• Decrease font size
  • Return font size to normal
  • Increase font size
U.S. Department of Health and Human Services

Animal & Veterinary

  • Print
  • Share
  • E-mail

Human Health Impact of Resistant Food-borne Disease

J. Glenn Morris, Jr., M.D.

DR. MORRIS: Thank you. It is a pleasure to be here. And if I can make the contraption up here work, I should be in business. I am going to need your help.


I just sort of wanted to follow up on what Fred presented. Basically, Fred gave the overall data on incidence of food-borne disease in the country based on the FoodNet estimates. And I would emphasize, the FoodNet database is really a fabulous database which has moved us forward substantively in our understanding of the occurrence of food-borne disease within this country.

I sort of saw my role as trying to look specifically at the issue of resistant pathogens and sort of asking the question, what happens -- what is the human health impact if you are dealing with a resistant microorganism as opposed to one that is sensitive to the standard complement of antibiotics.


The primary areas of concern are two-fold. First of all, there are concerns related to the direct transmission of resistant pathogenic microorganisms to humans. In other words, resistant Campylobacter, resistant Salmonella. And, again, I would focus on those two based on the data that Fred has shown and other data from a variety of sources suggesting that those are the major causes of human health problems associated with food-borne disease.

This is not to say that there are not substantive problems with other pathogens. But at least as an initial point of focus, these two appear to be a not unreasonable starting point.

I would emphasize, however, that there is also a second issue which relates to the transmission of genetic material or resistance chains to colonizing microbial flora. The concept here is that rather than -- or, you know, in the first instance, you are talking about a pathogen that can directly cause illness in humans.

In the second instance, you are talking about a microorganism which may carry a resistance chain which in and of itself may not cause illness in the patient at that point in time, but has the potential of introducing that resistance chain into the microbial flora of the patient. And, again, keep in mind that we as humans are covered with, filled with bacteria. We have a very intricate microbial flora. And this microbial flora becomes extremely important when you begin to talk about immunosuppressed patients and particularly patients who undergo transplants, organ transplants, bone marrow transplants or intensive chemotherapy.

What you become infected with when you are immunosuppressed as a patient is what you are colonized with. And so consequently, you are colonizing flora. And the resistance status of your colonizing flora becomes an extremely important element in terms of your risk when undergoing subsequent procedures designed to cause immunosuppression.

And I think the two microorganisms that have been the focus of concern in this category would be the enterococci. And there are potential concerns really to E. coli and other enterobacteria. E. c. -- again, the data in these areas are very poor, actually virtually non-existent. But I think these are areas that need to be kept in mind, particularly in the context of the increasing levels of antimicrobial resistance we are seeing in hospitals throughout the United States.


If we focus specifically on Salmonella and Campylobacter, the problems of antimicrobial resistance are initially related to failure of therapy. In other words, if you have a serious infection and, as Fred has pointed out, serious infections with these microorganisms do occur, particularly with Salmonella, and you have a resistant organism or multi-resistant organism, then that organism is not going to respond to therapy.

DT104s have attracted a great deal of attention as you all are well aware. These basically are strains that combine resistance to ampicillin, chloramphenicol, streptomycin, sulfonamides and tetracycline. And there are suggestions of increased morbidity and mortality associated with infections of these strains.

Again, some of these data are difficult to interpret because it is hard to sort out cause and effect relationships. But nonetheless, there are data suggesting that these strains do cause more severe illness or have the potential for being associated with higher levels of morbidity and mortality.

I think from a straight clinical standpoint -- and, again, speaking as a clinician who sees patients on a regular basis -- the suggestions of decreased quinolone susceptibility are probably of even greater concern. And, again, I would hope that many of you have seen the recent article that appeared in the New England Journal of Medicine on November 4th relating to the Danish outbreak of strains which showed a decreased quinolone susceptibility.

And their comment that these -- the patients who were infected with these strains, despite the fact that these strains were technically susceptible to Ciprofloxacin, that there was a "lack of clinical effect" of the quinolones.

