specified as moderate or �conscious� sedation.� This level of sedation does not impair protective airway
reflexes, nor require assisted ventilation.�
The procedure should be terminated (and the subject given reversal
medications) if a deeper level of sedation is inadvertently achieved.
2. The
policy for administration of procedural sedation indicates that infants fed
formula should be NPO for six hours, rather than four (which applies only to
breast fed infants). The protocol should be changed to be consistent with
UNC policy.
This
reviewer considers the fiberoptic bronchoscopy and the associated procedural
sedation to present greater than minimal risk.�
The data presented by the investigators suggest that the bronchoscopy
procedure (along with the procedural sedation) can be safely performed at
UNCH.� The UNC IRB determined that
bronchoscopy presents more than a minor increase over minimal risk.� However, no data nor specific adverse events
were cited in support of the IRB claim that the �risk of bronchoscopy in an
asymptomatic infant was of sufficient magnitude to be beyond that described as
a minor increase over minimal risk�� The
probability of any adverse event is extremely low, yet there are a few isolated
adverse events (i.e., laryngospasm, pneumothorax) which are of a potential
magnitude that this reviewer does not consider the bronchoscopy
procedure to present only a minor increase over minimal risk.� It should be emphasized, however, that these
events are rare, and can be handled by experienced clinicians without any
serious morbidity or mortality. This reviewer considers the approach to
procedural sedation outlined above to present a minor increase over minimal
risk in the hands of experienced pediatric clinicians.
III.
Comments on Permission Process (Undue Influence and
Coercion)
The permission form
should be re-worded to clearly indicate that the bronchoscopy and BAL will not
benefit the infant. Although Table 4 suggests that clinical decisions would
be based on the results of the BAL, there is little data to support the actions
taken under each scenario.� In the
absence of data, it is difficult to argue that the culture results may be
directly beneficial to the infant in the absence of another indication for
treating a pulmonary infection.�
Although the amount of money seems proportional to the
time and effort required for participation in the protocol, and a portion is
appropriately directed to the infant, one could argue that the possibility for
undue influence on the parent to enroll his or her infant should be altogether
eliminated.� The parents should be compensated for expenses only, as an infant
of this age would require adult attendance even if not enrolled in the
study.�
As opposed to what
is stated in the protocol, the study should be presented to a parent by an
investigator who is not the infant�s primary pulmonologist.� Especially in a setting such as UNCH
where bronchoscopy is a fairly routine clinical event, every attempt should be
made to avoid the misconception that the bronchoscopy procedures included in
this protocol are for the direct benefit of the infant.
The descriptions of
the procedures and risks in the permission form are not complete. For
example, there is no mention of the NPO period, the risks of the 2% lidocaine,
nor the specific risks of the medications used for the procedural sedation
(such as chest wall rigidity with fentanyl infusion).
IV.
Additional Concerns
Why are three as opposed to two bronchoscopy procedures
necessary?� The scientific purpose of
the protocol is to ascertain the pathophysiologic role of changes in mucin and
mucous.� The investigators, when
questioned, indicated that if the second bronchoscopy was delayed until after
chronic infection was established, it would be difficult to determine if the
changes were primary or secondary to the infection.� The investigators should
address the scientific necessity of the three bronchoscopy� procedures in the protocol.
An independent Data
and Safety Monitoring Committee (DSMC) should be established to review any of
the adverse events that occur during the course of the research.� If an adverse event occurs, no further
research bronchoscopy procedures should be performed until the AE is reviewed
by the DSMC and the research allowed to continue. This reviewer considers the
occurrence of any irreversible morbidity or mortality as a reason for the study
to be stopped immediately. No child should ever be placed at risk of
irreversible morbidity or mortality during research that does not offer the
prospect of direct benefit.
One issue to be considered is whether the storage of
specimens for unspecified research should be considered optional.� As the bronchoscopy and BAL does not offer
the prospect of direct benefit, the storage of specimens as part of the
research may not be considered coercive (i.e., failure to agree to specimen
storage will jeopardize the infant�s health care).� There is no direct benefit within the research that would be denied
if a parent did not participate in this research protocol.� However, it is still appropriate to offer
the choices of withdrawing the specimen from the tissue bank at some point in
the future, and specifying that consent should be obtained for any future
studies.� The materials reviewed indicate
that this issue will be handled in accord with the policies of the UNC IRB:
however, no information nor documents were provided on this point.� This
deficiency needs to be corrected.
