Foodborne Illness, E. Coli,
and Food Processing Issues

Table of Contents

E. coli Background
First E. coli O157:H7 Outbreak
Prior work with E. coli O157:H7 in Washington State
First E. coli O157:H7 Outbreak in Washington State
E. coli O157:H7 Jack in the Box Outbreak, 1993
DNA Fingerprinting
E. coli O157:H7 Odwalla Outbreak, 1996
Salmonella
Salmonella and Orange Juice
Salmonella and Alfalfa Sprouts
Campylobacter and Drinking Water
Student Questions

Readings

E. coli Background

A general feature of E. coli is that they are named according to various antigens. There is the somatic antigen, the O antigen, related to the body of the bacteria itself, and the flagellar antigen, which is called the H antigen. There are various permutations of the two antigens and one that is of particular interest, especially in the United States, is the O157: H7. There are many other types, some of which are non-pathogenic and others that cause equal problems to that of O157:H7.

[Escherichia coli O157:H7]

Here is some general background on E. coli O157:H7 and its spectrum of illness:

It can cause disease from asymptomatic infections to severe kidney disease with anemia called hemolytic uremic syndrome (HUS). The biggest problem is the simple diarrhea. There are many other causes of bloody diarrhea. Besides E. coli O157:H7, this includes Shigella, Salmonella, Clostridium difficile and others. A smaller subset develop HUS, which is a combination of renal failure, with severe anemia, called microangiopathic hemolytic anemia. HUS can be fatal especially to children. About 5% of the cases of E. coli O157:H7 develop this problem. HUS occurs more frequently in young children than in middle-age adults, and it is different from other forms of bloody diarrhea.

[Diarrhea, Bloody Diarrhea, HUS, and E. coli O157:H7 Infection]

The sequence of events with regard to E.coli O157:H7 is as follows. About three or four days after ingesting the bacteria, people develop a non-bloody diarrhea, which then develops into a bloody diarrhea in many instances. This either can resolve or develop into complications such as HUS. [Sequence of Events] From the time of exposure to onset of diarrhea can be days. This means a long delay before the illness comes to a physicians or health department’s attention. In general with infectious diseases, individuals have to become ill, they have to become ill enough to see a doctor, and the doctor has to obtain a laboratory test to be performed. Because of this, there is a considerable time lag between the time that people are exposed to the time that you have some sort of lab report. [E.coli O157:H7 Infection]

First E. coli O157:H7 Outbreak

The story began for E. coli O157:H7 in 1982. There were two outbreaks, one in southern Oregon and one in Michigan with bloody diarrhea of unknown cause, and these were associated with eating fast food hamburgers. They were not Salmonella, Shigella, Campylobacter, or any of the pathogens known at that time. The Centers for Disease Control and Prevention worked for many months on identifying the cause of these outbreaks. Eventually they found that it was caused by this certain serotype of E. coli, O157:H7. Prior to that time, E. coli was not known to be a disease-causing organism.

Prior work with E. coli O157:H7 in Washington State

We had been working on E. coli O157:H7 for some time. How do you study something before it emerges? I do think it is possible. Most of these diseases that we talk about as "emerging pathogens" make their appearance in subtle ways for years before they become major public health events. I think it is at that time we need to gather as much data about them in the event that they become worse problems.

In Washington state we began working on O157:H7 in 1984. We worked with Group Health Cooperative, a large health maintenance organization. We made an arrangement with them that for every stool that was tested for bacterial culture at Group Health Cooperative they would also test for E. coli O157:H7, which was a research test at that time. Then we did a case control study. We compared food histories of those who had become sick with those who had not become sick in order to identify risk factors. What we found was that E. coli O157:H7 was not unusual in Washington. It was as common as a well-known bacterium, Shigella, and that it was also associated with eating poorly cooked hamburger. This was in the absence of any discernable outbreaks.

Also in the mid-1980s, two researchers at the University of Washington, Drs. Phil Tarr and Peggy Neil, were doing a special study trying to determine the cause of HUS. At that time, no one knew that it was caused by E. coli O157:H7. Every child that was hospitalized with HUS was tested promptly for E. coli O157:H7, and it was found that if those children were tested soon enough, almost all of them were positive. E. coli O157:H7 is not the only cause of HUS, but for Washington State and for the United States, it is the primary cause.

