SGDEPARTMENT OF HEALTH AND HUMAN SERVICES

FOOD AND DRUG ADMINISTRATION

CENTER FOR FOOD SAFETY AND APPLIED NUTRITION

CONTAMINANTS AND NATURAL TOXICANTS SUBCOMMITTEE

OF THE FOOD ADVISORY COMMITTEE

VOLUME II

Wednesday, March 19, 2003

8:39 a.m.

4700 River Road

Riverdale, Maryland

PARTICIPANTS

Francis Fredrick Busta, Ph.D., Chair

James E. Heubi, M.D., Co-Chair

MEMBERS

Alex D.W. Acholonu, Ph.D.

Lawrence J. Fischer, Ph.D.

Marion H. Fuller, D.V.M.

Lawrence N. Kuzminski, Ph.D.

Ken Lee, Ph.D.

ACTING INDUSTRY REPRESENTATIVE (NON-VOTING)

R. Bruce Tompkin, Ph.D.

TEMPORARY VOTING MEMBERS

James Anderson, Ph.D.

Robert D. Baker, M.D., Ph.D.

Larry R. Beuchat, Ph.D.

Henry M. Blumberg, M.D., Ph.D.

Margaret E. Briley, Ph.D., R.D., L.D.

Laurie J. Moyer-Mileur, Ph.D., R.D., C.D.

Marguerite A. Neill, M.D.

Virginia A. Stallings, M.D.

Phillip Tarr, M.D.

Patti Thureen, M.D.

C O N T E N T S

Page

Opening Remarks 4

Questions from the Committee 7

Subcommittee Discussion of Issues, 47

Recommendations, and Response to Charges

P R O C E E D I N G S

Opening Remarks

DR. BUSTA: Good morning. It looks like we are all here on time and ready to go, bushy-tailed and no snowstorm, here in College Park.

If we could convene the committee meeting at this point. A few announcements. First of all, microphone style. It's the switch closest to the mouthpiece into which you speak, that is the off-on switch, the one that is closest to the end of the microphone. The bottom one, you just don't bother with. Just work with the top one, that is the on-off, and it has to be turned off when we are not using it because we get a lot of feedback if we don't.

The second important announcement. On the sheet that we got yesterday called Charge and Questions, the most recent draft, if you would strike "Draft," because those are our final questions that we are addressing today, so it is really not a draft. That is the set of questions that we are addressing.

That is the one that, if you recall, the third line in Charge 1, No. 1, is factors, not facts. That is how you differentiate that final draft, last draft, but it is no longer a draft, it is, in fact, the Charge and Questions.

A number of committee members have indicated that there are some questions they would like to ask for clarification and explanation this morning before we continue on with our discussion of the charges and our committee responses, so arbitrarily, we are going to take 30 minutes. We will limit it to 30 minutes of asking questions of the presenters from yesterday.

We really need to limit it to 30 minutes, so we will try to be succinct in our questions because otherwise, we could spend I imagine the whole day grilling various people for details, so I think just key questions, we will take 30 minutes and try to get some of those clarifications made this morning.

If the individual isn't here that needs to answer the questions, then, we will just have to go on.

Are there any other items that anyone on the committee would like to bring up?

The approach to the questions that we are contemplating, not necessarily final depending on your committee, for example, after the first 30 minutes, again discuss Charge 1 as we were doing just before 6 o'clock yesterday, discuss that as best we can, and the Chairs will do their best to summarize what that discussion was and what we think is a consensus and see if everyone agrees with that consensus.

If there are dissenting comments, then, obviously, those can be put in the record, as well. If that doesn't work well, we can go around with each individual making a statement, but we will try that approach first.

Is that all right with everyone? We will give that a try.

I see Dr. Tompkin with a microphone in his hand, poised and pointed, so I assume that you wish to start.

DR. TOMPKIN: I would just like to make a comment. As we progress with the discussion and as we try to reach a consensus, there will be some issues where we don't have all the data we need, and are we capturing the data gaps separately, because we can only take a discussion only so far in the absence of more data.

So, can we somehow create a parking lot for those kind of things as we move through, and then come back and discuss them further?

DR. BUSTA: I think that is an excellent comment, as always. We will do our best to capture those here, and I would think that Charge 2, No. 4, Critical Knowledge Gaps, would be the place to accumulate those, and that is toward the end.

The other thing is that you all should have gotten four handouts from the public comment group that are the slides that they used in the last presentations.

Questions for clarification.

Questions from the Committee

DR. FULLER: Two questions. One is if we can have anyone from the industry that spoke yesterday, one of the questions we had asked early on had to do with what were the steps taken in, I believe it was the Norwegian plant, that resulted in their not having problems further.

Then, the second question, totally unrelated, has to do with the temperature of mixing the powdered formula, and we heard that boiling water created problems with protein coagulation. I am curious, because we also heard that 70 seconds at just a few seconds resulted in significant kill, you know, what happens at 75 degrees C or 80 degrees C, do you still have that problem.

DR. BUSTA: Do we have an industry representative to respond to that?

DR. SMOOT: I will try to respond to the first question regarding the Netherlands plant. After the situation that occurred and the data that was used yesterday to demonstrate microbiologically improvements in the process and the environment, the primary actions that took place were in the hygienic design of the process equipment itself where there was an elimination of water used.

Part of the spray dry process is a water scrubber, so we found in the environment, there were pockets in areas that were designed into the process that brought more water into this dry processing environment.

So, there was considerable engineering redesign efforts to improve the process and eliminate the water that was currently in that type of process stream and to improve the management and use of water in and around that area, as well as the knowledge to further investigate breakdown equipment, design equipment that is easily accessible, you know, designing in the cleanability of the equipment.

These were the types of things that were the primary, as well as then enhancing our surveillance in terms of the efficacy of good hygienic practices that were being taken in the factory at that time, you know, a continuous improvement.

These are the types of things that, in that particular factory, as well as learning some best practices there, have been continually being spread throughout that particular organization. I am sure the learnings there have found their way throughout the infant industry, as well.

DR. FULLER: So, did you end up having to replace equipment or were they able just to make changes to existing equipment and water, and I understand you did say water use, as well?

DR. SMOOT: Both, there was both. I mean there was redesign of unit operations, as well as replacement of equipment.

DR. FULLER: Thank you.

DR. SMOOT: I don't believe I would be qualified to approach the second question.

There is a point of information from Bruce.

DR. TOMPKIN: The data that you provided were data on product.

DR. SMOOT: Correct.

DR. TOMPKIN: In addition to that, there was a lot of samples collected, I would assume, from the environment, and those data are not included, so whereas, in each year, you had 700 to 900 and some samples analyzed of product, that was supplemented with some additional environmental samples.

DR. SMOOT: Correct. The data that you I believe now have a copy of, in that one table, that was a focused study that ran concurrent to the normal process monitoring for finished product, which you have seen again cited in the Codex risk profile, and many of you have made reference to where there is a more stringent norm for coliforms, less than 0.3 npn/gram. That was the normal operating monitoring for finished product.

This dataset that you have access to was a concurrent focus study to look deeper into that, and then as Dr. Tompkin pointed out, as a supplement to that enhanced environmental monitoring, was also taking place at the same time, yes, and that data has not been shared.

DR. HEUBI: Just a quick question also about that data. Is that product that was actually stored in like a bag and then checked, or was it tested during the process of drying?

DR. SMOOT: I would say that I don't have intimate detail, but knowing the nature of the process monitoring, it would have been minimally into the big bag tote stage of the process.

DR. BUSTA: Dr. Kuzminski.

DR. KUZMINSKI: This is for Dr. Smoot, a question. On your slide yesterday, it's on page 4 of the handouts which were provided this morning, and thank you for that, the slide yesterday on current manufacturing intervention strategies, and you outlined there programs and practices, such as HACCP, supplier QA, et cetera, all the way from the raw materials to the finished product steps of the manufacturing process.

My question is related to this. If we consider the Tennessee incident in 01 as an event, were all of these practices in place at that point in time? I am not saying that your organization is linked to the Tennessee event, I just don't know--

DR. SMOOT: I would have to defer specific comment to that particular event to people more knowledgeable of the quality and safety procedures in place. I could not speak for that particular event.

DR. KUZMINSKI: Is there someone here, others from the industry, that could answer that question?

DR. SMOOT: Yes, I believe so.

DR. MARCH: My name is Dan March. I am with Mead Johnson. Yes, I can say that the HACCP programs were in place at the time. We looked at the quality of that product at release. It was probably one of our best batches microbiologically that we had produced under the standards that currently existed at that time.

DR. KUZMINSKI: Thank you. If these processes or programs were in place at that time, what different has industry done since the event in 01?

DR. MARCH: HACCP is an evolving process, and it is always open to improvement, so we have instituted improvement in our HACCP programs in the plants to take account for some things that we had not noticed in the past, such as use of water and minimization of water in certain processing.

We have gone to a higher level of HACCP, looking at different control points or controlling what we thought were critical control points to a higher level, so it is an evolving process.

DR. KUZMINSKI: And the sampling plant that was described on finished product yesterday, the 5 grams in 5 for the Enterobacteriaceae. That would be something new in terms of the evolving HACCP, to verify HACCP.

DR. MARCH: Yes, it would, exactly.

DR. KUZMINSKI: Thank you very much.

DR. BUSTA: We had a second half of a question on the lower temperature, 70 to 80 degrees C causing coagulation or clumping in the product. I know the statement was that very hot water caused clumping yesterday. Does that occur at 70 or 80?

DR. MARCH: I can speak to that. We had done studies with the boiling water and had shown that there is an immediate, as Dr. Buchanan's graph showed, there was an immediate loss of temperature when you add boiling water to product that is tempered at room temperature.

So, you have to take into account there is immediate loss of temperature, so therefore, in order to get the temperatures of what we would say pasteurization around 70 degrees where the kill is most effective, you do need to heat it up somewhat, 85, or something like that.

We do know that somewhere in that range between that temperature and boiling, there is going to be some effect on the protein, potential coagulation. Again, vitamin C losses will still be substantial in the presence of oxygen and heat.

Also, we do need to consider the potential burn hazard even at 75, 80 degrees to the preparer and also to the infant. Does that help?

DR. BUSTA: Dr. Stallings.

DR. STALLINGS: To clarify that, then, if we were talking about preparation in the home, we could ask people to boil water and add it to room temperature formula or in a nursery setting.

What would be the temperature that we would achieve in that sort of setting, are we going to be at the 70 degree killing temperature, or are you suggesting that if you did that--part of what we are trying to figure out is, is that effective in killing bacteria that would be in the product at the point of preparation.

DR. MARCH: Was that to boil the water, cool it, and then prepare products with it? I am sorry.

DR. STALLINGS: I thought the question may be a misunderstanding, it was about preparation of a dry powder, so if I were doing that at home, I would be boiling water, adding it, and you reminded us that you have an immediate temperature drop, and what is the temperature of a mixed formula, and it cools.

The question is are we going to get any killing if we use that.

DR. MARCH: Yes, there would be an immediate kill. If you used boiling water, there would be an immediate, once mixed, immediately, but the graph that Dr. Buchanan showed, showed that there was an immediate loss of temperature down to somewhere around 80, 85 degrees.

That is still too hot, and it showed by leaving that out over a period of time, the cooling curve did flatten out, so it took much longer then to drop down once it reached equilibrium with the product, it remained at a temperature that is still capable of causing burns.

DR. STALLINGS: What is an acceptable serving temperature just so we have an idea of what that number would be?

DR. MARCH: An acceptable serving temperature is probably going to be around body temperature, you know, 98 degrees or something like that--37 C.

DR. STALLINGS: Let's stay in one unit or the other. So, we are starting out with boiling water at 100. We mix it, we are at 80 something. We have got to let it, in an equilibration, come down to 30.

DR. MARCH: That is correct.

DR. LEE: On the same subject, you know, I bottle fed three of my own at home. One of the first things I do is I pour some of the formula on my hand to make sure it is not too hot.

Are we saying that these healthcare professionals are not capable of observing the temperature of what is being fed to the neonate? I am kind of missing--there seems to be some significance given to the potential burn hazard to the healthcare worker and to the baby, but I think most parents do this at home all the time, so I am just wondering what is different, because to me, it's a tradeoff to kill these bacteria.

DR. MOYER-MILEUR: I think in a NICU situation what you would have is this formula is being prepared in a formula room. They would have to be held I guess at a certain temperature and then cooled, and then taken to the unit and then rewarmed to room temperature in a Level 3 NICU.

Now, whether or not you are going to have people who are trying to mix it just prior to a feeding or not in a smaller nursery, I don't know.

DR. BUSTA: Dr. Fuller.

DR. FULLER: Thanks. I guess what I am trying to get at is--and I don't have the experience here--but I was thinking more along the lines of if we have identified a population that is very highly at risk, and yet we have also heard there is a very real need for some of these products, my question is can you not heat the water to, say, 85 degrees, yes, you will have a rapid dropoff, do you then have enough to mix with formula, you know, mix for one minute, let cool a minute, mix with formula a minute and then rapidly cool to the temperature either for storage or to serving temperature.

That is where I was going with that, and does that take care of the protein coagulation problem.

DR. BUSTA: I see no volunteers.

DR. THUREEN: I would like to just add this, because it is the same question.

How significant is the clumping issue, would it be enough to clog up a tube or clog up a bottle, or is it just an inconvenience, it might not be digested as well, and is the remaining protein of the same nutritive value, which is really the important question, because boiling, I mean it seems like a good way to get rid of the bacteria, but are we really going to be destroying the product, so that it is not effective for nutrition.

DR. MARCH: I don't believe that I can respond to the clumping and the degree of clumping, however, when we get into--going back to the previous comment--there was a thought of doing an incremental heating and cooling.

I guess what we are looking at there is some very complicated hospital protocols that could possibly become very confusing. I don't mean very complicated, but at least more complicated than it is now. I guess that is the danger of that. If not properly heated and cooled, you could be getting yourself into an incubation situation if you are not warm enough then.

That would be my fear of a microbiologist warming the product up 10 degrees doubles or I guess shortens the growth of bacteria. I guess I would discourage that in that it does complicate the matter.

DR. BUSTA: Dr. Heubi.

DR. HEUBI: I have a manufacturing question. Yesterday, Dr. Zink told us that there was not any major advantage to one of the other drying methods for formula, and now both Dr. Smoot and Dr. March have told us that in response to some of the issues that were raised, they reduced the amount of water used in this process.

Is it your opinion as industry that, generally speaking, the less water that is utilized, the safer the procedure is, and are you moving in the direction to minimize water in this process?

DR. SMOOT: Very succinct, yes. The issue of use of water, parts of the process, whether you are using steam heat or water cooling of some type, there is water somewhere throughout the system, so engineering designed to minimize that exposure to the process stream is one aspect.

The other, as we talk about spray dried runs, I believe Dr. Zink had mentioned yesterday his preference to run longer and dryer, I think even Dr. Tompkin had pointed out is very true. However, due to formula changes and allergen issues and cleaning, you have to address allergen issues with wet cleaning, so depending upon production schedules and changeover, you have to do some type of wet clean sanitation.

But again, working with our partners in sanitation practices, we are trying to even improve how we do wet cleaning in a controlled wet clean scenario. So, yes, by all means, it is something that we are working very hard with in industry.

DR. BUSTA: Dr. Thureen.

DR. THUREEN: Thank you. So, if we assume that you can't eradicate all E. sak contamination at the plant level, and that you could potentially eliminate it at the nursery level, I think most nurseries would be willing, with boiling, to go ahead and take that step particularly since most Level 3 nurseries have milk labs with fairly well trained people.

So, it seems like if you don't know, that there could clearly be steps outlined that would show effective means of heating and cooling, might require special equipment, but you could develop that, that nurseries, with all their costs, would clearly be willing to invest in something like that, so that there should be able to be research done that would identify the proper process for that, and if you get a good kill, then rewarming shouldn't be a huge issue for bacterial contamination, at least to E. sak, or not. I mean that may not be true.

There are clearly other bacteria in the process and you are going to introduce them during the process, but if you could have a good heating and cooling procedure, and then rewarming procedure, and then do studies to see if the protein is denatured or not, then, you could at least say this is an intact product, maybe we need to add some extra vitamin C after we finish the product or give the baby vitamin C supplementation, but it seems like if we can't clear up the problem at the manufacturing level, we could at least go a long way at the preparation level.

Clearly, there is going to be a lot of education involved in that, but it seems to me that that would be a part of the solution.

DR. BUSTA: That was not a question.

DR. THUREEN: It is a question in that there was a lot of no head shaking from that little group over there that this is not a realistic solution.

DR. BUSTA: Let the record note that she was pointing to the previous respondent.

DR. THUREEN: Several of them. There are some yes's and no's over there. Do you not think that is a realistic solution?

DR. GEHRIG: Good morning. This is Tom Gehrig with Wyeth Nutrition. My background is mostly in food technology, in that area. I have done some work in denaturing of milk proteins, soy proteins, and I guess if you look at the milk proteins, your wheys particularly, they start to denature about 60 degrees C, and as you go up the time-temperature relationship, they continue to denature.

The soy proteins are even a little bit more susceptible and my quandary would be that you are going to have to make it so precise on the mixing that, you know, we design these, we do a lot of protein efficiency studies to ensure that the protein is intact, that I think you would start to see, especially through the nasogastric feeding tubes, especially when the diemers go down, that you would have some problems with plugging of those tubes.

We know we have plugging in the nipples of the baby bottles, and I think it would be a big issue if we start taking these things up to 80 degrees and holding them for any amount of time.

I think it works, but, you know, from the industry standpoint, we spend a lot of time on our processes, especially on the evaporation, on the drying, to try and minimize the changes to the proteins, and I fear that if we take and reconstitute with 80 or 90 degree water, and hold it for any amount of time, we are going to be in some trouble.

DR. THUREEN: Thank you. That really lays the issue to rest for me, and I didn't think I had had a good answer before, so thank you for that.

DR. BUSTA: Dr. Lee.

DR. LEE: I appreciate that observation. If one were to anticipate reformulating with hot water, say, even as high as 100 degrees centigrade, one could design the proteins appropriately, so they are soluble, perhaps hydrolyze them somewhat, maintain your protein availability.