I would emphasize the importance of this because, again, from a clinical standpoint, the quinolones are our primary drug in terms of management of Salmonella. Salmonella is -- can be a very devastating infection, particularly in the very young and the very old. It frequently infects endothelial surfaces.

The quinolones have in many ways been miracle drugs with Salmonella. They show excellent cure rates. They penetrate into areas where you don't get good penetration with other drugs. And clearly the drug of choice for disseminated Salmonellosis are the quinolones.

In that sense, the DT104s are of concern. But you are not necessarily dealing with the drug of choice. When you begin to deal with decreased quinolone susceptibility, particularly when combined with strains that carry the DT104 resistance pattern, you have a very significant clinical problem because you are beginning to lose your drug of choice. And, again, Salmonella infections can be very severe and life-threatening.


In terms of Campylobacter, again, quinolone resistance, I think at this point there is good documentation that there are increasing problems with quinolone resistance in Campylobacter. Available data suggests that quinolone-resistant strains result in a longer duration of diarrhea. Data out of Minnesota, some of the FoodNet data, you do get several days of increased diarrheal illness.

However, we really don't have a good handle on some of the other health impacts. As Fred has pointed out, Campylobacter is not always an innocuous disease. And, again, our reporting systems are probably skewed to the more severe end of the spectrum. But nonetheless, you do see a substantive hospitalization rate.

And there are good data suggesting that an immunocompromised host, particularly patients with AIDS, Campylobacter can be a very severe illness. And in those patients, loss of the quinolones may become a very important factor. Again, speaking to someone who sees AIDS patients, I am very concerned about this ongoing difficulty and the specter of decreasing availability of quinolones as a first-line therapy for patients with disseminated Campylobacter infections.

I would also raise the question about the Guillain Barre syndrome rates. As you are aware, the predominant long-term sequelae and by far the most serious long-term sequelae of Campylobacter infections is Guillain Barre syndrome. And at this point in time, we don't have a good feel for what is going to happen if we lose our primary therapeutic agent against Campylobacter in terms of ongoing rates of GBS.

So we really don't have any data on these other health impacts. But I think that these are clearly areas that need to be looked at because they may make a profound difference in the way in which we deal with these data.


Other issues, there are suggestions that resistant strains may have increased virulence. Again, it is difficult to tease out the effect of increased virulence, increased numbers of hospitalization versus various types of reporting bias. But this had been suggested.

There are also issues relating to increased transmissibility of these agents, particularly in association with prior antimicrobial use. And it is very clear that if you perturb someone's colonic flora with prior antibiotics, it, you know, sets them up for infection with a multi-resistant strain.

And there are even suggestions that the infectious dose may be dropped, again, in the setting of prior perturbation of the colonic flora with antimicrobial agents which is not an uncommon circumstance these days. If you go to your physician, you may well get an antibiotic for something and that may well set you up for subsequent infection.


In terms of introduction of resistance chains, I think most of the focus there has been on the enterococci, particularly on the resistance to vancomycin, VRE, vancomycin resistant enterococci. I would emphasize the concept that I mentioned earlier is that the colonizing strains, the strains with which you were colonized in your intestinal tract are the strains with which you become subsequently infected.

And, again, we have shown this in several studies, following patients longitudinally. Once you are infected with a VRE strain, you basically are infected with that strain for life. The numbers may drop to low detectable levels. But if you are subjected to antibiotics or chemotherapy, that strain will re-emerge.

And if in turn you were at a severely immunocompromised state, that strain which may be untreatable with currently available antibiotics may well be the cause of your demise. So you really don't want to introduce resistance chains into the colonizing flora in patients.

I would also note that there have now been several studies pointing out the significant cost associated with vancomycin-resistant enterococci versus vancomycin-sensitive enterococci. Estimates vary widely, anywhere from several thousand dollars up to $90,000.00 or $100,000.00 per case. So vancomycin-resistant enterococci is a substantive concern.