First E. coli O157:H7 Outbreak in Washington State

In 1986, we had our first large outbreak of E. coli O157:H7, which was in Walla Walla, Washington. The outbreak was associated with a fast food taco restaurant. However, we were not notified of the outbreak promptly. We found the outbreak because three individuals with severe complications were transferred from Walla Walla to Seattle, and the Seattle physicians notified us of the problem. We traced them back to Walla Walla and found that there was an outbreak. We also found that there was an outbreak in a nearby nursing home as well. After the outbreak was thoroughly investigated, retrospectively, we found that cases of E. coli O157:H7 were occurring in western Washington at that time as well. That generated a lot of enthusiasm for E.coli O157:H7, and we made E.coli O157:H7 a reportable disease in Washington state. Washington was the first in the United States to add E. coli O157:H7.

All of these things laid the groundwork for the preparation and response to a big outbreak of E. coli O157:H7 in 1993. In addition, Washington State has a long history of being good at foodborne disease investigations. There were some very astute epidemiologists here who were very conscientious about foodborne disease. It had a reputation, back in the early 1980’s as being one of the better states with regards to foodborne disease investigations.

E. coli O157:H7 Jack in the Box Outbreak, 1993

Washington State became famous with regard to E. coli O157:H7 in 1993 with a very large outbreak. This was related to contaminated Jack in the Box hamburgers. There were 500 culture-confirmed cases and 100 probable cases in Washington State. About 150 of the cases were hospitalized and there were 45 cases of HUS of which 28 required kidney dialysis. Three children died related to this outbreak.

The Washington State Department of Health was able to identify the source of the outbreak only five days after our first notification. We were notified on the 13^th of January and we made our public announcement around the 17^th or 18^th.
[E. coli O157:H7 WA State 1993] This is a plot of E. coli O157:H7 cases, not by date of the onset of illness but by date that they ate the hamburger. These are not all the cases, but these are individuals who could remember what particular day they went to the Jack in the Box and ate a hamburger. It was a rather sobering thought to imagine what things might have been if we had delayed our identification by only a few days. Only weeks after the outbreak did we know the true incidence of disease, which was in the hundreds.

We were able to quarantine and recall about 250,000 hamburger patties, or about 60% of all the hamburgers that were implicated in this outbreak. If those hamburgers had been consumed, we estimate that the number of cases would have been at least two times more than the number that we had. As it was, that was the largest, and remains today the largest, foodborne outbreak of E. coli O157:H7 in the United States, but not in the world.

This outbreak is a good illustration of the importance of doing good field epidemiology – asking questions about person, place, and time. We were able to confirm the causative agent with a case control study. The other important point of this outbreak to remember is that this was not the first time we began working on this problem.

A major point with investigating outbreaks successfully is the principle that in outbreak investigations, you are never making your decisions with complete information. If in outbreak investigations we waited for everything to be finished, the outbreak would be over. We are always making decisions with partial information.

At the time of the 1993 outbreak, we had about 13 outlets of the Jack in the Box chain mentioned by cases. On the other hand, there were a total of 66 Jack in the Box outlets in the area. One of the puzzles we had when determining whether Jack in the Box was really related to the outbreak was why were only these 13 named? We went through every possible food item, other than hamburger. We tried to determine if it was anything that might have been distributed to some sort of subset of the chain that could be implicated rather than the hamburger.Even though we knew that E. coli O157:H7 is frequently related to hamburger, it is a big mistake to say, since it is probably related to hamburger, therefore it must be the hamburger.

There have been instances where there have been outbreaks of cholera, and someone said it is probably the water. Do not drink the water, and then it would be found that it was the bottled water, which everybody was advised to drink, that was the cause of the problem. We ruled out everything and we were not able to find any other explanation. We had significant relationships between hamburgers and becoming ill, so we made the public announcement. Eventually, the explanation became known. The answer was that it was just a matter of time. Eventually, almost every single Jack in the Box outlet in the Pacific Northwest was named.

I think another part of that equation is that if you get partial information from people that you know and trust and your ability to make decisions is a lot better. As a consequence, I think a lot of the ability to respond to emerging pathogens is establishing good relationships ahead of time and knowing who your team members are in terms of working on outbreaks.