It is really a process design question, is it not? I mean the protein is not adversely--correct me if I am wrong--the protein is not destroyed nutritionally by this boiling water syndrome. I mean you can solubilize protein in hot water.

DR. WALLINGFORD: John Wallingford, Wyeth.

I think this is an area that we need to work together on. We don't have all the information we need, so if I could recommend this be one of the parking lot items where we actually go out and get the data at what happens to the physical properties with formula after different temperatures of mixing and actually after subsequent mixing different temperatures of holding prior to feeding, I think this is an area where there is a potential for some improvement in the ultimate bacterial safety of the product.

DR. BUSTA: Dr. Tarr.

DR. TARR: I have two questions.

First, yesterday, you talked about microbial sampling. I forget what the exact numbers were, but I know the numerator was expressed in grams and the denominator was expressed in tons.

How validated are these sampling techniques, can we be confident there, in this part of the HACCP protocol, you are getting reliable data from your production line? That is my first question. I can wait for the answer.

DR. BUSTA: Dr. Smoot, it looks like he is being nominated.

DR. SMOOT: Referring to the validation of this type of sampling plan, currently, the sampling plans for coliforms in our finished product monitoring now is at the level of N equal 5, and the sample size equal 1 gram.

The proposal, continuing with a N equal 5, in other words, number of samples, is consistent with an ICMSF Class VI sampling plan for fecal coliforms. That would give you a certain level, and has through time been validated as an acceptable sampling plan for a bacterium of moderate risk.

So, what we have proposed from industry is to cast a broader net on our process control monitoring going from coliforms to the family of Enterobacteriaceae first, and then secondly, increasing the sample size from 5 grams to 25 grams.

So far we have found at least working with the N equal 5 one gram for coliform has been reasonably successful in monitoring and assuring the process control, and what we were proposing yesterday was a step moving into the more stringent with both the target of the sampling, as well as the sampling size.

DR. TARR: A follow-up to that then, and this might have been answered yesterday, the incriminated product in Tennessee, did that meet, exceed, or was under the cut points for what you would consider acceptable for your process?

DR. MARCH: Yes, it did meet the specifications at that time for coliform organisms according to the plan that Dr. Smoot just laid out.

DR. TARR: To meet the future protocols or the proposed?

DR. MARCH: Retrospectively, that would be a difficult question to answer because it would be hard to make the equation--there would be some challenges, but I have a feeling, and it is only speculation, that it would meet that.

DR. TARR: And then my second question unrelated, it looks like the breast milk substitute is an important priority for both medicine and industry, and we would like to keep that available, and it might have to remain in powder form for the near future.

How close, though, is industry towards coming up with a liquified higher caloric concentrated form that could be sterilized and administered safely? We heard yesterday that it was technically difficult, but how active is the scenario developing?

DR. BUSTA: It doesn't appear that we have anybody on that. For clarification, Dr. Zink said that the breast milk supplement was tested.

DR. ZINK: Going back into our little database of samples, two human milk fortifier products were tested and classified as transitional formula, so in the survey we did two human milk fortifier products were tested.

DR. TARR: What were the results?

DR. ZINK: They were negative.

DR. BUSTA: In regard to the plant that you showed the data on that were reducing the count, and the last data for 2002, the fourth one, there were 6 positives and all 6 were E. sak positive, and that appeared to me that as the plant was improving all of their cleaning processes and design, et cetera, it ended up selecting for E. sak.

Am I reading that wrong?

DR. SMOOT: Clarification. The data that was presented was on a percentage basis, so it was 0.6, not 6 positives, individual positives.

But, yes, as our experience with that particular factory, and across the 70, 80 that we have around the world, as we improve our hygienic practices and reducing water and driving down and monitoring the environment, the comment made yesterday, and I think it was borne out as well by Dr. Buchanan's data, is that this particular organism adapts well to warm, dry environments.

So, as we continually monitor the environment for the Enterobacteriaceae family, which is basically a biologic indicator for the presence of water, hence, we use it to assure we are keeping the environment as dry as possible, because they need water to grow.

High levels of EB indicate that you had water somewhere in the system at an unacceptable level. But with refining with this particular organism, E. sakazakii, is that out of this group of the Enterobacteriaceae family, it does appear to possess the adaptation to warm, dry environments, so as we are driving that number low, yes, we are starting to see exactly what you observed.

DR. BUSTA: So, in fact, another placeholder for research would be trying to determine why ecologically we are selecting for that type of organism and how we can modify the ecology to eliminate that kind of organism.

DR. SMOOT: I would say yes, that is a correct assumption.

DR. BUSTA: Peggy.

DR. NEILL: I think my question is somewhat headed in your direction, Les, but it combines some of the information that Don Zink gave us yesterday with his illustration of the process, as well as the implications of the most recent data that you were just explaining.

Don painted a fairly powerful portrait that by the time the product is coming through the dry sifter spray dryer, it has undergone several kill steps, and the subsequent suspicion, obviously not proven, is that this is a contamination of product post that step.

My question relates to the following. What are the technical obstacles for creating a sterile product post the spray dryer including conceptualizing that it would not have to be the entire product lot, production run, et cetera?

In other words, if we are steering our way towards identifying that there is only a subpopulation of recipients of formula that are at risk for this infection, then, one could conceptualize that you only need some of this formula to be sterile and directed towards that particular patient population.

DR. SMOOT: First of all, you know, we have been successful in industry in terms of environmental control at assuring the eradication for the most part from the process environment, other pathogens of concern, such as Salmonella and Listeria.

As we look at the Enterobacter sakazakii and its place as a member of the Enterobacteriaceae family, though maybe not totally convinced, but we are finding data on a weekly basis that it is a little bit more ubiquitous than what we maybe thought to begin with.

So, this organism has been able to eliminate it from the environment post-heating steps of these either wet mix, dry mix technology streams. The proposal as I understand it then, can we divert or can we provide some type of unit operation for a small production portion to go into almost a Class I00 sterile room type environment.

I would say on the scale of the industrial process, the total process, yes, that would be very difficult, is that something that we could again maybe work in partner, is this technologically feasible, is something like this possible, I wouldn't say no, it is not possible, but this is something the industry would have to study in terms of a risk-benefit there.

But you are basically talking about going to a Class 100 room type environment for post-drying process stream, and you saw the magnitude and size of the equipment to do that even after the cooler, after dryer, which is the first unit operation post-star valve on the bottom of the cone. To even put that in the frame of a sterile environment would be very difficult because what we are finding, that is a part of the areas that we need to improve anyway in terms of hygienic design.

There would have to be a considerable learning to go to that level of process control to this sterile environment.

DR. NEILL: I think my comment is that in the hospitals, we do this all the time in terms of having a triage of medical products and foods that are adjusted to the perceived risk of the recipient population.

I mean it's pasteurized egg product now in most hospitals, so that is where I am coming from in terms of a conceptual approach. I think what I am trying to ask is really much more at the level of the technologies.

I am intuiting that an ultra-high temperature step is likely to cause difficulties with the protein denaturation, so that you may end up with an insoluble product, and I don't think we heard a clear answer yesterday to irradiation, and all irradiation is not created equal because there is certainly different types.

What are the perceived obstacles in the physical-chemical nature of the formula that would prevent creation of a commercially sterile product? I think maybe that is a better way for me to articulate the question.

DR. SMOOT: I might have to defer to people more intimately involved, maybe Tom, to someone more involved in the actual process or in technology. Would you want to speak to this? I defer to Tom Gehrig, Wyeth.

DR. GEHRIG: I was responsible for design of our new factory largely, in part, in Singapore, which we started about a year and a half ago. We took a good look at this, and, you know, speaking to what Les said, the volume of the air that moves through one of these dryers or in the conveying systems is tremendous, you know, in the volume of like 270,000 cubic feet per minute.

What we do is try and install Hepa filters on the inlets to these dryers, but where it gets difficult is, you know, we can control on basically the transfer of the powder into the packaging operations, and we take a lot of care to ensure that that air is filtered especially going into the packaging hauls.

We set up basically a hospital room environment where employees that move into that area have to do full gowning. They have got masks, they have got gloves, full head gear, so in the particular zones where the powder can be exposed to the environment or to the employees, we actually zone kind of on a concept of 1 through 5, 5 being the most stringent zone.

The only people allowed in that area are the people working in that area. Where you get into some difficulties is when you have a breakdown in the line and you have to have an intervention, and the employees have to be very well educated when they go into this equipment, especially a mechanic or something, that those tools basically have to be sterilized or sanitized before they go into the system.

I think, on the industry's part, it is requiring us to really go back and look at our processes, kind of what Les was saying, that whenever there is an intervention, we write it down, we mark down what has been done, who goes in, that the equipment has been sanitized and cleaned.

But to create a clean room environment basically from the spray dryer south, so to speak, into the packaging line, I am sure it could be done, but it is going to be extremely difficult to do.

It has been our experience that these powders for the E. sak, we can get down to counts that are actually less than 100 CFU per gram, and we still have E. sak in there, so as far as the drying process goes, it is essentially sterile as far as the powder operation goes.

So, the bug is selective, and we could take controls, but I guess my concern is when you are operating a plant, we can't really operate to two different standards. I can't say this product is going to be diverted off and it's going to go through sterile product, and this other product is--we really have to operate to one standard, and especially if you are taking term infant formula, and it is being added in, we know this is being done for low birth weight, to supplement the formula, so we would have to produce really the term formulas also to that standard.

DR. BUSTA: Dr. Baker.

DR. BAKER: I would like to take that one step further. It seems that one possibility or a good way to get around this whole problem is to develop a sterile product. We have sort of beaten this sterile powdered formula into the ground.

Is there a way of providing everything that the sterile powdered formula can do in a liquid form that could be sterilized, so we just completely move away from powdered formula in the NICU setting, is that a possibility for the industry to come up with high-concentration formulas that could be added without adding a lot of volume is what it comes down to.

DR. VANDERHOOF: Jon Vanderhoof. I guess the question is could we as an industry convert our entire manufacturing process to providing nothing but liquid formulas worldwide, and I think the answer to that is no.

Could we provide selective products for the NICU as liquids, and I think the answer to that is most, and we are doing that.

Thirdly, there are probably some products that are going to have to be continued to be used as powders at least for the time being until we come up with better solutions for that.

We have heard a lot of interesting ideas with both some good and some bad in them, but I think we are all moving in the right direction. I think the likelihood going down the line is that most of the products that will be needed in NICUs will be available as liquids, but there are probably going to be some small numbers that will be needed as powders that we will probably just have to take extra precautions with.

DR. BAKER: I was only talking about NICUs, not about your whole worldwide production of powdered formula, and I guess the problem with the milk fortifier, human milk fortifier, which I guess now is not produced as liquid, but, for instance, the amino acid formulas, they presumably could be produced as liquid, is that not true?

DR. VANDERHOOF: Well, I guess I am the only physician here, and we don't have an amino acid formula, so I can't really tell you except that most of the ones that are available are powders, and the ones that I have had experience with using are powders, and you are seeing more and more use of these formulas in high-risk children.

Certainly, in my prior experience, up until six months ago, as a pediatric gastroenterologist in a transplant center, we used an awful lot of amino acid formulas in powder, and I don't see that practice changing very much, so your concerns are right on.

Whether that can be made as a liquid or not, I don't know. I think there are some problems making amino acid formulas in liquid. I can't really tell you what they are. The hydrolysates you can make as liquids, and we do that.

DR. BUSTA: I am doing a terrible job of chairing because we are already 15 minutes past the maximum, but Dr. Stallings, and after Dr. Stallings, just one more, so whoever thinks they have got the greatest urgency will have to signal me.

Dr. Stallings.

DR. STALLINGS: I just wanted some clarification. I am still struggling, and I am sure it is because my microbiology isn't that good, but the issues of sampling. When we were hearing the information early in the day, there seemed to be an important point to get to our bug of interest, that we needed to be sampling like at 100 grams for this particular organism to be able to say if it's there or not with some reasonable assurance.

Now, we are at a sampling of an N of 5 times 5 grams, which gave us a 25-gram per lot, and we know that we don't understand exactly what a lot is in the sense that probably varies from each production site.

But I still am not sure that what is proposed is an adequate sampling for the organism that we are interested in, and I understand we have gone from fecal, the most general term of fecal coliforms to the family, but we haven't gone to the bacterium that we are really talking about.

So, I would appreciate a little more understanding of why that response is most appropriate for addressing this issue.

DR. SMOOT: One of the questions I think the committee is here to look at, and we from industry, as well as other presenters, are trying to provide information to come to what is, if there is, a safe allowable level for this organism in a powdered infant product as it is delivered to be reconstituted.

The proposal by industry again would provide, not going over the enhanced more stringent from where we are at today, but what we are looking at is an absence of 25 grams type of sampling plan.

Based on information that industry has relative to our understanding of the relative risks involved with the delivery of a powder at a certain microbial level of this particular organism, we feel, in terms of a process monitoring control, if we can assure that the process of delivery product of absence in 25 grams in terms of a dose or a feeding provided to an infant, that you would provide yourself with a safe allowable level. I mean in a sense, that is what we are proposing.

Yes, it open for question and debate, but we are proposing that this is a starting point for this broad group of Enterobacteriaceae absence in 25 grams is an acceptable level.

DR. BUSTA: One last urgent. Dr. Lee.

DR. LEE: I actually am going to change the subject a little bit and ask the industry formula guys to think of the whole product line. I would like to know, particularly in these particular days of the global conflict, Tom Ridge has asked us in the food processing industry to place particular vigilance on the quality and wholesomeness and safety of our food supply.

I would like to know, without giving away any specific details, is our formula manufacturer and distribution network doing its best as far as security and homeland security can go particularly in terms of minimizing counterfeiting and diversion of these products, because, you know, feeding the babies is critically important to us in America, and it is a point of intervention that I think I would like to hear somebody tell me what is being done.

DR. COULTER: Julia Coulter, Mead Johnson.

I am responsible for our efforts around food security at Mead Johnson, and I can tell you that shortly after the 9/11 crisis, all of the industry was invited to come to Washington and talk to the FDA about a risk assessment process.

They rolled out an excellent methodology, which they asked us to apply, and I know that we have gone back and we have done risk assessments in our firms and identified gaps where there might have been an opportunity for an unwanted act to occur and have invested significant amounts of money in bringing ourselves closer to where we need to be as far as food security.

DR. LEE: Do you feel good about the trace-back, if I should happen to walk into WalMart and find something that I question whether or not it came out of your plant, do you think you can do that?

DR. COULTER: We all are regulated to have product traceability, so yes, we have that in place and would be able to identify back to point of manufacture if something occurred.

DR. BUSTA: Dr. Lee has this incredible capability of bringing us back to reality, what the threat is.

Subcommittee Discussion of Issues, Recommendations,

and Response to Charges

DR. BUSTA: Let us, if it's all right with everyone, return to Charge 1. We were discussing this as we closed yesterday, and I am asking if there are individuals who would like to comment or discuss, in a preliminary fashion, characterizing the infants at risk.

What I recall hearing yesterday, let me say it in the negative, I don't recall anyone saying that there is not a risk from E. sakazakii in powdered infant formula. I heard people say there is a risk existing, and then went on to go on to the next sentence there and basically identifying specific populations that were at risk, specifically, in the compromised infants of one way or another identifying those and some of the factors, wouldn't you like to have additional discussion.

Let me restate that. Have we adequately identified the infants at risk and the factors that make them at risk to this infection?

DR. HEUBI: I am going to make my prerogative as the Co-Chair.

As I gathered from what we talked about yesterday, it was the sense that pre-term infants who are less than 36 weeks gestation, up to the age of about 4 to 6 weeks of age, were an at-risk population, and that individuals, infants and children of any age who were immunocompromised were potential at-risk population. Not considered at risk were term infants greater than 36 weeks gestation.

The question I still was sort of left with yesterday was if there was concern about term infants who were in the NICU environment, and whether that was a consideration.

DR. STALLINGS: I think that is one of the important points. When I was jotting down my notes last night, it was we have incomplete information. We know, though, that, if infected, it is often a fatal infection or one that results in brain damage.

So, I think given that we don't know how to describe the immune status of newborns very well, certainly not in a clinical setting, that my advice at this point would be to say, if you are sick enough to be in a Level 3 nursery, which many people would include many term infants, particularly term infants with surgical diagnoses, that I would say that that, for a starting point, should be anybody that is sick enough to be in a Level 3 nursery, I would put at risk.

It also has some practical advantages that that is a nursing unit, it's a formula delivery unit, it's an organization that we understand.

My question then would arise, given the nature of healthcare in the U.S., and referral patterns and all the other things, is are we sure of who the babies are at Level 2 nurseries because in the past, many of those babies, we would have expected to be in Level 3 nurseries, but there is a lot of pressure to deliver some care particularly more standard neonatal respiratory care at a Level 2 setting.

So, that is the place that I don't have as much confidence in, but I am pretty comfortable that Level 3 nurseries should be a site of interest.

DR. BUSTA: Dr. Thureen.

DR. THUREEN: I think that extending the same guidelines to Level 2 nurseries is probably a reasonable thing to do, because it sets a standard since most of those are manned by Level 3 physicians.

Also, most of those babies in Level 2 nurseries are probably not going to be getting on any of these powdered formulas, near-term infants, and that the same kind of concocting is not going on in that population in the Level 2 nurseries and the Level 3, but just setting that standard and gearing the education to both Level 2 and Level 3 nurseries makes a lot of sense because if just simplifies the process and makes the issue less confusing.

DR. BUSTA: Dr. Moyer-Mileur.

DR. MOYER-MILEUR: My only comment would be on the age limit. I think when we are saying 4 to 6 weeks, we need to be careful because I think we need to say 4 to 6 weeks post-term, because I think it makes a big difference to a baby who is born at 23, 24, or 25 weeks.

DR. HEUBI: I have a comment. Does the FDA understand what this designation of Level 2, Level 3 is, so that we don't need to clarify that?

Patti, do you want to clarify?

DR. BUSTA: The request was to clarify it.

DR. THUREEN: Level 3 designation is for the most intensive care and generally, also includes the ability to do surgery and cardiovascular surgery.