Now, I will say that in this country, we have had substantive problems with vancomycin-resistant enterococci. This happens to be our own home-grown problem in University Hospital in Baltimore. And it is a substantive problem with deaths associated.

Now, of course, the thing in the United States is that we have not used the vancomycin analogues in animal feeds. And so consequently, this appears to be, speaking as a physician, of our own doing associated with our heavy use of vancomycin in the hospital setting.

Nonetheless, I think there are increasingly convincing data coming out of Europe that there can indeed be introduction of vancomycin resistance chains through the food supply and, again, demonstrating that these are possible transmission routes.

I think the other point that I would make with these is, again, the concept of endemicity. What has happened in the United States is that VRE has become endemic in patient populations. We are finding that 20 to 25 percent of all hospitalized patients carry VRE in their intestinal tract.

Again, for most of these patients, these are innocuous colonizations. They don't cause any problem. But, again, if you have got a patient with VRE who you subsequently try to do a bone marrow transplant on, they are at substantive risk that they will develop VRE bacteremia.

You get to the concept of thresholds on this. And my sense in watching the VRE epidemic progress -- and, again, it truly has been an epidemic which has progressed over the past decade -- is that it is very difficult to set a lower threshold; that once you begin to see the genes introduced into human populations, these will be amplified by use of drugs in humans.

And so the key factor is not introducing the gene into human populations in the first place because, again, I can tell you that we as physicians will be using these drugs when we have to. And when we do, then we will get amplification and we will end up with the type of situation that we currently have with vancomycin-resistant enterococci in this country.


The resistance to quinupristin and dalfopristin I think is probably one of the major concerns right at the moment. Again, as you are aware, this is a drug, analogues of which have been widely used in agriculture. There is a high rate of resistance in agricultural isolates. We have found a low rate of resistance in humans. Actually, there have been several studies which have documented this.

And I think the real question, and it is going to be an interesting natural experiment if you will, will be to see with the current introduction of Sinersid as a drug for human use within the past several weeks, whether we will see an amplification cycle in people of these genes and of these resistant strains that we are finding at very low levels in terms of colonization in humans at the present time.

I can tell you, there has been very aggressive promoting of Sinersid as a drug in this country. And so there is likely to be fairly heavy clinical use. And, again, I think this -- at least on the part of the physicians. And I think, again, it is going to be interesting to see what happens given the low level of resistance that we know is already present in the human population.

The gram negative microorganisms, again, as I said, there are no data. These are obviously areas of concern. Again, I can tell you, since I spend a fair amount of time watching levels of resistance within hospitals, there are substantive increases in resistance levels in gram negative microorganisms within hospitals.

Probably most of this is driven by physician use of antimicrobial agents. But I think there needs to be some awareness that there may also be some transfer of genes between animal and human populations which at least need to be thought of or looked at more closely.


So to summarize, in terms of the impact of resistant microorganisms, there are two routes that would appear to be a major concern: the direct transmission of resistant pathogenic microorganisms and, secondly, the transmission of genetic material or resistance genes to colonizing microbial flora.

There is a clear health impact associated with resistant strains. But at the same time, as has been repeatedly said, there are substantive data gaps and there is a clear need for further quantitative data and further modeling in these areas. Thank you.


DR. STERNER: Questions for Dr. Morris Please go to the microphone.

MR. : Just a couple of points. Do you think -- you gave nice examples of the physician-driven vancomycin resistance. Right And I just wonder if another illustration of that might be if you look around at different countries in the world who don't use vancomycin as much as here, who don't use treatment of Campylobacter early on, they have much less resistance problem. And I think we have to be careful not to blame the animals too much or put the blame on our own doorstep.

For example, the use of gentamicin for 30 years in animals has produced no resistant strains in man at all that we have found at present, although the use again may select for those resistant strains later on. So I just wondered if you look around geographically, if you think it confirms the hypothesis that physicians do have quite a lot to do with this.

DR. MORRIS: I think there is absolutely no question speaking as a physician that physicians play a very substantive role in development of antimicrobial use in human populations. I think, however, there is the issue both of the ongoing physician use of these drugs and the potential introduction of resistance chains in the human populations where within which there can then be subsequent amplification.