The study should
only be performed if there is compensation available for any physical injury
that may occur as a result of participation in the research.� Contrary to the letter dated February 7,
2003, the UNC IRB is precisely in a position to alter institutional and/or
state policy on an ad hoc basis, especially when considering approval under
�46.407.� Compensation for research
injury is one of the essential �sound ethical principles� that must be met for
any research to be considered under �46.407.�
The waiving of professional fees is insufficient, as this is a small
part of the total expense of caring for any research-related injury.
V.
Should these studies be performed in older subjects
prior to infants?
Given the fact that infants with CF become colonized
and/or infected early in life, the scientific questions asked by this protocol
can only be answered through
performing a bronchoscopy during the first months of life.� One can ask, however, whether it is more
appropriate to study infants who present with either meconium ileus,
respiratory symptoms or failure to thrive, rather than the �apparently well�
infant diagnosed with cystic fibrosis through screening techniques.� If, for example, infants with meconium ileus
underwent a bronchoscopy at the time of surgical intervention, what would be
the impact on the risks of the research, on the ability to answer the
scientific question, and on the feasibility (i.e., recruitment of an
appropriate sample size) of the research?�
In addition, do we believe parents of an infant newly diagnosed with CF
will have the capacity to provide research consent under the duress of emergent
neonatal surgery?� Is it ethically less
(or more)� challenging to enroll infants
with CF under these circumstance?
The grant application (signed 7-18-97) has limited
applicability to the protocol outlined in the materials.� In the grant application, bronchoscopy and
BAL appears limited either to children with cystic who are less than 3 years of
age undergoing a clinically indicated bronchoscopy or infants born with
meconium ileus who would undergo bronchoscopy and BAL while undergoing surgical
repair of the bowel obstruction.� It is
not clear why the investigators expanded on this subject population to include
asymptomatic infants diagnosed with CF.�
There are two possible reasons.�
First, the need to perform follow-up BAL raises the issue of a
non-therapeutic bronchoscopy anyway.�
Second, there may be insufficient infants born with meconium ileus at
UNCH in order to achieve an appropriate sample size in a reasonable time period.� Performing
the first bronchoscopy at the time of an initial surgical procedure minimizes
the risks of the BAL, and should be done whenever possible.� Parents should be able to provide informed
and voluntary permission in spite of the duress of an urgent surgical
procedure.� Permission should be
obtained from someone other than the clinicians caring for the infant.� This reviewer, however, would not limit
enrollment to infants with meconium ileus, given the safety record of the
investigators in performing pediatric fiberoptic bronchoscopy.
VI.
Application of the Criteria for IRB Approval
A.
Apply the
general criteria of 45 CFR 46.111.
Provided that
changes in the protocol are made in accord with the italicized recommendations
(above), the risks to subjects are minimized by using procedures which are consistent
with sound research design and which do not unnecessarily expose subjects to
risk.� This reviewer considers these
changes in the research protocol to simply document the precautions and
approaches that the investigators are likely already using in their clinical
practice, given their excellent safety record.�
When the bronchoscopy can be performed safely (i.e., clinically
indicated), it is appropriate to use procedures already being performed on the
subjects for diagnostic or treatment purposes (�46.111(a)(1)).� The selection of subjects is equitable,
taking into account the purposes of the research and the setting in which the
research will be conducted (�46.111(a)(3)).�
The risks to subjects are reasonable in relation to the importance of
the knowledge that may reasonably be expected to result, as there are no direct
benefits to the subjects.(�46.111(a)(2)).�
Nevertheless, when some or all of the subjects are likely to be
vulnerable to coercion or undue influence, such as infants, ... additional safeguards
have been included in the study to protect the rights and welfare of these
subjects (�46.111(b)), requiring us to apply the additional protections found
in 45 CFR 46, Subpart D.