So, some of the Level 2 units, it varies around the country, and what is considered Level 2 unit in different places is very different, because some units will vary and take care of small babies with intensive respiratory needs, but because they aren't able to provide surgical services, are considered a Level 2, so people were refer to those as 2 1/2.

That is not an official designation, but some Level 2 nurseries only take care of nearly well babies, so a Level 2 nursery is hugely variable, and that is why applying it to Level 2 and Level 3 nurseries, the same standard of Level 2 and Level 3 nurseries would cover the entire gamut of different types of Level 2 nurseries that are in this country.

Did that answer the question?

DR. BUSTA: Is that clear? Okay.

Any further comments? As a Chair who has only learned all about this, this week, it really sounds quite complete.

Co-Chair?

DR. HEUBI: The question is now do we want to go around and get each person's consensus, attitude about this, so we can have it for the record before we go to the next charge, I guess that is the next question.

DR. BUSTA: Finalize this one.

DR. HEUBI: Yes, we could get away from this one, I think.

DR. BUSTA: A real vote.

DR. HEUBI: A real vote, we could do a real vote.

DR. BUSTA: The industry people, of course, do not vote. You will recall from yesterday that Dr. Beuchat had a special event that he had to go to this morning, so he will not be voting.

I see Dr. Fischer with a microphone in his hand.

DR. FISCHER: Before I vote on this, I want to make sure that I understand, that there is no risk to immunocompromised full-term babies.

DR. HEUBI: I don't think that was what my comment was. When we actually started this, I actually included that in the criteria that was included, yes, immunocompromised of any age.

DR. THUREEN: Would you restate that?

DR. HEUBI: What I stated was that for infants who are pre-term infants less than 36 weeks gestation, up to the age of 4 to 6 weeks post-term are at risk, also, immunocompromised infants and children of any age.

We also entered that infants of any gestational age up to 4 to 6 weeks post-term age in Level 2 or 3 nurseries.

DR. THUREEN: But I think your comment is, is there no risk to the healthy term infant. No, I don't think we can say that. We believe that there is minimal or very low risk, but that we can't absolutely say there is no risk.

DR. BUSTA: Dr. Neill.

DR. NEILL: An amendment somewhat to what Patti was just saying might be to include the phrasing "an even lower risk." In other words, we have said there is a risk to the subgroups, but we can't peg it exactly, but whatever it is, it is even lower for the term healthy child.

DR. HEUBI: I would accept that as an amendment.

DR. BUSTA: If there is a risk for a full-term infant, it would be lower than the group that we are describing.

DR. HEUBI: Right, and I think the other point that has been raised, the information that we have been provided actually doesn't allow us to actually assess the risk of either one of these currently because most everything that has been described to us has been case series that doesn't really give you any kind of denominator, it's all sort of like the numerator without a denominator.

DR. BUSTA: Dr. Thureen.

DR. THUREEN: We could say there is a low but as yet unquantified risk for all these groups, but in this order, or then we categorize them by what we think is relative risk.

DR. BUSTA: Did you take that as an amendment, too?

DR. HEUBI: Absolutely.

DR. BUSTA: Dr. Tarr.

DR. TARR: Do we need to go down the route of qualifying the risk as low, medium, or high, all we have to state is there is a risk, yes or no, and are we asking for trouble by putting a modifier on the risk?

DR. BUSTA: Are there other opinions on that?

DR. HEUBI: I am a little reluctant to quantify the risk because we just don't have a good answer.

DR. TARR: And "low" has a connotation to it, so does "medium" and "high."

DR. BUSTA: Dr. Thureen.

DR. THUREEN: As a clinician and knowing pediatricians in the community and interacting with these people, I think that pediatricians need to know that for their practices, we believe that this is not completely risk-free, but really a safe product by most general standards, and that this really is referring to high-risk populations. Our deliberations on prohibitions or recommendations against minimizing use or how to use these formulas really refers to just the high-risk populations.

So, a statement about well term infants, I got calls about this all the time when it first came out, but does it apply to babies in my office practice, and this would answer that question, and I think that is what people out there want to know, I think we have to make some comment about relative risks in different populations for the practicing physician.

DR. HEUBI: The answer may be that term infants are at lower risk.

DR. THUREEN: That's right.

DR. HEUBI: That's all, because we don't exactly know what the risk is to the pre-term infant for sure.

DR. BUSTA: We had some nodding by several of the committee as to lower, what Dr. Heubi related.

Other comments? Okay. Are we ready?

All in favor of the description of our response to Charge No. 1, raise your hand.

[Show of hands.]

DR. BUSTA: It appears unanimous.

Charge No. 2. Dr. Heubi says we are now halfway there. No, that one has four parts.

If there is a meaningful risk, and we have established or indicated we have a meaningful risk for selected populations, how can this risk be addressed?

No. 1. What intervention strategies can be used in infant formula manufacturing processes and plants? Discussion. Dr. Kuzminski.

DR. KUZMINSKI: Thank you, Frank.

We have heard a lot from industry on the programs and practices that they do in the manufacturing environment, and taking HACCP to a higher level, as they have described, is the new activity since a point in time in 01.

I fully agree that HACCP, as a practice, is an evolutionary practice based on learning, and you keep taking it to higher levels. I understand the complexity in the manufacturing process and in a manufacturing facility.

I firmly believe that there is no silver bullet here in terms of if we do this, this problem will now be solved, but I do believe that very serious effort--and I don't doubt the effort, I believe these plants have been taken apart and put back together again in the last couple of years in order to find out more knowledge as to where this bug comes from--but I do believe also, knowing the size of some of these units processes, unit operations in these, that there is room for process development research in terms of food safety and hygienic design especially with these humongous pieces of equipment that these processes employ.

So, is there a specific intervention that can be recommended? I guess I must say I don't think so, but I think a coordinated full court press needs to be continued and maintained in concert with the agency as to how HACCP continues to be taken to a higher level on the manufacturing floor by the industry.

So, in terms of a specific recommendation for an intervention, I can't make one, but in terms of a blanket recommendation of taking HACCP to a higher level, I think there can be more collaboration with the agency by the industry as to here is what we are doing.

I can recall five or six years ago that the agency put out a call to the food industry looking for volunteers actually, volunteer companies that might come forward and put HACCP into their food plants.

That is a model that perhaps could be used with this specific industry as to how this industry is taking HACCP to a higher level on a collaborative level.

DR. BUSTA: Other comments? Dr. Fuller.

DR. FULLER: My thoughts are sort of fuzzy, but let me throw them out for discussion anyway. I am wondering if some recommendations to develop, I don't know, an industry guidance document that would speak more specifically to whatever the specific measures might be, but something that could be developed in concert with industry. I think they have the experts there.

We have heard some discussion about design, engineering designs, as well as process designs, that can help in this area. The use of hygienic zones apparently has stepped up since at least the Tennessee event, if I understood that correctly.

I think yesterday we heard, as well, something along the lines of product or ingredient specifications coming into the plant, you know, some of those things. Again, I mean I am just trying to recall some of that.

The other thing that one of my questions, another question I wanted to ask, I think FDA mentioned that there were some GMPs that were going to be reopened for comment. It was not clear to me whether those GMPs addressed the powdered infant formula as well.

I think we heard yesterday in one of our opening comments that it did not or the existing GMPs did not, I wasn't clear on that, but that might be another avenue for addressing some of these things, as well, in a longer term.

DR. BUSTA: Requests for comment.

DR. TAYLOR: We, as an agency, apologize for the microphone situation. As soon as we get funding from Congress, we will buy more microphones.

In terms of reopening the comment period, in fairness, we will reopen on the entire rule. Obviously, we are looking to refresh the data and get better input for anything that might have happened since 1996.

So, we will not just be taking comments on one particular area, the full rule is open, but clearly, we will be asking questions specifically targeted in areas that we think have become increasingly problematic.

In 1996--if you would like copies, we can certainly get it for you--we proposed GMPs, as well as the definition for quality factors. This was in response to the statute that does allow separate GMPs for infant formula.

It is broad reaching, it goes across the entire terrain of GMPs. It is a very long, extensive rule.

DR. FULLER: And it does cover the powder formula.

DR. TAYLOR: It covers all.

DR. FULLER: Great. Thank you.

DR. BUSTA: Comments? Dr. Fischer.

DR. FISCHER: Under Charge 2, No. 1, we must talk I think about the monitoring procedure again. The industry has proposed this monitoring scheme and sampling scheme, and basically, the scheme starts out with a change from a quantitative analysis to a presence or absence test.

I don't think we have heard why they want a change from a quantitative test to one that is just yes or no, they are there or not there, and I guess I am not experienced enough in sampling to know what is the best thing to do, but my gut feeling is that I would rather know how much is there, if it is there, and work from that aspect rather than work in a presence or absence mode.

I guess I am not very keen about the switch that is proposed by the industry here, and I would much rather think about, if we are going to go to a presence or absence, I would say there should be an absence of all of the offending organism there.

The trouble with going to presence or absence, I think, is simply what is the sensitivity of the test. If you made a test of inadequate sensitivity or used a test of inadequate sensitivity, you could have almost anything there.

So, I think the first thing we ought to do is decide whether, in fact, we want to recommend that the change in the type of test from quantitative to non-quantitative, if you will, the presence or absence, my recommendation was not to accept, I think, this presence of absence test.

DR. BUSTA: Response? Dr. Tompkin.

DR. TOMPKIN: Well, actually switching from a quantitative to a qualitative does increase the sensitivity, and actually analyzing the full 25 grams for presence/absence is a more sensitive approach to it.

I don't think, as a panel here, that we can define a sampling plan that is or is not appropriate, and I think that that should be the responsibility of FDA really working with the industry, but I think we could indicate that there is a need for review of the sampling plan as it existed and as has been proposed for the agency to consider that and arrive at a proper sampling program.

The sampling program that was proposed by the industry does fall in line actually with history in terms of Codex and International Standards for this class of product. Moving to a presence/absence does increase the stringency of the sampling plan, so it is a step forward in that sense, but this alone is not the true indicator.

It is only an indicator of the conditions under which the product was produced. That is all it is doing, it is not providing a specific level of confidence that a pathogen is or is not present.

You have to consider that this sampling plan that has been proposed and that has been used historically as part of a total food safety management system, and that is really to follow on an earlier comment that the HACCP is being enhanced, but really what the industry is doing--and this is not just the only industry moving in this direction, other food industries are moving in the same direction--you have to consider that it's the supporting documentation that occurs whether it is HACCP and the GMPs.

But in this case and in other cases in the food industry, it is the environmental sampling that really provides an additional supplemental level of confidence, whether it is, in my case, my past, sampling for Listeria and the environment was tremendously important in our ability to produce lowest risk possible, ready-to-eat meat and poultry products.

That worked very effectively for us as a company, and the industry has moved in that direction. This is similar in that sense, in that controlling the environment is critical to managing the food safety risk and that by having an effective sampling program in place to measure that level of control is an important aspect.

I posed that question to the industry, and certainly FDA has access to the procedures, how those procedures are applied, and so on, so it is not a secret thing. I would say that you have to consider the whole package rather than putting your reliance on an indicator test on the finished product.

DR. BUSTA: Dr. Fuller?

DR. FULLER: No.

DR. BUSTA: Dr. Blumberg.

DR. BLUMBERG: I think it gets back to the risk again because the sampling, the procedures they are using now, 1 gram or this 5 gram, in the outbreaks, they took 100 grams to be able to detect E. sakazakii, so for healthy babies, there is low levels there, it probably doesn't make a difference, but for the high-risk infants, the current sampling probably isn't good enough, so I think it just gets back to the population at risk and whether you want to use this product or switch to another product in the neonatal ICU, like a liquid product that is sterilized.

I think that gets back to the issue, because I think saying, well, we have sampled at 1 gram or 5 grams, it is not sensitive enough, it sounds like, to detect E. sakazakii using those sample methods. For most people, it probably doesn't matter, but for the high-risk infants in the neonatal ICU, it may make a difference.

DR. BUSTA: Dr. Fuller.

DR. FULLER: Thanks. I will make my comment.

We have I think heard that the powdered formula cannot, at least for the moment, be manufactured to produce a sterile product at this point, so we know that there is going to be risk, and absolutely we have identified the population we think at risk.

The sampling done, if I am understanding correctly, the in-plant sampling, be it finished product or environmental sampling, is done as process verification, that is, to ensure that whatever manufacturing processes are in place, they are minimizing the points at which there is exposure to the product and the amount of bacteria, the bacterial content or load, or what have you.

So, I think--and this is where I don't have the expertise--I don't know if that 25 gram plus or minus is an adequate process verification or not. I think maybe that is probably what is giving all of us some difficulty because the sampling that we were hearing about in the infant formula, I believe is for a different purpose.

I mean the sampling that we have been hearing about with the 25 gram is to determine whether the procedures in that production line are adequately controlling exposure. We don't know what those infective doses are or anything like that. It just is that process working to minimize that risk.

I don't know how to get to, and I don't think we have the expertise here, to determine if that is adequate for that purpose or not, but I think that is the point of discussion for both of those issues, and whether you release product that contains, I think is another issue all together and I am not sure how to answer that.

DR. BUSTA: Dr. Stallings.

DR. STALLINGS: Thank you, Dr. Fuller, I think that is very helpful. I was sitting here thinking if those of us who really do clinical practice or a group of neonatologists in nutrition support people, if we told them the story, I think that is one of the critical issues.

That is why I am not reassured by the industry appropriate sampling for product leaving the door when I know that we have a fatal bug that, at least as I understand the data that were presented yesterday, the only way that we have some reassurance as to whether it was there or not is the 100 gram sampling.

I think that is where I have a real quandary, and I think we are going to be, or the agency is certainly, and industry going to be facing a period of time where we don't have enough information, we don't know if we can change more manufacturing processes, we don't know if we can quickly get the sterile products through things that are just good practice and not regulatory.

But what are we going to do for the year or two while we know this exists, and if somebody said, look, you know, is it safe to feed this product to this baby in the NICU, assuming that it is not going to clog the tubes and all of that, how do we reassure them.

If I were presenting this as a seminar to my colleagues, I would say, well, it sounds like from my infectious disease colleagues, the way to be reassured is a sample that is significantly greater, and we don't know any denominators. That has come up again and again, we really don't know the exposure rates, we don't have enough information to do risk assessment, so what do we do in the meantime.

So, really, I see this as sort of the crucial next issue is, is there something we can do even if it's in an interim stage while more information in the animal studies and the infectivity and the dose, and all of the things will probably will come up out of the No. 4 section, is there something that we can be doing now that would give reasonable reassurance for clinical care for the high risk group. You know, that is where I think from an infectious disease point of view, what can we do now, is there anything.

DR. BUSTA: Dr. Thureen.

DR. THUREEN: This doesn't answer this question directly, but I still think the most important thing we are going to come away with is educating physicians, nurses, et cetera, that these are not what these products were intended for, and if you choose to use them in a different way that they are not intended for, these powdered formulas as additives, that you are running a risk of potentially putting this patient at risk for an infection. It is a low risk, but it's there, and that there is so much more that we have to learn, that you just have to realize the risk and you are taking that risk until we get more information that can be more definitive.

DR. STALLINGS: As a follow-up, the only thing that I see as the exception would be the breast milk fortifiers, and right now there is a sterile liquid product available and some of the others are powder, if I understand the product lines correctly now.

Then, like you said, everything else, there is sort of an alternative, but as was presented yesterday, and most of my life has been spent trying to argue for good growth velocity, you know, in children with chronic diseases, so I think it will be a hard balance if we are trying to take off products that people have learned to use effectively to get good growth. So, it will be a challenge.

But what about the breast milk fortifiers?

DR. THUREEN: I think the people feel they are going to have to use powdered breast milk fortifiers because the liquid concentrate is good, but it just dilutes the breast milk and adds too much volume, so I think that they are still going to be there no matter what recommendations we make, people are still going to use fortifiers, but in terms of all the other concoctions that come up, I think it could change practice.

There will be units that even with all the education and knowing all the facts, will choose to continue their current practices of making up their own formulations, but it has got to be an educated decision, I think, and I don't think the education is completely out there.

DR. BUSTA: Dr. Tarr.

DR. TARR: Assuming we are staying focused on the high-risk, hospitalized infant, is it possible to filter sterilized fortified breast milk, pass it through a millipore filter? Will there be too much retention once it is solubilized?

DR. BUSTA: From my experience, it is hard to filter more than about 10 milliliters of milk through a given small membrane filter, so that would seem difficult.

DR. HEUBI: In response to that, we have done not things with milk, but with other volumes, and you are right, I think it would add another technical aspect to this that would make it very difficult for staff to do it timewise, because I think Frank's point is a good one.

It takes a good while to push things through a millipore filter that is the size that you are going to trap bacteria, because breast milk formulas, the breast milk additives are still a black box, we have no idea how microbiologically safe they are. There is no data yet.

DR. MOYER-MILEUR: I think we need to remember when we are using powdered human milk fortifier, we are adding it to mother's milk, which does have immunologic properties, and that maybe some research needs to be done to see what bacterial counts are after this product is added to mother's milk.

Now, it needs to be done in a number of ways, because mother's milk is stored in a number of ways. Sometimes it is given in a fresh state, while other times it has been frozen and then thawed, but I think we really need to keep in mind that mother's milk does offer some protection to the infant.

DR. BUSTA: Dr. Neill.

DR. NEILL: I think I have been sitting here finding myself like everyone else, sliding into Question 2, under Charge 2, which I think most of us saw yesterday as being the central conundrum to this meeting.

I suspect our discussions so far have gotten us to a point that--throw it back at the Chair--we should struggle to articulate the responses for 1, and that is something along the lines of there is intervention strategies you can do, but we really cannot say that we could HACCP our way or test our way to safety for this at-risk population, because I think that is what we just said.

We can go back and add some supplemental sentences that talk about the need to do it nonetheless, but even so, we all just heard through much of yesterday that it can't get us to a point by which we would say the albeit low risk has been now made quantitatively lower to a fashion that it is acceptable to feed the product to the at-risk population.

But in terms of I think trying to keep our eyes on the horizon, I think I am seeing that we have made the statements that we could make at this point for Question 1 under Charge 2.

DR. BUSTA: Dr. Fuller.