There is no question that physicians in this country through their antimicrobial use practices amplify resistance. The question is can we change that. I can tell you, having been very active with CDC programs related to judicious use of antimicrobials, being very active in our hospital in trying to restrict antimicrobial use, I would love to tell you that we are going to be able to successfully change the prescribing practices of physicians in this country. But I can't.

I can tell you that all of the efforts we've made to date to try to make a substantive impact in the prescribing practices of physicians have not worked that well. And so I think there needs to be a recognition that while, you know, we may not like it, there is, indeed, going to be amplification once genes are present in the human population. And that amplification is clearly going to be driven by human use of antimicrobial agents.

But I think that -- when you are talking about risk assessment and about threshold levels, I think it has to be recognized that there will be that amplification pattern.

MR. : Could you explain further your concern for Guillain Barre and fluoroquinolone resistance, what that connection might be

DR. MORRIS: As I said, this is extremely -- you know, these things are extremely speculative in that one of the -- Guillain Barre appears to be basically an immunologically mediated response to infection with specific strains of Campylobacter. The question is whether early treatment of Campylobacter might in some way abort that response or might have some impact on Guillain Barre.

This is entirely, completely speculative. I have no idea. But I think again, if you were beginning to look at health impacts to design risk assessment models, I think there needs to be a recognition that the lack of an effective first-line drug against Campylobacter may have an impact further downstream on long-term sequelae.

And I think, again, if you look at the medical impact of Campylobacter, by far the greatest costs are associated with Guillain Barre as opposed to the acute diarrheal episodes. I am not saying there is any association. I am simply saying if you think about what needs to go into a modeling process, that don't forget the downstream sequelae and the possible impact of the lack of an immediate, effective, first-line drug.

Now, again, erythromycin is available. But the quinolones have been awfully nice. And at least at this point in time, I would say that by far the standard practice pattern is to use the quinolones.


MR. : Glenn, that was a very nice talk. I would just like to provide some follow-up on your comment that our efforts to influence human physician prescribing practices haven't worked that well. See, this is actually a very complicated area as you may know. There is out-patient and in-patient and different specialties. It is kind of like saying drugs on the farm. You know, I mean, there is just such great variation.

We have had challenges particularly in the hospital prescribing practices. But we are in the process of learning how to influence the primary care prescribing practices. And we have a number of intervention projects that are starting to show benefit. We had a workshop of these projects -- well, primary investigators of these projects in June. And a report of this workshop is going to appear in the American Journal of Public Health.

So this is difficult. It involves patient education, behavior of the physicians and other parameters. But we are actually starting to learn how to do it.

DR. MORRIS: I concur. And, again, there is some beautiful work being done in trying to change physician practices. Nonetheless, if you are talking about a risk assessment model for today, I think it has to be recognized that there is an inevitable physician amplification component of it.

I sincerely hope that ten years from now, the impact of that physician amplification will be substantively less. And, again, we are actively working on that and I know a number of centers are. But for right now and probably for the next three to four to five years, it must be recognized that the physician amplification component is unfortunately, and I emphasize unfortunately, an inevitable component of any type of modeling that you do.

DR. STERNER: Thank you, Dr. Morris. It is always reassuring as a veterinarian to hear the words to the effect of, "Physician, heal thyself."

Our next speaker it might said has a history of professional parapetic peregrination, or more properly interpreted, that means that his resume looks as if he wasn't able to hold a steady professional job.


Will Heuston is a veterinary epidemiologist experienced in risk assessment, risk management, risk communication in industry, government and academia. And I might add on a personal note for those of you who have concerns over the transmissible spongeiform encephalopathies, that he was one of the visionary people who probably 12 years ago now was on a committee that saw fit to keep that problem from rearing its ugly head here in the United States and the devastating impact that it would have had on the cattle business in the United States and Canada. With that, Will, I will introduce you to talk about risk assessment. Put it all together.