DNA Fingerprinting

What has happened in Washington since 1993? One of the things that was done with the isolates from the Jack in the Box after the outbreak was to perform DNA fingerprinting on them. [DNA fingerprinting] The isolates are stretched on a plate, in the upper left hand corner and you grow the bacteria. Then you make plugs of the bacteria and digest the bacterial DNA by an enzyme that chops it up in a very systematic way. You run that through an electrical field and you get bands that look like this in the slide. [Alpine, Wyoming Outbreak of E. coli O157:H7] If the bands are similar those isolates are probably identical.

DNA fingerprinting was available as a research procedure at the Centers for Disease Control and Prevention back in the early 1990’s. When the isolates were tested from the Jack in the Box outbreak, it was found that the isolates from the meat matched the isolates from the humans. But more importantly, there were also E. coli O157:H7 cases that had occurred in December, a month before the outbreak was identified, which had the same pattern. We thought it was nice that we were able to prevent hundreds of cases of E. coli O157:H7 by identifying the meat, but it would have been even nicer if we had prevented the outbreak altogether by being able to identify the outbreak even earlier.

The technology of DNA fingerprinting became more available by the mid-1990s. There were two different types of tests. One is called pulse field gel electrophoresis (PFGE), which we do at the state laboratory. The other is called RFLP, which is done by a microbiologist at the University of Washington, Dr. Mansour Samadpour. It is a different type of test that basically does the same thing as PFGE. We had the benefit of both methods in this area.

Another thing that was done was that a network of laboratories around the country was established in the mid-1990’s which performed pulse field gel electrophoresis. All of these laboratories were performing the same procedure in a standardized way at the same time. Washington State was one of those states. The method that is used by all of these states was developed here and was given to the other jurisdictions. The images of bacterial fingerprints are transmitted through the Internet to a database in Atlanta, Georgia. Because of this, it is possible for us to produce an image here in Seattle and see if there are similar patterns that are occurring elsewhere in the United States.

E. coli O157:H7 Odwalla Outbreak, 1996

In October of 1996 there was an outbreak with only two cases of illness that had the same RFLP pattern back. [Outbreak timeline] More cases were identified and these also had the same pattern. Interestingly, there were about three E. coli O157:H7 cases that had an RFLP pattern that was different. We thought these were unrelated to the other cluster and excluded them from the case control study. A very small case control study was done. Ten of ten cases drank Product A apple juice and zero of nine controls drank Product A apple juice which was significant. It was extremely helpful to be able to exclude outliers from the outbreak data by RFLP, because if we had ten of thirteen and zero of nine, we would not have had a significant result. Because of this we were able to implicate apple juice as the source.

Our conclusion was generated out of a combination of good epidemiologic fieldwork plus molecular epidemiology. Initially, there were two different types of apple juice that were consumed. One was Odwalla apple juice. The other was Starbucks apple juice. So we were not really sure if we had a common source problem until someone found out that Odwalla was at that time the sole supplier of Starbucks apple juice. If you drank apple juice from Starbucks, you drank Odwalla juice.

After the case control study was done, there was a product recall the day before the Halloween holiday. We were under time pressure to make a decision because we had this vision of little children going out collecting candy and apple juice at parties. [Outbreak timeline] We made our decision with the case control study. We did not wait for the positive culture. This is the principle of needing to make decisions with partial information, although if you have a well-conducted case control study and statistical significance, there is a good presumptive piece of information. It was found that we were not dealing with a local problem. Odwalla apple juice is produced in California and distributed to Canada, as well. [Geographic distribution] This is the distribution of cases.

[Dates of pure apple juice consumption] This is an example of the source of this outbreak. The red is the lot of apple juice that was positive. The vertical part of the arrow is when the apple juice was produced, and the horizontal part is when it was consumed. These are dates of consumption when the dates of consumption related to the outbreak were known. The green bars are uninvolved lots.

[Distribution of Oct 7 apple juice] This is another distribution curve with regards to the Odwalla outbreak. The yellow bars are the amounts of apple juice distributed at various locations on the West Coast. The red bars are numbers of reported cases of E. coli O157:H7. I think that you can tell that there is a pretty good correspondence to where the apple juice was distributed and where the cases occurred.

Salmonella

Next I will be talking about a couple of Salmonella outbreaks. Salmonella has been around for a very long time and is a cause of bloody and non-bloody diarrhea. With Salmonella, as with E. coli, it is possible to serotype it. For Salmonella the various serotypes have different names. They have names like Salmonella seattle and Salmonella san diego. Which serotype causing the infection does not make a great deal of difference on how sick a person becomes. They are epidemiologic markers that we are able to use to trace outbreaks using pulse field gel electrophoresis.