DR. FULLER: Okay, let me throw something out and really get us going. I think, to follow along what you are saying, we certainly can't HACCP our way to a sterile product, and I don't think anybody is attempting to do that.

We have talked about a number or heard about a number of intervention strategies that industry has talked about. They have told us that here are some things we can do.

I guess, for me, the place that I am still having the hardest problem gets back down onto that one issue of if you are sampling your finished product and understanding that you are sampling that for process verification, and I don't know how to do that sampling in a perfectly meaningful way, but I think that the bottom line is if you find something, do you release the product, and I wasn't clear in yesterday's presentation. I think what I heard was that in some cases, even with the sampling regimen, whether it is more or less sensitive, that you would release product with a positive finding.

If that is the case, you know, I question that. I would like to hear some discussion on that. Otherwise, I think we are saying, yes, we heard the industry present, here are the intervention strategies, we talked about some of them, and then I think this, that get to me anyway, is the one place that I am not sure we have addressed, and my question is do we need to address it.

DR. BUSTA: Dr. Kuzminski.

DR. KUZMINSKI: I sense that there is some frustration here on the committee's part in coming to a finished statement for Question 1 under Charge 2.

Perhaps a way to address it is to build on what Dr. Fuller has said, that perhaps there should be more transparency at least to the agency as to what industry is doing to heighten the level of taking their internal intervention programs to a higher level.

An ability to communicate to the agency that a sampling plan has, whatever the sampling plan is, they proposed one, a common methodology to be followed by the industry, but whatever it is that we, as an industry, have X level of confidence that this sampling plan indeed represents a lot, and if we find this, this, and this, that we have X level of confidence that the product either meets or does not meet specifications down that decision tree.

That is what I haven't heard here in the presentations, and for me, in a simple way of looking at it, it would help me understand the goodness or badness, if you will, of the thoroughness of the program.

DR. BUSTA: Comments? Dr. Thureen.

DR. THUREEN: I think your points, both previously and now, combined with Marion's, are excellent and really get to the heart of the matter, and that our trying to figure out a sampling plan that would reflect a lot's purity or risk is not part of this committee's assessment, but should be done between the agency and the industry with statisticians who can clearly, I would think, come up with a sampling plan that would reflect a lot, I mean that's got to clearly be out there, and that we should just stick with general guidelines and that particular point be addressed in a combined agency-industry effort.

DR. BUSTA: To reiterate the question, No. 1, what intervention strategies can be used in infant formula manufacturing processes and plants.

Dr. Acholonu.

DR. ACHOLONU: I seem to be getting the feeling that we are talking about methods rather than answering the question, what intervention strategies can be used. Industry has helped, they have responded enhanced environmental monitoring and controls using HACCP principles.

I feel that does the job, and then they have another statement here, complementary final product testing for Enterobacteriaceae. We are looking for method, not the actual practice what to do.

A general statement what intervention methods can be used, and there are 1, 2, 3, 4. That is what I feel we should be doing, and I am of the opinion that industry has done a good job of suggesting those kinds of responses.

Thank you.

DR. BUSTA: Dr. Neill.

DR. NEILL: Well, I think Bruce is going to kill me for doing this, but, Bruce, could you speak a little bit about the difference, intellectually, not statistically or methodologically, for testing for process control and finished product testing, since I think your comments in many other venues have been enormously instructive to many of us, myself included, and it might be helpful if you could try to speak to that now.

DR. BUSTA: Dr. Tompkin.

DR. TOMPKIN: There is historically in the past, microbiological sampling of individual lots was the approach that was used to determine whether or not individual lots were acceptable, and therefore, whether or not a food operation was in compliance.

There has been a movement toward a more holistic approach, and essentially that involves sampling, for example, from the environment as one source of data, and to collect the data, but then track it and plot it in a manner that you can follow trends and with the idea that we indicate, the data would indicate that there are some positives in the environment.

I think we have to accept that, but the question is what is the number, what is the prevalence rate, and how low can you go in any given operation and track it at that level, and if you do see a trend showing an indication of a higher prevalence rate in this case, then, you determine why.

It can reach a point where the data indicate that you should shut the operation down and essentially reclean and restart. So, you are controlling the process and using the data to track it.

That can be done over time in terms of months, and you have to look at the data, don't just collect the data and put it in a file, it has to be used in a meaningful way.

In a similar manner, the testing of finished product, those data also should be tracked, so we are talking about 25 grams total being analyzed with 5 subunits. If you are up one this week and another one the next week, that indicates that there is an intermittent level of control and your goal is all zeros.

So, by tracking those values, you can again get a measure of trends and whether or not your process is in control. Those can also be supplemented by, as we heard yesterday, that a production run may involve 20,000 pounds over, let's say, five days just as an example.

To understand the process, that means that you probably should, well, would be prudent to collect samples at the beginning of the run and at various intervals throughout the production.

That allows you to know whether or not you have a problem at the beginning as a result of cleaning, as we mentioned yesterday, or if something is occurring in the process that, over time, leads to additional contamination.

That information allows you to determine whether or not a five-day run is acceptable or not in terms of control. So, that is another means of collecting information. You wouldn't need to do that on a continuing basis, but it could be done on an intermittent basis, whatever is appropriate to the specific facility.

Also, in the event you do have an intervention, mechanical breakdown, and you have to go in and do some work on an operation, that would be a time where additional sampling would be a prudent thing to do, don't just rely on what you have been doing, do some additional investigational sampling of the environment and/or the product.

So, all these kinds of information then can be used to come up with a comprehensive understanding of whether the process is in control, and there are different approaches from a statistical standpoint, then, that can be used to use the data from a statistical standpoint and arrive at some level of confidence of control.

Is that helpful at all?

DR. BUSTA: Dr. Baker. We are way past a break. Does your question need to be or can you wait until after the break?

DR. BAKER: I think this is quick.

DR. BUSTA: Okay. Go ahead.

DR. BAKER: I have two questions actually. One of them is I understand that the sampling is a process sampling, it is not meant to assure that the formula that is produced is safe, and I understand that a negative test does not mean necessarily that every can in that batch is free of E. sakazakii.

So, that is one thing, but then if you get a positive and then you are planning to send this out to nurseries where you know there may be a risk, that seems to me that is a kind of a different world than the question what you are doing. With your eyes closed, it may be all right, with our eyes open, is it okay. That is one thing.

The second part of the same question is, is there a way that you could do the 100-gram test in a sub-lot and supply that to Level 1 and 2 nurseries as opposed to the rest of the world?

DR. BUSTA: First of all, I think that sampling plan did not imply that if you found E. sakazakii, that you still would ship it. I think the response was in regard to the Enterobacteriaceae and whether that was positive, and then whether it was a high temperature organism implying fecal coliforms, and the decisions would be being made on that, not on classifying E. sakazakii, if I recall that sampling plan that was proposed.

DR. TOMPKIN: The point at which any testing for E. sakazakii was not defined in that sampling program, and I think the group is having some difficulty with the idea that you are coming up with a positive sample for Enterobacteriaceae in the lot, in one of the five subsamples.

That is still a more stringent, much more stringent sampling test than has been used in the past. When it comes to a sampling plan, the sampling plan is internationally recognized with the five subsample units, and so on, as an indicator test.

So, there is a statistical and international recognition for that program. The decision as to whether you ship with a C equals 1 or a C equals 2, the C equals 1 is accepted, it has been accepted in the past even with the older procedure, analytical procedure.

When it comes to the C equals 2, it is debatable, and I agree it's fuzzy, what does all this testing, what does all this evaluation going to consist of, and I think this is an area where FDA is going to have to have some input into that.

This is a proposal, and it has to be considered also that what we are undergoing is a process of reaching some level of testing, some level of control, and so on. What we are going to propose or recommend today is not what is going to happen hereon, because as new information comes along, there may be other needs for testing.

So, I think that the C equals 2, if that is a problem for you, I think that that is where FDA, in fact, FDA will have input into this proposal. That is all it is, is a proposal.

With that, I just should say that all the testing in the environment does include testing for E. sakazakii, as well as for Salmonella, and those are larger samples.

DR. BUSTA: We will take a 15-minute break and we will come back. Thank you very much.

[Recess.]

DR. BUSTA: Dr. Heubi.

DR. HEUBI: I was asked at the break to help to clarify our response to Charge 1 in terms of what we, as a group, perceive the risk to the normal healthy term infant, and we qualified it as being lower, but I think if it would be agreeable to the committee, we would probably place it at the level of minimal, but we don't know specifically what the risk level is.

DR. STALLINGS: As I remember the discussion, part of what we were trying to stay away from were words that had a lot of value judgment to them, because we have no data, so that is where I think we came up with, you know, there is we recognize a real risk here, and the other one is lower, but I guess I would react to calling it minimal makes it sound like we have had information to evaluate.

Now, from what we know, I would bet that it is minimal, so I think it is just an issue of how the committee wants to deal with that language in the absence of data.

DR. HEUBI: Alex. I am not going to try your last name even though you went through all that.

DR. ACHOLONU: Acholonu.

DR. HEUBI: Acholonu.

DR. ACHOLONU: Please forgive me. I need some clarification here. I would like industry, maybe Dr. Zink is here, to tell us what is meant by final product testing. Do you do this test when you have put the baby foods in containers, or you have in a big container and you just come and take about 5 grams or so, and if you do, do you take it from the top, do you take it from the bottom, what is going on, what do you mean by a lot, testing of lot? Can you explain that to me?

DR. HEUBI: Can we hold that for just one moment? We want to finish Charge 1.

DR. ACHOLONU: All right.

DR. HEUBI: We will not forget it.

DR. BUSTA: Dr. Thureen.

DR. THUREEN: Well, I would still go back to saying there is a risk, but as yet unquantified, and it just says we don't know what the risk is, and then put them in the relative order and just say upfront we don't know what it is and admit it, and not try to get too detailed in the definitions of what minimal and moderate meant.

DR. STALLINGS: To keep our dialogue, if I heard the Co-Chair's question, was could we change that rank order to minimal.

DR. HEUBI: Actually, I just was suggesting that. I think it is more from the standpoint of relative to the question of the mother of the healthy term infant who calls and says is it safe to feed my baby powdered formula, and I guess that is sort of where I would come down to say is this risk material enough that we should be presenting it as a significant risk.

DR. STALLINGS: Are you saying like on the label, because this is a discussion that is going on to help the agency continue discussions with industry versus on a label, and if we were talking a label, my personal opinion is the labels need to clearly state that these aren't sterile products, but I wouldn't put anything on a powdered formula intended for use for normal infants that would discuss these risks.

I thought we were talking about educating primarily health providers for the high risk group, so I don't want to cause undue alarm either.

DR. BUSTA: Dr. Tarr.

DR. TARR: I am uneasy as a health professional stating what the public needs to know and wants to know. I am differentiating between those two. I would have no qualms about feeding my children powdered reconstituted formula being able to put in perspective all the possible risks that are out there.

Yes, if we have to label food as being irradiated, which is still presumably quite safe, I would say that irradiated food poses considerably less risk to the public, very low risk compared to the low presence of a potential bacterial pathogen within powdered formula.

If you then ask the public which would you rather know, and in putting it on that scale, I think that they would probably both say that they would like to be informed, but as a health professional, I feel a little bit imperialistic in stating that.

Would this be an appropriate question to ask Dr. Briley as the consumer representative or to open this up to a wider, we need more data about public desire to be informed of this small risk?

DR. HEUBI: Margaret, did you want to comment?

DR. BRILEY: I don't believe I have ever had any data that would tell me exactly how the consumer would respond to this, but knowing the consumers in our part of the world, they make an awful lot of impact in asking a lot more questions than the normal group would be in the boonies, for example.

They are very informed consumers and they go to their physician or their care provider with questions and written down, and I feel like that, as a consumer, I would want to know. I would want to know exactly, and I would find out.

I would call the industry if I did not feel like I had gotten complete answers, and that is not an uncommon thing, that happens in our part of the country.

DR. HEUBI: Patti.

DR. THUREEN: May I ask a question? What will be done with the recommendations that we come up with today, will they be put on food labels, is it just going to be general recommendation?

DR. TAYLOR: If I can just go back to Charge 1, which I think feeds into your question about the recommendations. Obviously, we are asking for scientific input to help us make some decisions, and clearly the industry is here, and there is a role for dialogue there, as well.

I think the important thing to realize is that when there is an advisory given to the agency, we then take it and move it into a regulatory mode, and there is a step in there that considers that.

I think the question in Charge 1 is not what you tell consumers, but can you characterize the public health issue around this. I am sure you all know we get press questions basically saying we have a lot of mothers out there with full-term infants, what does FDA say about issues related to powdered infant formula.

So, I think the short answer is I don't think at all the intent in the charge was to go down the direction of labeling risk on infant formula, so I think that would personally be a very bad idea, regulatorily a very bad idea, and that was not the question on the table.

Perhaps the question is if you are focusing on immunocompromised children, don't forget there is a large population out there that is interested in knowing about full-term healthy.

DR. HEUBI: If there is no additional comment, I guess I will re-read my proposal for Charge 1, and that was that for populations at risk for potential with exposure to powdered formulas for E. sakazakii include pre-term infants who are less than 36 weeks gestation up until they are 4 to 6 weeks post-term, infants of any gestational age in NICUs Levels 2 or 3, and then immunocompromised infants of any age, and for term infants, the risk is considered to be less than it is for pre-term infants, but not zero. I was going to say not non-existent, but that is a double negative, less than pre-term infants.

DR. BUSTA: Yesterday, I said something to the effect of and if the risk exists, it would be less, because we don't know if that risk does exist, we do not have that in a quantitative fashion at all.

Dr. Thureen.

DR. THUREEN: I like that, just what you said.

DR. HEUBI: Me, too, I like that, too.

DR. THUREEN: I think we should add that.

DR. BUSTA: So, for full-term healthy infants, if the risks exist, it would be less.

DR. HEUBI: That is correct.

Can we call for a vote again?

DR. BUSTA: Dr. Tarr.

DR. TARR: "Would be" sounds authoritative, and in the absence of data, I am not sure we can absolutely say would be less.

DR. BUSTA: Maybe or appears to be less.

DR. HEUBI: We will use the Tarr-Busta amendment.

DR. BUSTA: It appears that it would be less. Is that what you are saying? Okay.

DR. HEUBI: We had better quickly vote before we are done here.

DR. BUSTA: As I said, two hours to get to yes, and we all agreed.

DR. HEUBI: Can I call for a show of hands with my proposal with the amendment, the famous Tarr-Busta amendment or Busta-Tarr? Thank you.

DR. BUSTA: We are back on No. 1. I understand that one of our colleagues has a proposed statement--I am sorry Alex's question.

DR. ACHOLONU: Just to become surer of what we are trying to articulate here, as I said before, we have industry talking about this question of complementary final product testing, and I heard the word "lot" used, and what I want to know is from the industry, this complementary final product testing, is it when you have put all the stuff in containers or when you have the very big container and you have taken some sample from there, put it into containers that would be shipped out for the public, if when that is answered, when you take samples to do final testing, if it is a big lot, do you take it from the top, do you take it from the bottom, and is the sample random, or is there any special method you use for taking samples that you test?

DR. BUSTA: Do we have a respondent?

DR. ZINK: I will just give you a brief discourse on finished product testing. Once a powder is manufactured and it comes out of the final, comes through the sifter and it is in its final form, it is, in fact, at that point finished, but, of course, there is opportunity for further contamination until it is sealed in a can.

These terms aren't precisely defined by microbiologists. Certainly, once a product is hermetically sealed in a can, all would agree it is finished product at that point. Some might even refer to testing it before it goes into that can as finished product.

Other terms of art used are line samples and things like that, but the truth is in practice, manufacturers are often doing testing all along the way.

For example, in terms of microbiological testing, if a manufacturer is going to store this in large sacks or totes or drums for a period of time, they likely will do some testing on it at that point, again to see if there is a problem, then, they avoid the added cost of putting it in can and then finding out there is a problem.

Also, another reason for doing that is if they test it and find that it's okay, and then they go ahead and fill it into cans and test it again, and this time find a problem, that helps them narrow it down where such a problem might have occurred, but all manufacturers test finished product after it's in the can.

Sampling schemes for this vary, but generally, they will fall into two categories. I think most manufacturers pull cans from the can-filling line at predetermined intervals. In other words, they might say to their staff to pull one can every 15 minutes or pull two cans every hour, some scheme like that.

Those cans go into the laboratory. They are hermetically sealed, labeled cans usually. They go into the laboratory, and then the laboratory will take those cans into some kind of a biological cabinet or contamination-controlled area, and they will aseptically open those cans, they will take their samples for chemical and microbiological testing, and perform those tests.

Now, those cans, at that point they are destroyed, they don't go back to the line, and any remaining powder is destroyed. It goes to animal feed or something like that, so this is destructive testing. It is not unusual for a manufacturer to pull, oh, my, more than 100 cans perhaps from a lot that they are testing.

Now, you asked about the definition of lot. This varies somewhat, but generally, a lot of product is an amount of product that is produced under substantially identical conditions, and this often means a calendar day's production.

A lot may be composed of many batches, and a batch is something truly is a homogeneous blending unit, if you will, and very often in the coding on product cans, they actually identify the batches that go into a specific lot, and all of this is traceable back to specific lots of raw materials, if you will.

Does that answer your question?

DR. ACHOLONU: If you happen to find some contamination, what do you do then?

DR. ZINK: Well, it depends. Let's take a situation where you are a manufacturer using a dry blended situation, in other words, you have produced some of the product, maybe the fat and protein component, you have wet blended it and spray dried it, and now you are going to blend it with a carbohydrate function, carbohydrate, such as maltodextrin, and you have tested that maltodextrin, and you have audited the supplier of that maltodextrin, you have visited their plant, you have a great deal of confidence in them, and everything looks fine. You have tested your protein and your fat component before you blend them together, and you blend them all together and now, all of a sudden you have found, let's say, E. coli in the finished product. Where did it come from?

Well, I think most manufacturers would put that product on hold and then they would launch an investigation, and that investigation would probably go back and look at much larger sample sizes of all the components.

It would probably involve some environmental testing of the equipment that touched those since then. What they would be trying to do is find the smoking gun, if you will, and figure out what caused that to happen.