B. Assess the risk presented by each
intervention or procedure in the proposed research.
As discussed above, the bronchoscopy and procedural
sedation present more than minimal risk, and thus cannot be considered under
�46.404/50.51.� The bronchoscopy and
procedural sedation present greater than minimal risk (�46.405/50.52 or
�46.406/50.53), and thus require us to evaluate the possibility of direct
benefit to the infant.� This reviewer
agrees with the UNC IRB that the evidence for a prospect of direct benefit is
insufficient to consider the bronchoscopy under �46.406/50.53.� Although the procedural sedation, taken
alone, presents no more than a minor increase over minimal risk
(�46.406(a)/50.53(a));� the sedation is
only a means to an end and does not result in any knowledge to ameliorate the
infant�s disorder or condition (�46.406(c)/50.53(c)). The fiberoptic
bronchoscopy presents more than a minor increase over minimal risk, and thus
requires consideration under �46.407/50.54.
�
Although not stated in the regulations, one could argue
that a procedure not offering the prospect of direct benefit yet presenting
greater than a minor increase over minimal risk should nevertheless meet the
other criteria listed under �46.406/50.53.�
The bronchoscopy is an essential part of this research protocol, and
would likely yield generalizable knowledge about cystic fibrosis which is of
vital importance for the understanding or amelioration of cystic fibrosis (�46.406(c )/50.53(c )). Premature
death from CF occurs as a result of chronic bacterial lower airways infection,
leading to bronchiectasis and eventual respiratory failure. This research could
serve as the basis for developing new therapeutic agents to improve lysis and
hydration of mucus. In addition, the research presents experiences that are reasonably commensurate with those inherent
in the infant�s actual or expected medical situation (�46.406(b)50.53(b)).� Apparently,
the majority of infants diagnosed with CF at UNCH undergo a
clinically-indicated bronchoscopy by nine months of age.� Finally, based on the review and discussion
of the protocol, this reviewer believes that the research does present a
�reasonable opportunity to further the understanding, prevention, or
alleviation of a serious problem affecting the health or welfare of children.� (�46.407(a); 50.54(a)).
C.
For all
categories, consider the requirements for parental permission (�46.408;50.55).
With the
italicized modifications outlined above, this reviewer determines that informed
consent (that is, parental permission) will be sought (and appropriately
documented) from each subject's legally authorized representative, in
accordance with, and to the extent required by �46.116 and �46.117.
(�46.111(a)(4,5))
D.
When
appropriate, there are adequate provisions for monitoring the data collected to
ensure the safety of subjects. (�46.111(a)(6))
E.
When
appropriate, there are adequate provisions to protect subject privacy and to
maintain data confidentiality. (�46.111(a)(7))
With the italicized changes outlined above,
this reviewer determines that these two criteria for IRB approval are met.
Final Recommendation:
����� This reviewer finds that the research
under consideration does not satisfy the conditions of �46.404, �46.405,
or �46.406.� However, after the
recommended modifications, the research �presents a reasonable opportunity to
further the understanding, prevention, or alleviation of a serious problem
affecting the health or welfare of children; (ii) �will be conducted in
accordance with sound ethical principles; and (iii) adequate provisions are
made for soliciting the assent of children and the permission of their parents
or guardians, as set forth in �46.408.�
����� Consistent
with sound ethical principles guiding pediatric research, and assuming that the
modifications recommended above are in fact made, the exposure of infants with
cystic fibrosis to the risks of pediatric fiberoptic bronchoscopy in the
proposed research is scientifically appropriate and necessary, and can be
performed safely.� It should be
emphasized that this determination is limited to the UNCH site based on the
documented experience of the investigators.�
The protocol should not be performed by less experienced
personnel. The infant subjects will be chosen to minimize the likelihood of
individual harm and maximize the likelihood of gaining knowledge that furthers
the understanding, prevention, or alleviation of a serious problem specifically
affecting the health or welfare of other infants with cystic fibrosis.� The risks of the research to the infants are
balanced by the importance of the knowledge gained.� After some modifications, the research optimizes a parent�s
capacity to give permission, and to understand the anticipated experience and
risks of the research.