The fate of the product could very well be sealed if it exceeds a regulatory or a manufacturer's internal requirements, they likely would destroy it, but nonetheless, the investigation would proceed.

DR. ACHOLONU: Thank you.

DR. BUSTA: You have a proposed statement for 2, Section 1?

DR. FISCHER: I am working on it.

DR. BUSTA: Do you want to try it?

DR. FISCHER: It's a general statement, so hopefully, that will do the trick.

It would start out by saying something like this. Industry should continue to improve the manufacturing process with an ultimate goal to produce a powdered formula product containing essentially no pathogenic organisms.

So, what we are doing is setting an ultimate goal, striving for doing better all the time, not to say that it has to happen, but we ought to be going in that direction, and I think probably from the sound of it, they are.

As part of the manufacturing process, a product monitoring scheme that accurately detects for each lot the quantitative level of offending pathogenic organisms, including E. sakazakii.

This information would be provided to the agency prior to release of the product. So, what I am saying is that for each lot, they would have an idea of the number of colony-forming units per 100 grams of the product, which would be provided to the agency.

Now, I am thinking about E. sakazakii, but it might be other organisms if that comes up, that is a problem. Further on, the industry has already suggested a level which they consider a safe level, if you will, and we could talk about that when we get to that part, No. 3, I think.

Then, there are these other things, other parts of the process that can be improved, have already been discussed, that is, monitoring of the environment and improvement of the environment, and so on. We could put language in there for those kinds of things, as well, or we could leave it out, I don't know.

But basically, what I am trying to do is saying that the goal should be to produce the best product possible, which I know they want to do, and that for the purposes of regulation, if you will, that they provide a quantitative level of the bacterial content of selected organisms to the agency, and the agency, of course, will have some number in mind that is acceptable or not acceptable.

DR. BUSTA: Other comments? Dr. Thureen.

DR. THUREEN: I agree with the first statement, but I think the very first sentence you said, that the goal would be to strive to set a process to having no contamination, we may or may not get there, but that would be the goal that you would work towards.

The whole second part, I don't think we should go into any specific numbers, and also make a general statement that the means to get to this point would be commonly decided upon by industry and the FDA with experts that the FDA would choose to involve in manufacturing processes to set up a plan that would be general.

It could apply to all the members of industry that they would use, and just leave it very general, in that order, that FDA and industry would come up with a plan. We would not make any specific recommendations because we don't have the expertise on this committee.

DR. BUSTA: Other comments? Dr. Fischer.

DR. FISCHER: You mean not provide them with a number?

DR. THUREEN: No, industry, I mean industry and the FDA and the experts that the FDA have at hand, like Dr. Tompkin, would help come up with a reasonable plan, and not leave it up to this committee, who still have a lot of questions about the whole process, but really get several experts in to help with that plan.

DR. FISCHER: So, what would be bad about saying that the level of bacterial content or pathogenic organisms in a particular lot ought to be determined as part of the plan?

DR. THUREEN: Right. Well, I presume that that is what they put in their plan, that they would determine levels, monitors, that there would be ongoing assessments of is this an acceptable level, that it's a moving target, a moving plan, and part of that would be included in it, but I think that is sort of a given with everything that we have heard, that that would be part of the plan, I presume it would be a given, that that information would be included in whatever plan combined industry and the agency came up with.

DR. BUSTA: To my knowledge, if one detects a pathogen in a product like this, it is considered adulterated and unacceptable for sale.

Am I correct on that? So, consequently, quantitatively, all you can do is say less than 1 in a given amount of product, because that is all you can measure. So, I am having trouble with saying report the numbers of pathogens when I don't think if you have got pathogens there, the product doesn't move.

Dr. Kuzminski.

DR. KUZMINSKI: In regard to Dr. Fischer's statement as amended by Dr. Thureen, I would agree with that basically. I think we need to keep the recommendation at a fairly high level, more general level, and trust that the technical capabilities of both the agency and those that they can draw on, and the industry, which are enormous in this case with the companies represented, will come to the technical capability and plan that we are talking about here.

I think also the recommendation, the verbiage, recognizes that, in my own personal opinion, recognizes that industry is on the right track here, but I think also, again my own opinion, that there needs to be some, because of the seriousness, may be a rare event, but when it happens, it is really serious in terms of fatality.

There needs to be some accountability, and I put it earlier in terms of transparency, collaborative with the agency, as to what the continuous improvement program is within the industry.

DR. BUSTA: Dr. Blumberg.

DR. BLUMBERG: I was wondering if there could be clarification about this issue about contamination because clearly, this product is not sterile, and it is just a level of detection you are getting into.

I was wondering in the statement that if you detected an organism that had to be--

DR. BUSTA: Pathogen.

DR. BLUMBERG: A pathogen.

DR. BUSTA: If you detect a pathogen, it is adulterated.

DR. BLUMBERG: Clearly, this is not sterile, there is pathogens there.

DR. BUSTA: Just saprophytic organisms, they are not pathogens.

DR. BLUMBERG: I guess in the right host, it's a pathogen.

DR. BUSTA: Dr. Neill.

DR. NEILL: I think we have a lot of confusion on the medical side of the table, and probably over there, because what is a pathogen for one person may not be a pathogen for another, but they are both a microbe.

Don, can you help us with what is the current assessment of the agency's--maybe it is not correct to say the agency's--what we are not asking is what is current practice, I don't think, on the part of the industry, but if a microbe is present in powdered formula on microbiological testing done for whatever purpose, process control, et cetera, but it is in product, what is the current meaning and significance of this?

DR. BUSTA: Do we have anybody in conformance here?

DR. ZINK: I just love these questions. I told you that your capacity to cause mayhem was increased in this job. We used to have a term in clinical microbiology called "frank" pathogen, and that was an organism that everybody agreed was likely to cause illness in man, things like Salmonella, Shigella. The list of frank pathogens only gets longer, it never gets shorter. I don't recall anyone ever saying that, okay, we were wrong, this Salmonella isn't a pathogen anymore. That has never happened.

So, we are continually faced with new organisms that may not be all that new, they predate us, but we find organisms that seem to have a potentiality to cause disease that was not previously recognized.

This is, in fact, the situation we are in with Enterobacter sakazakii. In this specific case within the agency, a Health Hazard Evaluation Board met and determined, in fact, that this organism constitutes a health hazard.

Does that mean it constitutes a health hazard to every individual? No, but it means that it has risen to a level where the agency feels that its presence in a food could constitute a hazard that is going to affect at least some portion of the intended population and that they need protection from that.

That is not to say that everyone who might consume this product will become ill. That doesn't even happen with the frank pathogens like Salmonella and Shigella, but it becomes difficult.

You could carry this argument on out and cite cases of almost every member of the Enterobacteriaceae has caused illness in some host or another, in some reported time, and I think it becomes a difficult judgment at some point for a group of experts to say this organism has now risen to the level of what we would call a pathogen.

I am sorry if that sounds a bit nebulous, but the topic is a bit nebulous.

DR. STALLINGS: In follow-up to that, did you really answer the question, what organisms today, if they show up, would constitute that they are adulterated and the product would be discarded?

DR. ZINK: Certainly, if we find Salmonella in infant formula, I think we are going to consider that to be adulterated. If we find Listeria monocytogenes in there or any other of the more widely recognized pathogens, we would consider that to be adulterated.

Now, put into a difficult situation, we have been taking that view for Enterobacter sakazakii, and I think we are looking to this group to comment on that.

DR. STALLINGS: My sense from the committee over this day and a half is we believe that organism has risen to that level, that if it were there, that if it were there, then, it represents a significant risk.

I guess that is one of the things we need to be sure of, because from that flows the whole concept of what is adequate surveillance from the public protection point of view, which is different from the well-developed surveillance from a manufacturing point of view, and it seems like that is exactly what we are struggling with.

DR. ZINK: We struggle with this, too. If it becomes clear to the scientific community that Enterobacter sakazakii is a pathogen, then finding that in a ready-to-eat food like this, I believe we would regard that as an adulterated product.

DR. STALLINGS: So, just to continue that, I mean if we were here talking about another dried food product that was not going to a group we have defined as high risk, then, this organism wouldn't go there, whereas, you might say Salmonella found in--I don't know if this is true or not--dried powdered milk for general consumption in baking, and all of that, but in infant formula, this organism has risen to the status of a pathogen, and thus, we are looking for a different surveillance, not just a manufacturing surveillance.

DR. BUSTA: Dr. Lee.

DR. LEE: I agree with everything that has been said about food, but I also think that the formulas we are talking about have a special niche, maybe a step above food. Somebody I think the other day said we are halfway between plasma and food because of the way it is being used, and that the neonate really doesn't have the same ways of handling this material as somebody with a normal digestive system would or fully developed digestive system would.

So, I think we are applying a higher standard here. I certainly would include E. sak on that list of deleterious incidental ingredients that would flag an adulterated status if present.

DR. BUSTA: Dr. Tarr.

DR. TARR: I am doing a little bit of Poisson distribution in my head. With 0.6 percent of 100-gram samples have E. sak in it, from The Netherlands, that means 100 percent of each ton has E. sakazakii in it, does it not? When we look at small quantity testing as it goes out the door, does absence of proof mean proof of absence? No.

Then, we have to assume that everything could potentially have E. sakazakii in it and should undergo a sterilizing step before it is given to--rather than relying on recalls, which are high publicity, debatable interventions and to preserve public health, should we not rely--I am not saying that something shouldn't be recalled--but should we not use as the mainstay a reliable sterilizing step prior to consumption rather that use recall as a safety net, as an assurance of safety?

DR. BUSTA: Would someone like to respond to that before I get to Alex? Dr. Kuzminski.

DR. KUZMINSKI: You had a key phrase there, a reliable step prior to consumption, and I think the charge in Item 2 under Charge 2, in terms of recommendations to feeding institutions where the formula gets prepared in a finished step for the infant, that, to me, encompasses the phrase that you used prior to consumption.

DR. TARR: It would seem to be an alternative to reliance on recalls and declarations of adulteration.

DR. KUZMINSKI: I don't think it's an alternative. I think it's a continuous step in the manufacture and consumption of the product, of a dried infant formula.

It's a personal viewpoint, a step that starts at a raw material supplier, and the process continues through the manufacturing and distribution system to the hospital, where it is prepared in a finished form for consumption by the infant. I view that as a continuous process in the life of the product.

DR. TARR: I agree.

DR. KUZMINSKI: It's not preparing it with hot water, and we are going to debate instructions in a while. I don't believe that is an alternative at all to a recall situation as you have described.

DR. BUSTA: We have had a hard time staying on No. 1. We have talked a lot about No. 2, we have talked a lot about No. 3, and we have got one proposed on No. 1, and we keep looping back or downward, so what I would like to do is just hold on the proposed No. 1 and go on to No. 2 and get some of that off of our respective chests and go on. We can't seem to hang in there, and we keep hanging really down on No. 3, as well.

Alex.

DR. ACHOLONU: I was just going to comment on your statement. Microbiologically, a pathogen or pathogenic substance is one that is detrimental to health. There is not a way you can consider it microbiologically. If it is not, we call it a commensal [ph], one that gets into you, but doesn't do any harm. It is either a commensal or it is a pathogenic or parasitological or microbial organism that is detrimental to health.

Having said that, when Dr. Busta talked about the discarding of contaminated specimens or substances, if it is contaminated, you cannot go ahead and distribute it to the public.

We must understand that we are dealing with bacteria here. When you talk about parasites, it takes time for them to complete their life cycle, but when you are dealing with something like bacteria, they multiply in a short time.

You can have one bacterium in a sample, and it will stay there under propitious conditions, under optimal conditions, will multiply and become plated, so I feel that when we have established the fact that E. sak is a pathogenic organism, we have spent time discussing what it does to human beings, it is pathogenic, and if we find one that is contaminated, my feeling is that it should be discarded.

Thank you.

DR. BUSTA: Dr. Fischer.

DR. FISCHER: So, we are on No. 2, right?

DR. BUSTA: Let's give No. 2 a try.

DR. FISCHER: I just want to make sure. I think another intervention is labeling, labeling the product. I noticed that industry recommends that we eliminate the use of powdered products for feeding the at-risk population. I don't think anything is wrong with that, I think that's a good idea.

We also discussed the possibility of labeling the powdered product to indicate that it is not sterile, and I don't think that's a bad idea unless labeling it as such will reduce the care with which the product is prepared.

I mean if you label it that it's unsterile, will people begin to just feel that they don't have to take extra precautions in preparing the product for use. I don't know whether that would happen or not, other people may know, but I think that it wouldn't be a bad idea to label it as a non-sterile product unless there is a real down side. So, that's the labeling issue.

DR. BUSTA: Dr. Heubi.

DR. HEUBI: If we make that recommendation, we should also recommend that there be some caveat saying that every effort is being made from a manufacturer's standpoint to reduce the amount of bacteria in the product.

If we make the statement that it is non-sterile, then, we should make some comment also, so the public doesn't say, oh, this is full of bad stuff, that the comment is made that every effort is made by manufacturers to reduce the presence of bacteria in the product.

DR. BUSTA: I have a question for those of you that work with a product like this. If it is stated on there that the manufacturer is making every effort to minimize the microorganisms in the product, but it is not sterile, would that mean something to the handlers, would that register, would it be like the label on liquor or cigarettes?

DR. MOYER-MILEUR: Cigarettes, yes.

When you say handlers, a lot of times these are technicians that have been trained in formula preparation, so the fact that there is a label warning on there really may not make much difference to them, and a physician may never see that label warning, so I don't think that saying this product is not sterile, that's fine to put there, but the however, we are doing everything we can to make it more sterile, I don't think would be helpful.

My other comment is that the thing to realize, that we can train technicians to have good practice and good skill in making these preparations, but I think you need to go further up the food chain to the person who is ordering the product, and they are the ones who require the education piece.

I think that if you work within a nursery situation, you realize that there is a lot of independence in the decisionmaking on what infants are fed, and that if there is not--and at times even within specific nurseries, not agreement on how best to feed these infants, so I think, and I would put this to the industry to help with the education of the healthcare providers, that when they are making these decisions, they need to make thoughtful decisions and realize that when they are opting to use a powdered product, that it is not sterile and that they are opting to use a product that may not be in the best interests of a high-risk infant, and that there are safer products available for use. But I really think that the education piece needs to come at a much higher level than the technician preparing the product.

DR. BUSTA: Dr. Stallings.

DR. STALLINGS: Commenting on that and what Dr. Lee said, I think the reason that you would change the label is because you are going to change institutional policy. It is going to have nothing to do with the resident ordering the formula, it is going to have nothing to do with the technician.

If we create a label and a set of knowledge and working forward that says this shouldn't be done for these reasons except under extraordinary circumstances, then, in essence, Hospital Nutrition Committees and hospital lawyers, and if we can get to that level, and say this is what we need to be doing now, then, we will get compliance, because the policy and procedure that the technician is going to inherit--and I agree with you, I think the facilities are good and the people are well trained, and they are trying to do a good job--that is what we are really talking about.

If we are going to change practice, we are changing practice at the level, and this kind of product is right between a food and a drug, and in this setting, it is much more like a drug than it is a food, we don't have any trouble in this environment treating this just the way we would in many ways an I.V. product.

I mean these are very high risk infants, this is an essential compound for them to get to survive, it is what gets them off of I.V. feeding and all of those inherent risks, but we just need to recognize this isn't about technicians.

If we put things on the label and we educate the nurseries, it will be a policy decision and the trickle-down will be that residents may or may not learn at this point in their careers that this may not--we are still trying to figure out this organism--but they will learn the process of prescribing the least risky product they can to take care of the baby.

That is why the labeling is important. If I try to get my committee and my lawyers to do something, this is how we change it from it's my personal opinion as a pediatrician taking care of babies in a nursery, to this isn't in the best interests of children as a group or the hospital, and we are working with industry to find the right place. So, I think the label stuff is very important. I don't know what the language is, but that's the only way we are going to change care.

DR. BUSTA: Dr. Fuller.

DR. FULLER: Thanks. I think I would sort of try to throw a couple things out. First of all, we are talking about an educational or an outreach to various user groups, specialty groups, the neonatologists, the NICU nurses, and whatever other specialty populations, so we have got that piece, and along with that piece would go the things that might include how often you change the tube, you know, the hang time of the tubes, and all that great stuff.

So, you have got that piece. I think that another potential intervention strategy we might want to ask ourselves, and that is looking at the current label. There are two pieces I would look at on that, and one of those is the instructions for preparation, are they adequate as they are now, can they be more uniform, do we want to make any enhancements or changes or recommend anything there.

Then, a correction I think to what Larry said earlier, I don't believe industry said eliminate the use, I think they said minimize the use in the at-risk populations, so a piece of that educational would be to identify that these products should be used only when there is no other product available yada-yada, and then the last thing is whether or not to add the piece on the label, is there benefit to saying that the product is not sterile, or is that better addressed through this educational outreach, does putting that on the product raise a red flag to the consumer and say, ooh, I shouldn't use this.

I would say if we are going to--well, I will stop here.

DR. BUSTA: Dr. Baker.

DR. BAKER: I would say about Question 2, all of these things are really covered in the recommendations that were put out by the CDC and the FDA, and now the new revised ADA guidelines for institutional use of non-human milk. I think the ADA included human milk, so all of these things are already set out. I think all we need to do is reinforce them and say that we go along with those guidelines.

That does not include the labeling, and I do think the labeling should include a statement that this is a non-sterile product if no other reason than just for transparency, but also for education, and that is the other piece of the recommendations. Somehow we have got to get to the people who are actually ordering these formulas and reinforce what has already been said, that you ought to find alternatives whenever possible to let them know the risks.

DR. BUSTA: Dr. Tompkin.

DR. TOMPKIN: There are some issues with regard to labeling, whether it is labeling the intact container or labeling the carton or the inserts that may be placed in a carton, and there are different approaches to that.

But what we are talking about, I believe, is the labeling or communicating information to those who are healthcare givers of the higher risk population. The CDC did, of course, as you just mentioned, provide some guidance, but it does not state in here that--this is the Morbidity and Mortality Weekly Report statement anyway--it does not say anything that this product is not sterile, and, in fact, it communicates a confusing message, "Trained personnel should prepare powdered formula under aseptic technique," which assumes you are dealing with a sterile product, and that is not the case, so I think the message is not clear, and it should be improved upon.

So, this was, in fact, the summary interim recommendations, and they should be further enhanced based on what is occurring here.

DR. BUSTA: Dr. Kuzminski.

DR. KUZMINSKI: I think if we are on Charge 2, Question 2 on the page, I am addressing that one. I think there are two areas that are important in my view for other intervention strategies.

One is on preparation instructions on the label to encompass the most current information on temperature of water, on mix, on holding time, on cooling it down to a certain time, hang time, whatever. There is newer information that we have heard today and yesterday. That is perhaps not reflected on the labels of products that are out there in today's marketplace. So, that is one area.

The other area for an intervention strategy, I believe builds upon a couple of recommendations that we have had in the material given to us. One is the elimination of powdered formula to at-risk populations.

Industry has proposed a definition of what they think at-risk is. It is different from what this committee thinks, and I think this committee could build upon the industry recommendation of eliminating powdered formula and fold in its definition of at-risk.

I think where the fuzzy area is, is in terms of--and I have learned more about pediatric nutrition in the last two meetings of this committee than I think I had ever dreamed of knowing--but where the fuzzy area in my layman's way of looking at it is in healthy full-term infants as to yes, there is a risk, it's lower, but do we restrict that large population from use of powdered infant formula.

DR. BUSTA: Dr. Thureen.

DR. THUREEN: Addressing that question, I don't think it's practical or reasonable to eliminate from the healthy term population from a cost standpoint, that much of the population that use powdered milk could not afford a more expensive product, and I think we would be jeopardizing healthy infants by taking that off the market.

DR. BUSTA: Dr. Stallings.

DR. STALLINGS: Just as a point of clarification because we found that the three of us were disagreeing, the response to the charge that was mailed out before we came, it actually used the term eliminate the use of powdered products in response to this charge to Question 2.

When Dr. Merritt presented yesterday, and the slides we got this morning, it uses "minimize," so just for the future, we need industry to know, there may be a response if the Chair would recognize, so that I think would be helpful because sometimes these documents do come back to haunt us, and we want to be sure we are on the same page.

DR. BUSTA: Yes.

DR. GHILARDI: Ron Ghilardi [ph] with IFC.

The wording is a continuation of the same sentence, and I think that may be the confusion. In the Response to the Charge, industry identified as a Charge 2 response, "Eliminate the use of powdered products for feeding the at-risk population (hospitalized, premature, low birth weight, and immunocompromised individuals to the extent possible, as FDA already recommended)."

So, in other words, it is either eliminate to the extent possible or minimize, and the clarification relates to that population, and I think we were very close to what the committee was identifying as the at-risk population.

I think you further specified it when we referred to both low birth weight and immunocompromised, I think you further identified what that means, but I didn't see any difference in that.

DR. STALLINGS: So, for follow-up then, do you see that this language is any different from the current practice, because I saw this as a move to be more firm in our commitment to get powdered product, and that is where the issue of eliminate used in this very narrow setting versus minimize, but I see it in the context.

But is industry proposing anything different from what it was from the interim proposals from CDC and others that we have been working with the last year or so?

DR. GHILARDI: I think we are emphasizing that that has a need for further communication. I don't think that that is in practice everywhere, and we are saying that that should be something that is in practice everywhere.

DR. BUSTA: Other comments on No. 2? Dr. Baker.

DR. BAKER: I just wanted to answer that comment about aseptic technique. Aseptic technique, to me, does not mean that the product you are dealing with is sterile, it means that you are dealing with it in a certain way, but it doesn't mean necessarily that it is sterile, and I don't think that was the intent of the statement, that they were claiming that it was a sterile product.

Do you want to comment on that?

DR. BUSTA: Dr. Tompkin.

DR. TOMPKIN: Well, as a microbiologist, I wouldn't imagine using aseptic technique for something that is not sterile. It just doesn't seem like something that I would do. I would use caution, care, but I wouldn't take it into a hood or do anything like that.

DR. BAKER: The ADA, the new ADA recommendations are going to be aseptic technique, I believe, and they do recommend a hood when possible.

DR. BUSTA: Dr. Beuchat.

DR. BEUCHAT: To that issue, it means perhaps obviously, two different things. It might mean handle the product in a way that you don't contaminate it, that is, handle it aseptically, but it may also mean to the reader, a microbiologist, that it is already sterile and you need to keep it that way.

DR. BUSTA: I think this has been a very interesting meeting.

[Laughter.]

DR. BUSTA: The only time I have had experience like this as a food microbiologist is working with engineers. They are sort of off the other end.

It is really astonishing that identical terms conjure up different concepts in practitioners' minds depending upon what your background is and where you are practicing it, and that is a perfect example of some of the experiences we have had.

I think that we have worked on No. 2 quite well and actually stayed on it, which was better than we did on No. 1, and No. 3, I think is of greater depth and we can handle without some additional energy.

I am recommending--oh, my Co-Chair says do you want to get consensus on No. 2 before we move on.

Dr. Kuzminski.

DR. KUZMINSKI: I thought we agreed that our Chair would synthesize as the discussion went along, and I would be happy to repeat what I repeated, but I will take a shot at it if I can recall with all the other discussion going on.

There are two opportunities or two areas of opportunity for intervention strategies. One area is in the area of labeling and labeling instruction to reflect current knowledge on precautions in preparation, the need for certain water temperature, the need for certain holding times, cooling temperature, et cetera.

But the point there is that the label should reflect current knowledge.

The second area of intervention strategy is to build upon two pieces of information that we, as a committee, one we have generated, and the other, the industry has brought forward: one, to eliminate powdered infant formula from populations at risk, and that definition of populations at risk would be the one that this committee has generated with the precaution as raised on the healthy term availability of powdered formula to that population, as Dr. Thureen has raised.

DR. BUSTA: May I add that when you say eliminate, "wherever possible?"

DR. KUZMINSKI: Sure.

DR. BUSTA: So that we don't flush that. And may I add to the first aspect, there seemed to be a great deal of emphasis on not only labeling of the product appropriately, but transmitting the information on that preparation procedure to the entire, I think food chain was a very appropriate word, all the way up, educational material all the way up the administrative system all the way throughout the hospital, so that everybody acts appropriately on that information, if that is an appropriate addition to that, because I heard a lot of that.

Dr. Fischer.

DR. FISCHER: Let me see. I want to make sure we get some agreement or not on labeling it as a sterile, as a non-sterile product. Do we want that label on it?

DR. BUSTA: I heard a lot of people say be sure to have non-sterile on that or this product is not sterile, whichever is the appropriate wording, indicate that it is not sterile.

DR. FISCHER: It is being honest to do that.

DR. BUSTA: Yes. It emphasizes that.

Other comments on that summary? Dr. Blumberg.

DR. BLUMBERG: I think it's important to include that term you mentioned, "wherever possible," because it sounds like for breast milk, that is a real issue with fortifier, because right now it doesn't sound like we have any fortifiers that are sterile.

DR. BUSTA: Right, and I think we have to say "wherever possible," and it could be at the end of the phrase or at the beginning of the phrase, but I think it has to be "wherever possible," otherwise you are proposing an activity that may not be able to be carried out.

Other comments? Dr. Beuchat.

DR. BEUCHAT: I apologize to the committee for not being here earlier, but I pick up now on the meaning of the word "eliminate" versus--

DR. BUSTA: It's too late.

DR. BEUCHAT: Eliminate all viable cells through techniques that are available. You can detect dead cells. I think that we probably are thinking in terms of, if it's the word "eliminate," we are talking about viable cells.

DR. BUSTA: It's eliminate powdered infant formula wherever possible, and you can get back to that other theme on Question No. 1.

Dr. Thureen.

DR. THUREEN: I just have a question. Do you think that it would be useful or advisable to put on the label, like the web site, the ADA/FDA web site, so that people could go to reference for any materials that get updated as time goes on, or would you say if you have questions about this, call the manufacturer, or that gets outdated that it is not worth putting on a label?

DR. BUSTA: Personally, I would hesitate to get into the details of the label because I don't know what can be or cannot be, and there is a lot of designation of size, all that.

DR. THUREEN: That's fine.

DR. BUSTA: Any other comments?

DR. KUZMINSKI: Just a comment. I just would be interested to learn what the practitioners feel that the preparers of the formula, from preparers of the bottle of food for the baby from the infant formula mix, powdered infant formula mix, do they now think that the powdered infant formula is a sterile product?

DR. THUREEN: I would say yes, I think there is a huge misconception, and I think that people just don't know, and it's just a presumption that if it is for babies, and it comes in a container, it is going to be given to babies, it's a sterile product, anything that you pop open like that is a sterile product. It is just an assumption. There is no education--I mean they just don't know.

I took a poll and I asked people if it was sterile or not, and the majority of people presumed it was sterile. I mean at all levels, from the technicians to the physicians, they presumed it was sterile, the majority of them. It just never occurred to them to even ask that question.

DR. BUSTA: Dr. Tarr.

DR. TARR: I agree with Dr. Thureen, and even if they would defer on the question of sterility versus non-sterility, if you asked them is there a possibility that there is a germ in there that could harm a child the answer would be "no" almost uniformly, whether or not they understood the technicality of sterilization.

DR. KUZMINSKI: You see, the difficulty I have, and I agree with truth in labeling, the difficulty I have with putting the statement of whatever the wordage turns out to be, this is not a sterile product, is may imply that this is a defective product.

If the wide preconception, as you have described, is that it has been a sterile product, and if that is now on the label, is this something different, has the process changed that makes a product less good than it used to be, because I thought it was sterile, and now they are telling me on the label it is not sterile.

DR. BUSTA: Dr. Tarr.

DR. TARR: What about a more inclusive, but still accurate term, powdered formulas are not subject to sterilizing conditions as a statement. In that way, it doesn't say this product is not sterile inferring that there could be a product out there that is dry and powder and is sterile.

DR. STALLINGS: To continue the discussion, I think we have all committed that this has to have an educational component to it, so we are going to raise a level of awareness that will naturally, you know, well, what is the big deal.

This will have to come across that this class of products has gone through a different understanding that it potentially has risk, that had been there all along, and where it is now telling you that we are clear of them, there is a plan to minimize the risks both industrially and clinical care.

So, I mean this isn't going to be stealth. If we go ahead with this, and the few people that need to know this, which again are a select number of people in a healthcare organization, are going to hear it and going to make the policy changes and educate the people.

There will be an element of concern that is natural for people who think that, well, people who have been doing their very best to understand, but it's the nature of new knowledge.

I mean we just have to capture it within that, but I am sure we are going to cause some concerns, and I am sure if there is anybody in the country whose baby was diagnosed with this unusual bug, you know, they may think about it, but there is no avoiding that just by nature of having this panel and these alerts that have come out over the last couple of years, this is a different topic.

So, I think we have to say, I mean we have to state the facts. I think there will be the perception as Larry was talking about, that something that we did trust, you are going to be uncomfortable about and that our job is to move it quickly back into trust and select products that you need.

DR. BUSTA: Dr. Baker.

DR. BAKER: I feel like one of Larry's problems with this is that it is somehow labeling one product as being inferior. If the statement were something like powdered infant formulas, all of them are not sterile products, powdered infant formulas are not sterile products, then, it wouldn't single out that particular brand as being not sterile, it is just all powdered formulas, and I believe that is what we heard the last few days.

DR. BUSTA: No, am I correct in that we are addressing this specific group of products for this group of at-risk population? This would be not on all powdered infant formula. Are you proposing for all powdered infant formula?

DR. BAKER: I would say all powdered formula.

DR. BUSTA: Okay. Co-Chair, are you going to get us to consensus, so we can go to lunch?

DR. HEUBI: We need to go to lunch, but I tend to now agree with the concept of this issue of putting a label that all powdered formulas are non-sterile, and I think that takes care of my question about manufacturers making every effort to make them as clean as they can be, because that actually defuses this concept that everyone is the same in terms of how they are handled.

DR. BUSTA: Dr. Tarr.

DR. TARR: Maybe time for one more clarification. Pasteurization may be a better term than sterilization because of the differences--let me back up. A pasteurized product is not sterile, so we would have to extend that to all pasteurized products is this is not sterile. So, perhaps we should state this subject has not been subject to pasteurizing conditions using pasteurization as the noun rather than the sterilization.

DR. HEUBI: I am a little concerned about whether that nuance will be lost on the public totally.

DR. TARR: But technically, well, I would like to throw that open for discussion because we are inaccurate. There is a disconnect between saying this product is not sterile, but a pasteurized product is also not sterile, so perhaps it is equal to pasteurization, yet it is not, going up to the level of pasteurization.

DR. HEUBI: In contrast, though, I guess what we are contrasting is with liquid concentrates and rate of feed are all sterile.

DR. TARR: Are they sterile or are they pasteurized?

DR. HEUBI: Commercially sterile.

DR. BUSTA: We have got a wonderful conversation going. Dr. Lee.

DR. LEE: I think we just made a jump, and I want to just point out that we are consciously doing that. Someone said that we are going to label all powder formula as not sterile or containing microorganisms or whatever.

I am not quite so sure that is in the best interests of public information and knowledge or the industry or the science, because if you look at foods, most foods are not sterile, we are not labeling them in such a way.

I really think that this communication has to occur to the at-risk groups and the care providers to that at-risk group, but to globalize it and say all formula has to be labeled in this way is a bit of a leap to me that is not justified at this time.

DR. MOYER-MILEUR: The exception would be that some of these powdered term products are being used as additives to breast milk for pre-term infants in some nurseries, so it makes it very difficult to single out which products would be found.

DR. BUSTA: Dr. Tompkin.

DR. TOMPKIN: Since all this material is coming off one stream in some cases, not in all, but in some cases, it is one product stream, and the question is some of it is being used in a hospital setting.

Inserts into the carton is one way of dealing with that. Providing the information, you can place an insert into it. It is not an uncommon practice even with foods to put an insert into a carton as instructions for preparing the food prior to serving.

That way, you could target the information for the intended user in that specific setting, and you can build as much information into that as possible, and it should, of course, be complementary and in agreement with the American Dietetics Association, and so on, so it is a complete message.

DR. BUSTA: Do you want to get consensus or do you want to ruminate on it?

DR. FULLER: My sense was we were pretty much in agreement on the first two issues, and the issue of label, what should be on the label is the sticking point.

I think we have heard some differences of opinion. We have heard some people state that labeling should include a statement that powdered infant products are not sterile or have not been pasteurized or something along that line, and then we have heard a differing opinion that says no, that information is better placed in an educational program or outreach or perhaps, although in an insert, but I thought that was labeling also, so I am not sure of the fine points there.

One point maybe Ken could explain because I am having a little bit of difficulty. I heard that the liquid products are sterile or commercially sterile or pasteurized. Can somebody help me with the differences between those terms?

DR. LEE: If you take a commercially sterile can of, you know, I don't care, but beef stew, and subject it to high temperatures in a warehouse, it is going to grow, because it does have microorganisms in it that are not pathogenic. Did I get that right, Frank?

DR. BUSTA: Has anybody got a chalk board?

DR. FULLER: I only raise the point because we had the discussion about if we put something on the powdered product as not being sterile or something of that nature, should you also extend that same to the liquid formula, which is also not sterile, but may be commercially sterile, and that was why I was raising that issue.

DR. BUSTA: To my knowledge, I.V. fluids and things of that type in a hospital are sterile. Thermophiles, high temperature organisms, or whatever, theoretically, I mean essentially, they don't exist. They have been superheated to eliminate all of that.

In commercially sterile food products, it indicates that anything at ambient temperature, that can grow at ambient temperature is not in that product, and it could be a variety of ways.

If the product has a lot of sugar or a lot of salt, it is commercially sterile and could have a lot of organisms in it. It is free of pathogens, and it may have viable organisms that can grow at these high temperatures, as Ken indicated, so that in some cases, a food product like that would not be usable if you wanted to do I.V. because there may be saprophytic organisms that could not grow in that product. If you put it into a vein, you might get just an unusual growth that you wouldn't want to do.

Pasteurized is to eliminate generally for something like milk, is to eliminate any pathogens in that milk, and pasteurized is sometimes used merely to reduce the population of spoilage organisms.

DR. FULLER: Okay. My last comment would be if we are going to put anything on the label, let's be accurate with whatever it is in terms of what terminology we use.

DR. BUSTA: So, in synthesis, I heard that when we are asked the question are there other intervention strategies, I heard that it was essential to label in some way, either by insert or directly on the container, depending on the message that you intend to deliver, that the specific handling instructions for these individual products and their exceptional needs for the populations that are at risk, I heard that this included the handling practices, as well as the instructions on the product not being sterile.

The insert or additional educational materials are required and an additional education program is required for the staff all the way through to the upper administration and the facilities to understand what these labeling systems mean and what the important parts of the preparation procedures are, so that they are carried out appropriately, not just by the individual workers, but being instigated and enforced by the entire administration.

Dr. Kuzminski.

DR. KUZMINSKI: I think you have captured the first area of intervention.

DR. BUSTA: No. 1? Okay.

Dr. Blumberg.

DR. BLUMBERG: I know we don't want to get into the exact wording, but that is probably important because if it's said it is not pasteurized, I think people are going to think, well, this is like--

DR. BUSTA: It is not sterilized.

DR. BLUMBERG: Right, but if somebody used that word, I think there could be a lot of confusion depending on how the wording ends up, so I am not exactly sure what we are agreeing to.

DR. BUSTA: I was hoping to leave it loose enough, so that the people who are in communications and understand what the public understands would select whether this product is not sterilized, this is not sterile, this product contains bacteria, whatever words would appropriately get the message across to the clients would be the term.

I am not sure I am in a situation or we are in a situation where we know the best way to send that message that this is not a sterile product.

Dr. Stallings.

DR. STALLINGS: I would agree with that, because I think there really is an art to deciding on label and educational materials, and I think it is probably beyond both the time and the talents that we have.

So, I think the big question for the committee is have we captured the idea, and the agency can work with communications and space and proper place with that experience.

DR. BUSTA: Dr. Tompkin.

DR. TOMPKIN: This communication, it's for those healthcare givers who are preparing and handling these products in high-risk populations, this is not for the general population at large. I want to make sure that we understand who is being communicated the concept of non-sterility.

We are really communicating this to a special health group.

DR. BUSTA: And that is what I heard, that it would be accompanied with educational materials, so that one does not get a backlash from the handlers or the workers or the medical staff, or whoever.

Not unlike the aseptic technique, if aseptic technique meant don't contaminate it, and you remove that terminology, some people might get quite cavalier. So, I am assuming this comes with an educational program to make sure that the staff all the way through realizes that care needs to be at a higher level than it was.

Dr. Blumberg.

DR. BLUMBERG: Just to clarify, are you saying to label all products or just products that are going to be used in the neonatal intensive care unit?

DR. BUSTA: My understanding, the last that I heard is for products for the at-risk populations, so if it was infant formula, straight infant formula going to hospitals, that it would have inserts or labels, but for anybody dealing with an at-risk population.

DR. HEUBI: I think the problem is, is that you can't separate out products that just go to the hospital versus those that go to the public because we know, as Laurie pointed out, that there is supplementation of products with powder that is intended for term infants, that is given to pre-term infants.

So, it sort of like we are going to have to either go all the way or not is sort of what it boils down to, and I am now thinking in my own mind, after I have heard all this conversation, about whether we should just ignore this and not include anything about sterility at all, because I think it has now gotten to the point where I now know facts that I didn't know before.

In fact, Ken has pointed out that even products that are liquid products are not sterile, which then are you going to put something on them saying that they are not sterile?

So, then, I get into the situation where maybe we have opened a Pandora's Box here that I don't want to let the little genies and folks run out of because you are going to be dealing with cleaning this up for years after we make recommendations about this.

So, in part, although I thought this was a great idea to begin with, I am willing to admit that my mind has been changed somewhat. I am not sure that I think it is now appropriate to include something about sterility.

DR. BUSTA: Dr. Fuller.

DR. FULLER: I agree with you. I think that I agree with what we had come to with that one piece of label change. I think that the issues that Dr. Kuzminski first stated about using what we know about preparation of the product that we have learned over time, that the current label instructions might be enhanced.

I don't have any problems with changing those labels. We have to keep in mind the product is going to more than one user group, and the product has to be an appropriate label for all.

I think the stress has to be on the outreach and education to the medical community that are involved with the use under these very specific conditions where the risks are, and I don't think there is a problem with inserts that address those risks and those instructions. I think that is fine, but I am not convinced that putting something about sterile or non-sterile is where we need to be.

DR. FISCHER: Dr. Stallings.

DR. STALLINGS: Not to belabor the point, but I think one of the things that I have learned throughout this is something from a healthcare point of view about something I perceived as sterile, and, you know, the definition of commercially sterile versus naively thought to be sterile is new understanding.

But the formula that is liquid, we think of in a healthcare setting as sterile, and if that were to be punctured we wouldn't use it, if something inappropriately were to be put in it, we wouldn't let somebody stir it or mix it without it being a sterile instrument, or the blenders are cleaned, and all of that.

So, I know we are dealing a little bit with semantics, but if we go back to our job, it is really to protect the public--in this case the public are these little, tiny babies--what is it that we can do to really meet that.

I think I do believe that without it being in big red letters and alarming, the fact that these aren't sterile is an important new fact. Now, could we get that same information across if we had an educational program? Probably. And then the question is can we sustain that change in practice or we sustain it long enough until the product is changed, and all of this really goes away.

But the only reason I like the labels is it is part of how we learn when you are unfamiliar as a healthcare provider with something, part of how you learn about it is you glance at the label and it puts it in a place.

So, I am not looking to alarm people, but I agree that all of the powdered products have a real potential to be used in the nursery, at least for the foreseeable future, so I sort of favor whatever the most straightforward bland, you know, non-sterile product is something that I am accustomed to seeing with everything I use in an intensive care unit, it is either sterile and I put on gloves, and it goes into that range, or it is not sterile, and I am still careful.

So, I am still voting for non-sterile statement.

DR. BUSTA: Well, we had a synthesis, but it dissolved on sterile versus non-sterile, or non-sterile versus no indication of non-sterile. That is what I think we had.

Some people want to state this is a non-sterile product in some fashion. Others say it probably shouldn't, that may cause more problems than not putting the information on the product.

Dr. Tarr.

DR. TARR: I would like to ask a question. Are the liquid formulas pasteurized, do they meet the definition?

DR. BUSTA: Oh, absolutely.

DR. TARR: Then, why don't we state that the difference is these products or this group of products are not pasteurized.

DR. BUSTA: They are pasteurized.

DR. TARR: But the powder.

DR. BUSTA: The powder is pasteurized.

DR. TARR: That counts as pasteurization? If something is pasteurized, but then is subject to contamination or association with something that is not pasteurized, does that render it non-pasteurized?

DR. BUSTA: No, otherwise you wouldn't have pasteurized milk. Milk is pasteurized, then packaged.

DR. FULLER: Getting to that point, are they currently labeled as pasteurized, the powdered? No, I am talking about the powdered product, is that labeled as pasteurized? I didn't think so. And the liquid is labeled as sterile. I am seeing nodding heads out there.

DR. BUSTA: It is not labeled.

DR. FULLER: Okay.

DR. BUSTA: Those words are not used, to my knowledge.

DR. FULLER: So, we have got neither of these products currently have any labeling to address the issue of sterility or non-sterility, and we are suggesting that we identify the powdered as being non-sterile, and we just ignore or leave alone the liquid product, and let it go with perception.

I want to do the right thing here, believe me, I want to do the right thing, and I am just wondering if labeling is the way to do it.

DR. BUSTA: I am going to hold this to no more than five minutes, because as a good food microbiologist we have product incubating next door.

[Laughter.]

DR. BUSTA: At low concentrates, the dose makes the poison, as you all know.

Dr. Tarr.

DR. TARR: May I ask you to consider asking us, in a two-tier question, should there be some public declaration on the product that there is a categorical microbial difference between powdered versus liquid substance, and then, secondly, if the vote is yes, that we then defer to the FDA to work on the appropriate wording?

DR. BUSTA: All right. That is a question you are asking me. Let's see, I am Chair, I am Co-Chair.

DR. LEE: I will second that motion that we basically introduce the concept that there ought to be a communication that the liquid is different, perhaps cleaner, all right, and leave it to the healthcare professionals how to communicate that in whatever words are necessary.

DR. BUSTA: He just made the point that we may not be able to choose the right words.

Dr. Neill.

DR. NEILL: A point of clarification perhaps from one of our FDA colleagues in the audience. In full recognition of the fact that the road to hell is usually paved with those good intentions, are there different recognitions of the definition of the term pasteurization as applied to a food label versus the legal and regulatory interpretation of the term pasteurization.

I ask in the concept that it was my understanding from a long time ago that the only legal definition of pasteurization was for a particular time and temperature combination as applied to fluidized milk.

DR. ZINK: You know, it just gets better. Within the processed milk ordinance, there is a definition for pasteurized. I don't profess to be an expert in food law here, but I am not aware of any other definition of pasteurization anywhere.

Now, certainly the agency is confronted under the new farm bill of considering more broadly other processes, and that has to be done kind of on a case-by-case basis. You know, there is a process where we can be petitioned to decide whether or not something is pasteurized by the process, but right now the processed milk ordinance is the only one that defines the term pasteurized.

DR. BUSTA: And that is by a state-by-state decision, right?

DR. ZINK: Well, the states enforce it, but the ordinance defines it.

DR. BUSTA: Dr. Tompkin raised his hand, Dr. Beuchat has the phone, so whichever one.

DR. TOMPKIN: We might be about to say the same thing, but in the case of citrus juices, if they have not been subjected to a 5D reduction process, they must be labeled non-pasteurized, so as you go through, you can come up with some other examples.

DR. BUSTA: I believe there is a pasteurized for eggs.

DR. ZINK: There is a process for eggs, for milk or dairy products. Any other product, we would have to define it.

Dr. Beuchat.

DR. BEUCHAT: Just to add to that, to communicate that there is confusion outside this committee or at least concern about the definition of pasteurization, the National Advisory Committee for Microbiological Criteria for Foods has been charged by subcommittees with an issue entitled Redefining Pasteurization. That is a presently ongoing activity.

DR. BUSTA: We have 30 seconds. Dr. Tarr, would you like to restate what you said?

DR. TARR: Yes. We should vote yes or no. Are we in favor of stating that the powdered formulas are categorically different from a microbiologic standpoint than the liquid formulas?

DR. BUSTA: All who want to vote--

DR. TARR: And publicizing this.

DR. BUSTA: Those who want to say yes to that statement, raise your hand.

DR. TARR: The proposal, that there be a statement on or in the package to the effect of this powdered substance is categorically different from a microbiologic standpoint than the liquid substance.

DR. BUSTA: The first time was even better, but what you are indicating is that, with the appropriate wording, it is to indicate to people that powdered product has a different treatment that makes it less microbiologically hazardous, or whatever, than the liquid formula.

DR. TARR: And hazardous is a buzz word.

DR. BUSTA: Dr. Lee.

DR. LEE: I like it.

DR. TARR: Different rather than hazardous.

DR. BUSTA: All those in agreement with his statement, who want to say yes to his statement, raise your hand.

[Show of hands.]

DR. BUSTA: Eight.

All who want to say no, raise your hand.

[Show of hands.]

DR. BUSTA: Five.

Lunch is served. I was six.

[Whereupon, at 12:30 p.m., the proceedings were recessed, to be resumed at 1:15 p.m.]

AFTERNOON PROCEEDINGS

[1:15 p.m.]

DR. BUSTA: We have one hour and a half, and we are going to try to do our very best.

In No. 2, I am going to give an option and ask for your indulgence in listening to this option which I have heard comments about one way or the other, and see if you can jump at that, and if not, we may just have to put it onto a shelf and hold it for a while.

There was an 8-7 vote on that one option, and in the interim, I am learning a lot about what is involved in trying to come up with a labeling issue, and that labeling is another world that we have not addressed here, and if communications between clinicians and food microbiologists use terms differently, labeling is even really off in another sense.

So, I would like to propose the following synthesis on No. 2 - are there other intervention strategies: that we suggest to FDA that they prepare educational documents that can be used in attaching to the appropriate infant formula materials, especially for those for the children at risk, and when used for the specific children at risk, that would alert all individuals whether it be mothers or hospital personnel at any level, to the absence of sterility of those products or the special handling needs of those products, stated in a way that they will understand it and in a way that will deliver the message that it requires special handling without causing a panic or a loss of use of those essential products, that reflects the handling procedures, it reflects all the recent handling procedures that are recommended, and it gives a system that can be updated regularly to any new understanding and information that comes out.

With that, I am not specifically saying that it is labeling, but I am emphasizing that it should indicate in those materials, in educational materials, that these are not sterile products and have to be handled specifically.

Dr. Baker.

DR. BAKER: Just to clarify, this would be addressed to hospital personnel, as well as lay mothers, parents, who are making up formula.

DR. BUSTA: It is people that would use it as indicated where they would be supplementing mother's milk with formula, places in those situations.

DR. BAKER: So, some formula would have this, and some would not?

DR. BUSTA: Or the message would get to individuals that would be using formula for potentially risky children, children at risk.

Dr. Tarr.

DR. TARR: I think that that is close or at where we should be for the healthcare professional. I think it doesn't address the issue of the general public and the purchase of an item that we are now agreeing is not sterile.

Will you have a second proposal for that population or is this one notice fits all?

DR. BUSTA: You have a proposal for one notice fits all? This is big enough for me right now, this section.

Dr. Fuller.

DR. FULLER: I would like to suggest, and everybody shake your head if I have gone in the wrong direction, an alternate. What I would suggest is I think we were in agreement where we were on 2(2) with the exception of the labeling, which we took this vote and we had a very close vote. The committee did not agree on whether the product label should address sterility.

I would suggest that we agree to everything else and on that issue, we just note that we, as a committee, were not able to reach agreement, and recommend that FDA look at that issue and, if appropriate--I mean they are the ones that know what should and shouldn't be on labels and making that decision--obviously, some people feel very strongly that it is worthwhile, and if it is appropriate and appropriate terminology is out there, but to look at that issue and address it as they feel it should be.

I mean the committee has said this is something important, we have noted that to FDA, I think in our deliberations, and I think we just best move on.

DR. BUSTA: Dr. Stallings.

DR. STALLINGS: I think when we were earlier talking about labeling, the idea at the FDA, I mean the food label is such a precise, highly regulated space, that I think the idea was that people using the product would have this information.

I don't know enough, it's like the product insert, packaging, there are a lot of different places that you could put this information, and if that is captured in what everybody is saying, I think we, at least I, had moved a little bit towards the idea that this was something I didn't understand, and if I didn't understand it, a lot of people probably didn't, and if we use the principle of transparency as fact, but not to institute fear.

I mean it is the idea that now that we have gone through this process, it should be made publicly available in some way, labeling may not be the right term to talk about what we are doing.

DR. FULLER: I think that is the point, that's exactly, and I think if we could just agree with that, and move on.

DR. BUSTA: Dr. Blumberg.

DR. BLUMBERG: I would like to go back to the point that you brought up, about the steps that FDA should take as far as education, and maybe split it into two components.

Maybe that is for consensus reasons, and maybe there is other reasons, but one is that FDA should provide outreach to physicians to make them aware that powdered formulas are not sterile, and these are not generally the preferred agents in the neonatal ICU setting. Some kind of very strong statement that would follow up what is in the MMWR, and that be modified perhaps, the guidelines that were in the MMWR.

DR. BUSTA: That would be fine to add onto what I said. Do you want to vote? All in favor of what I said?

DR. FULLER: Yes, with those add-ons.

DR. BUSTA: With those additions.

[Vote taken.]

DR. BUSTA: Thank you. No. 2 is gone.

No. 1, we have two individuals that said they had some items to put forward.

Dr. Neill.

DR. NEILL: I have written what are four separate statements that could be deleted in part or in whole, in which I have attempted to capture the spirit of the scientific content as we discussed the Question 1 under Charge 2.

That question was what intervention strategies could be used in the plants.

Intervention strategies which reduce bacterial presence in powdered infant formula should be used in manufacturing processes and plants. These include, but are not limited to, prerequisite programs to assure the microbial quality of raw materials, hygienic design and maintenance of equipment, hygienic zoning in plant design, and continuous use and improvement of HACCP programs and their verification.

The committee encourages the development of a microbiological sampling and testing program through joint efforts by industry and the FDA, the purpose of which is to assure great clinical safety of this product. It would be highly desirable to formally assess the contribution of such a microbiologic testing program when added to the intervention measures described above.

The committee recognizes that with currently available processing technologies for powdered infant formula, the risk for illness due to E. sakazakii cannot be completely eliminated for the at-risk populations specified above (as in Question 1). Whether the additional interventional strategies described herein can achieve this result is not clear.

Recognizing the important clinical purposes for which powdered infant formula is used among the very populations most at risk for E. sakazakii infections, the committee strongly encourages and enjoins the powdered infant formula manufacturers to develop product formulations which combine the attributes of maximal growth promotion and microbiologic sterility for use in these at-risk populations noted above.

DR. BUSTA: Dr. Fischer.

DR. FISCHER: I like that one.

DR. BUSTA: Okay.

Dr. Fuller.

DR. FULLER: I think we just need that the maximum growth was of the infant, not the E. sak.

DR. BUSTA: Dr. Beuchat.

DR. BEUCHAT: The very last point that you made, the last sentence, microbiologic sterility?

DR. NEILL: I chose the words I chose for their specific meaning and offer them to you, then, for your consideration and debate.

DR. BUSTA: Could I expand on that? I think that you are referring to your earlier comment, or at least I interpreted that earlier comment, that in the right situation, nearly any microorganism can be a problem for this at-risk population that we are dealing with.

DR. NEILL: I would also be making the point of trying to be forward thinkers for the committee, and that E. sakazakii is a currently identified problem, and we really at this point don't know if there are other microbes for which we do not yet have knowledge that they pose a problem and, in fact, have a specific association with infant formula. So, there was a method to the madness in terms of trying to choose the words I did for that purpose.

DR. BUSTA: Dr. Lee.

DR. LEE: On that point, I would agree with the use of the word "microbiological safety" because we may someday indeed find out that there are desirable microorganisms to have in this product.

DR. BUSTA: Dr. Tompkin.

DR. TOMPKIN: I like it, too. I think it was well considered and it is broad enough and yet addresses in a specific manner the real intent. I think it is very well done.

DR. BUSTA: Are we willing to go with microbiological sterility?

DR. NEILL: Yes.

DR. BUSTA: Will you accept the modification to microbiological safety? Okay.

DR. BAKER: Can I just point out one thing?

DR. BUSTA: Yes.

DR. BAKER: Safety is a judgment, sterility is not.

DR. BUSTA: Dr. Kuzminski. Two minutes on this and then we are done.

DR. KUZMINSKI: If we hinge it on microbiological sterility versus safety, I really question the ability of the technology to deliver in a powdered infant formula product at this point in time that kind of product.

I didn't hear a time frame in your statement, which is good, because I think when we were talking about it earlier, we set an ultimate goal of zero content of problematic microorganisms, so I think your statement is a good one, and I like it.

I am a little discomforted by it, I am just trying to put my finger on why I am discomforted by it.

DR. BUSTA: It is late in the afternoon.

I would like to speak for safety in that sterility may not consider toxins that came through the product, that it could be sterile but still have microbial toxins in it, and it would be sterile, but not safe.

So, I would like to--as well as the viable wanted organisms--so, I would like to speak for microbial safety, microbiological safety.

All in favor of that very well constructed set of four comments, raise your hand.

[Show of hands.

DR. BUSTA: Unanimous unless there is somebody against. Thank you.

No. 3. Is it possible, based on available information, to specify allowable lower levels of microbial detection of E. sakazakii in powdered infant formula, and do allowable levels vary by risk characteristics of the infant?

Open for comment. Dr. Baker.

DR. BAKER: I would say no and yes.

DR. BUSTA: And as I read the question, those are contradictory, but outside of that, it's a paradox.

DR. BAKER: Given the present information that we have, there is no way that we can set lower allowable limits. I don't think there is enough information to do that. So, no to the first.

Do the allowable levels, if there are any, do they vary in infants? Well, we have sort of already said that they do, said that there are at-risk groups and not at-risk groups, so we can assume that they would vary group to group.

DR. BUSTA: What I am hearing is that with the information that we have, we really can't make a judgment on No. 1, and there is no question in your mind, as a commission, that No. 2 would vary, because we have identified groups. Okay.

Dr. Beuchat.

DR. BEUCHAT: This may be nit-picking, but I need clarification of the meaning of this question. Could it mean--I don't think it does--are we talking about methodology here, "to specify allowable lower levels of microbial detection," I don't think it means lower levels of sensitivity in terms of detecting the organism, but rather the number of the organism per unit product.

DR. BUSTA: I am seeing head shaking in the back that says this is not related to methodology, but is related to number of organisms in a food product that would be allowable.

Dr. Kuzminski.

DR. KUZMINSKI: I think an underlying point to this question is do we consider E. sak a pathogen, and if the answer to that question is yes, can you allow any level of a pathogen in the product.

DR. BUSTA: One of the greatest tricks is to answer a question with another question, and I don't know.

Dr. Stallings.

DR. STALLINGS: To continue the discussion, I think over the course of time, we have decided it's a pathogen in the high-risk group we are discussing. So, I would agree with the information we have at hand, that there is no way for us to recommend a lower allowable level, and if we can't do that, I don't see that there really is a second. Until we have the kind of information that would allow you to do a real risk analysis, I mean that is sort of where that one ends.

DR. BUSTA: Dr. Tarr.

DR. TARR: As far as I can tell at this point in our knowledge, all we know is it's a genus and a species, we have no idea about subtypes. So, until we gather more information, we have to assume that all are potentially pathogenic. Perhaps we can refine that in the future.

DR. BUSTA: We had some information yesterday, but that was in mice.

DR. TARR: Right now I would have to propose that unless and until further data emerge that there are clearly pathogenic subtypes that could then be subject to a second tier of screening or testing we should be safe with E. sakazakii.

DR. BUSTA: So, you are consistent with the earlier comments that we really are not in a position, we don't have enough information to respond to the first one, and the second one, therefore, is a moot point.

All in favor say aye.

[Ayes.]

DR. BUSTA: Opposed?

I am sorry I cut you off.

DR. BEUCHAT: That's fine. We do have, I think, pathogens in infant formula. Bacillus cereus, for example, isn't a zero tolerance, so it wouldn't be setting precedent. I am not arguing for or in favor of a limit, but I think Bacillus cereus can be present at a certain level, and not be considered adulterant. Am I wrong?

DR. ZINK: I feel better about this one. Oftentimes you hear people talk about zero tolerance. There really isn't any such thing. We test for pathogens using a sampling plan. The sampling plan, if it's well designed, is based on what we feel we need to be able to assure ourselves in order to manage the risk.

I mean when you do microbiological testing in the context of a quality control procedure or a regulatory agency, it is a form of risk management. So, if there is something that is very, very risky, it often has a very stringent sampling plan.

If there is something that has relatively low risk, it has a much less stringent sampling plan, and when someone is talking about zero tolerance, what they really mean is that according to whatever sampling plan you apply, if the result is negative, it's okay; if the result is positive, it's not okay, and it would be adulterated.

There is no way that you can look at a 40,000-pound lot of infant formula and say I am 100 percent certain that there is not a single cell of any given pathogen in there without actually testing all of it for the pathogen thus rendering the status of the lot moot because you have grown it all up.

So, I think, you know, one of the central questions we face is what sort, given our perception of risk here, what sort of sampling plan, if you will, is appropriate for risk management of this type of organism.

There is sampling plans that a regulatory agency might use, there is sampling plans that are manufacturer might use for quality assurance. I mean some of you on this committee know this better than I.

Even a broader question, is microbiological testing for this organism a risk management strategy that you would recommend.

DR. BUSTA: If you find Bacillus cereus in this product, is it considered adulterated?

DR. ZINK: I believe in the 1996 ANPR, we proposed a limit of 1,000 per gram--what is it--of 100. So, anything more than that, we would be likely to consider it adulterated, anything less than that, we would be likely not to.

DR. BUSTA: We voted on No. 3. Now we are on No. 4, right?

PARTICIPANT: We said that we did not have the--

DR. BUSTA: As I recall from two and a half minutes ago, we said that at the present time, we don't have sufficient information to make any kind of decision as to whether there is an allowable lower level of microbial detection or number of microorganisms, E. sakazakii in powdered infant formula.

We don't have information to know if there is some allowable level, and that consequently, do allowable levels vary by risk characteristics of infants, we said, well, being that we identified a risk group that was different from the general infant population, it is probably yes, but it is dependent on the first question and therefore it is moot.

That is what I just called for a voice vote, but if we need to raise hands, that is why I thought how we dispensed with that No. 3. Is that all right? All right.

No. 4. Dr. Blumberg.

DR. BLUMBERG: I think one thing that has come out is we really need better data on rates of E. sakazakii infections, and as has been pointed out, all we really have is numerator data from reported cases in the literature, and obviously, not all cases are reported.

What I would encourage FDA to do, and perhaps the group would agree, is that there are currently available population-based surveillance activities that CDC has called FoodNet, which I think is part of the Emerging Infections Program, and I would encourage that FDA work with CDC for these groups to do population-based surveillance to establish what are the rates, because I think if they did this for a year in the surveillance--they have six or seven surveillance areas, population-based surveillance areas scattered around the country--that they could come up with some data for us on what are the rates of infection with this organism.

DR. BUSTA: That's the first item on No. 4, right, critical knowledge gaps.

I am proposing that we make No. 4 a shopping list, and we just go on, and I calculated that if we had one hour to do this, which is exactly that, we would have approximately four minutes per person.

Dr. Stallings.

DR. STALLINGS: I will do my shopping list and then let everybody else talk.

So, the understanding of the level of occurrence of the organism in the powdered products with acceptable sampling, so this isn't for assurance of the product line, but as a separate study. Some understanding of the dose required for infection in the best animal model that we could get.

A review of current clinical practices for diagnosing an infection with this organism, and I am concerned about two things, one, that there could be partially treated meningitis because of the way we practice in a high-risk infant, you often give antibiotics before you know the organism.

The issue of clinical microbiology labs with the new automated systems, are those fully capable of doing everything that we want them to do.

DR. BUSTA: Dr. Heubi.

DR. HEUBI: I just wanted to amplify what Henry said, that is, we need to make sure that those include the at-risk populations that we are looking at, and also that the microbiological techniques actually allow us to be able to identify E. sakazakii, because that is another issue, as well.

DR. BUSTA: Do you have other ones on the list?

DR. HEUBI: Not immediately. I will think about it.

DR. BUSTA: Next volunteer. Dr. Briley.

DR. BRILEY: I don't believe we have enough data on the breast milk formula with the mother's milk combined in terms of the organism, as well as use, and that is where the consumer might be using it more.

DR. ACHOLONU: Alex Acholonu. In light of we were told that we don't have any knowledge of the reservoir hosts, I am suggesting that we conduct some experimental studies, may use some laboratory animals, and find out the reservoir hosts.

The next thing that I feel we should go into is the question of effective control, what would be the effective control measures.

Thank you.

DR. BUSTA: Dr. Fuller.

DR. FULLER: I will add my two cents. I think additional work on identification of pathogenicity factors, some of the host susceptibility factors that lead to disease, the issue of colonization versus disease I should say.

Then, an area that we talked about earlier, as well, was the potential for use of other microbes for competitive inhibition within the product, as well.

DR. BUSTA: That would include probiotics?

DR. FULLER: Right.

DR. BUSTA: Dr. Tarr.

DR. TARR: I have four elements I would like to propose. First, a needs assessment pertaining to consumer and provider knowledge base regarding microbial safety of these products.

No. 2. Optimal therapy for the infected child. I am concerned when I see brain abscesses and don't quite know the best way to address that.

No. 3. What is the spectrum of disease in the human host, does it include gastrointestinal disease in particular.

I just want to clarify Dr. Briley's comment. What exactly is the safety of breast milk supplements? Is that what you are asking? That seems to be a void in what we have heard in the past couple days.

DR. BUSTA: Dr. Lee.

DR. LEE: I put a short-term and long-term spin on this, just two items. One would be I think in the relatively short term, we can determine whether or not reconstitution with hot water is advantageous or not. That is just something that could be done technologically, so I will defer to the research community whether or not that is something that is advantageous. I think that can be done in a short period of time.

More on the long term is the issue of competing flora and what is the impact of having other bacteria present in inhibition of pathogens.

DR. BUSTA: Dr. Tompkin.

DR. TOMPKIN: With regard to doing a risk assessment, information is needed in terms of exposure, what is the potential exposure in hospital settings, as well as other mother foods, and in the environment. I don't think we have a real good feeling for that.

In a brief survey that was done in The Netherlands, for example, analyzing 16 sweeper bags, vacuum sweeper bags from households, just analyzing a 10-gram sample, they found 5 of those to be positive, so it is present in our home setting, as well as in hospital settings.

The other thing that is necessary for a risk assessment is the infectious dose, has been conducted in Canada, and that should be pursued, so that we can better understand that aspect of this ailment.

DR. BUSTA: Dr. Thureen.

DR. THUREEN: I would suggest that we have more information and actual preparation, things like the rate of colony increase, such as we got from the CDC that is not published, that type of data to help better guide preparation in hospital, cooling time, heating time, minimal cooling and heating times, things like that, so we can have very specific recommendations for preparation.

DR. BUSTA: Dr. Beuchat.

DR. BEUCHAT: Two questions or two areas that need I think research attention, both of which would provide I think information, better allow us to discuss levels, numbers in the formula, one is to determine from the isolates of commercially produced or processed formula, whether or not they are pathogenic or not, what is the level of the virulence, the possibility being that some are avirulent, others are highly virulent.

Secondly, under all ranges and combination of conditions to which the dehydrated formula would be exposed in the hospital or in the home, temperature release times, this kind of information on growth, survival, death behavior, I think would be of value in any discussions that would lead toward a recommendation for levels in the product.

DR. BUSTA: Dr. Fischer.

DR. FISCHER: I would like to add to that, if it hasn't been added before, an investigation into the inactivation of the organism be done particularly I would like to see the irradiation work done to see how effective that might be in others.

DR. BUSTA: Do you mind if a piggyback on that and add pulsed electric field and high pressure processing, and some of those non-destructive, low thermal treatments.

Dr. Acholonu.

DR. ACHOLONU: You are almost right.

I think it may be necessary to go into differences in susceptibility between the pre-term and the term infants, find out if there are some differences in susceptibility.

Next, I did make a reference to the possibility of prenatal infection. It may be good to go into checking mothers to see, since we have neonates being the group that is infected most, to find out if all of the infections come from the environment or from the powdered milk, or if some are coming from the mothers.

We have established the fact that this organism has no respect for age, that it affects both the young and the old. If that is the case, there is a possibility that mothers may have the infection, but it may be asymptomatic in them. So, I am suggesting that we look into the possibility of having prenatal infection.

DR. BUSTA: Dr. Kuzminski.

DR. KUZMINSKI: Just linked to the comments on hyperbaric and pulsed-field radiation, just in more general terms, in terms of process development research, looking at ways especially in the existing technology, process flow that employs spray drying to research methods of inactivation post-spray dryer.

Secondly, I am not that close to the microbiological detection methodology, but just in conversation here, it is not a real-time yes or no answer. It takes time, 72 hours, and that time kind of time frame, perhaps research into detection methodology to give more rapid knowledge of presence or absence.

DR. BUSTA: Dr. Tompkin.

DR. TOMPKIN: I think it would be advantageous to clarify whether or not a positive stool sample does, in fact, reflect colonization.

DR. BUSTA: Dr. Neill.

DR. NEILL: I would be interested to know whether E. sakazakii can exist in a viable but non-culturable state and for which detection methodologies then would need substantial revision.

I would also be interested in the enhanced recovery methods that might be necessary for detecting thermally injured E. sakazakii.

DR. BUSTA: Could I piggyback on that and also look at dehydration injured and other type of injury.

Dr. Beuchat.

DR. BEUCHAT: Asking again for definition of colonization. To me, discovery or detection of the organism in the stool doesn't mean colonization necessarily. In some of the comments in discussion that I have heard, the impression is that to some, it means the mere presence of the organism. To me, it means the actual establishment within the infant of a colony that is not easily removed or sloughed or gotten rid of through normal processes.

So, I am asking for a definition, is colonization that I have described, is it the same way that you would describe it as a person looking at the clinical area.

DR. BUSTA: This is for information, I am assuming.

Dr. Blumberg.

DR. BLUMBERG: I would say if you recover it from the stool of a patient, they are colonized with it. I mean you are not going to be doing invasive procedures to find out if somebody is colonized.

So, if you can recover it from a patient from a clinical sample, and the patient is not infected with the organism, then, the assumption is they are colonized with that organism.

DR. BUSTA: Dr. Tarr.

DR. TARR: I would like to ask how you differentiate innocuous passthrough, if it's in the formula, versus actual replication and some residual staying power that the word colonization would define.

Furthermore, I would even say that there is a chance that it could be not recoverable from the stool, get in the mucous layer of the gastrointestinal tract and colonized.

DR. BLUMBERG: I would agree with that. I mean the absence of the organism there doesn't mean it could not be colonized, it just means you couldn't recover it.

DR. TARR: I am not certain how to handle, though, passthrough in a stool, you know, a temporal wave coming through.

DR. BUSTA: Dr. Tompkin.

DR. TOMPKIN: In order to obtain maximum information in terms of the prevalence of organisms like E. sakazakii and processing environments, there has to be an atmosphere in which industry is encouraged to aggressively detect it.

A finding of a positive for this organism, and others perhaps, suggests by one interpretation of regulations, is that the products then are being produced under conditions whereby contamination of the product may occur.

So, in order to get the maximum amount of information in terms of control and information on prevalence, some flexibility in the interpretation of the regulation for that specific organism would be desirable in order to gather as much information from across the industry in a manner in which it can be shared and made available for review and use in developing policy.

That is almost not a question, but I tried to make it a question.

DR. BUSTA: I have a couple to add. I think we need to study the full ecological aspects of this organism in nature, in the plant, and in the human, and what unique characteristic permitted to do the things it is doing, becoming dominant in a dry situation, what results in its capabilities there, whether it has some unique aspects of coming in, in certain dry ingredients, and why it might exist under those conditions, the possibility of growing at low water activities.

Some of the information indicated that it resisted dehydration, but there is a great amount of information in its ecological relationships that does not exist even its competitive nature against other organisms in similar situations.

I would also like to see that relationship between these ecological characteristics related to its pathogenicity in that some of its characteristics may make it uniquely suited to establish itself in the feeding system, and I don't necessarily mean in the infant itself, but maybe it has a great deal of ability to form biofilms.

If you recall some of the information about a sediment or precipitate, which to me says if cells are falling out, it means a lot of extracellular material are being produced, and that may permit it to be an excellent biofilm producer, and it may be colonizing, in fact, a feeding tube, not the infant, and then sloughing off toxins.

So, there is a tremendous amount of opportunity there that I think I can see researched. It, in turn, relates to hygienic design. I am not sure about the hygienic design of the feeding systems, the tubes, et cetera, or if that is even considered very much, if it's self-cleaning or what it does.

Maybe the food industry that is trying to find all the hygienic designs of the dehydration systems might be able to contribute to hygienic design of feeding tubes and vice versa.

I think someone else has pretty well covered the others.

Alex.

DR. ACHOLONU: I was having a chat with Dr. Tompkin and we were talking about the question of allowable levels of the organism in the can or any container, and he left me with the impression that the organism doesn't increase, it stays there.

The thing is it is in a dry condition, it is in an inimicable environment, it is not able to grow because the condition is not conducive for replication, but then how long does each stay that way, is it one month, is it three months, is it six months?

Let's say you have a baby food can that was bought today. If it stays for six months, will the organism still be viable? In short, we may need to study the longevity of the organism. That is one thing I wanted to suggest.

Another is we talked about Enterobacteriaceae. Is it possible to have multiple infection, different species occurring in only one person or one baby? If this occurs, does it exacerbate the pathological conditions?

Those who are interested in pathology may want to look into this, what happens when you have E. sak and E. cloacae?

Thank you.

DR. BUSTA: Dr. Fischer.

DR. FISCHER: I would like to propose, if it hasn't been proposed already, that the microbial ecology of the organism in relation to the developing flora in the GI tract be understood. Certainly, the other flora might inhibit or enhance the growth of this organism or perhaps the unit's virulence.

So, I think this microbial ecology of development needs to be done.

DR. BUSTA: Dr. Tompkin.

DR. TOMPKIN: I think it would be helpful to do a survey of the literature including patents with regard to technologies for reducing populations of gram-negatives such as this in dry products.

For example, in the 1970s, we actually built a small batch process or system for pasteurizing gums and stabilizers in a plant we had in Kearny, New Jersey, and it essentially involved controlled humidity, I forget what the level was, and then heating at 160 without leading to denaturation or product degradation.

That was intended specifically to kill Salmonella, so killing enteric organisms in dry powders is a difficult task to retain the functionality and other features, but there may be other opportunities that have not been explored.

DR. BUSTA: Have we exhausted the research? Critical knowledge gaps and research priorities relative to the need to address issues about the presence in powdered infant formula?

DR. FULLER: That's it.

DR. BUSTA: Well, we could have spent another 15 minutes discussing to get to the same end.

I, Frank Busta, the only reason I know how to pronounce my last name is there is a rock star who uses it as his first name, apologize to most of you for mispronouncing your name at one time or another.

Thank you all very, very much for a really highly engaged committee meeting. Each and every one of you was highly engaged in the entire process and very involved.

I would like to thank on behalf of my Co-Chair, all the FDA staff and especially Don Zink, the industrial representatives who stayed on and responded to a lot of our questions, and all the help and assistance from everyone who is here.

It was I think an excellent committee meeting and we look forward to seeing you again.

Thank you all.

[Whereupon, at 2:13 p.m. the meeting concluded.]

- - -