Salmonella and Orange Juice

In the summer of 1999, we got a call from a microbiologist at Children’s Hospital stating she has three cases of Salmonella C2. Later we serotyped it, and it was Salmonella muenchen. One of my epidemiologists started working on the problem. The cases were an interesting combination of individuals. One was a child from Alaska, and that child happened to be in Seattle only for a few days. The child previously had cancer and was here for a follow up. The child stayed over at Ronald McDonald House, went around town, ate at a few places, went home, and then had this problem. Another was a child who actually lived in this area. The third was an adult from north of Seattle who had a congenital agammaglobulinemia. We interviewed these individuals and one interesting feature was these persons had consumed a blended fruit drink called a smoothie at a particular chain in Seattle. However, not all the cases drank the same smoothie.

Smoothie Ingredients

There are all kinds of smoothies. The diagram above illustrates the ingredients used in the various smoothies served at this chain. We began thinking about the smoothies that these three individuals drank. They also had some friends who were with them who also drank smoothies, and some got sick and some did not. We compared the smoothies that these individuals drank. There were two items that the sick people seemed to be exposed to, orange juice and orange sherbet. We continued the investigation the following Monday. Our lab did PFGE of all the case isolates and they had an identical pattern.

Public Health – Seattle and King County began working on a further investigation, and were finding additional cases during that time. Then we contacted the CDC with the results, because if it was the orange juice, this was produced outside of Washington State. That means that it is an interstate problem and we need to talk to other agencies, like the CDC and the Food and Drug Administration. About 10 minutes after we had this conversation, we got this excited callback letting us know that the Oregon Health Department was also seeing an outbreak related to Salmonella muenchen. Their outbreak was related to a breakfast buffet. They were actually able to calculate statistical significance because they had a group of individuals at the buffet and could do a cohort study. The two items implicated were French toast and orange juice. We asked where is this orange juice from? It turned out to be the same orange juice source that we were working with, the Sun Orchards orange juice in Tempe, Arizona. Oregon was able to perform PFGE on the isolate and was compared with the Washington strain. It appeared that we were dealing with the same strain.

We then began investigating the food chains involved, and we were able to get Salmonella from the blenders and the dispensers. Subsequently, there were a number of isolates from other states that also matched. Some of these are Salmonella muenchen that do not match the outbreak pattern and the others with the arrows are the ones that do match. Eventually it was found that this outbreak involved 20 states in the United States and three Canadian provinces.

Next, the Food and Drug Administration tested orange juice from retention tanks, or storage tanks, at the facility, and those were positive for Salmonella. Actually, it was positive for about ten different varieties of Salmonella, including Salmonella muenchen. We found that not all of the orange juice was squeezed in Tempe, Arizona. In fact, a significant part of the orange juice was squeezed somewhere in Mexico and shipped by tanker truck up to Tempe. The orange juice from Mexico was then blended with the United States orange juice, bottled, and distributed to the 20 states and the three Canadian provinces. The Food and Drug Administration also obtained a sample from a truck before the truck unloaded its orange juice and that was positive for Salmonella muenchen. This isolate had the same fingerprint as the outbreak isolate.

We do not know how many people truly became ill. In Washington State, we had 100 culture confirmed cases of Salmonella muenchen. We did study some cohorts of individuals, groups at banquets where the orange juice was served, and about half the people who consumed the orange juice became ill with diarrhea – not necessarily culture confirmed diarrhea, but diarrhea in association with orange juice exposure. Calculating from the amount of orange juice distributed in Washington State and estimating that half of the individuals drank a glass of orange juice became sick, we estimate that there were about 10,000 cases of Salmonella muenchen related to this outbreak. We think that for every culture-confirmed case that we had, there were about 100 other individuals who actually were ill.

The moral is to do good investigations on the local level because you do not know exactly where they may go. Outbreaks may turn out to be extremely important as far as national events. Initially, this was a report of three cases from a microbiologist to a state health department person on call on a weekend in Seattle. The trail eventually led all the way to Mexico. This also emphasizes another factor of emergence, trade and travel!

Salmonella and Alfalfa Sprouts

The main reason I want to talk about this Salmonella outbreak is that it illustrates the strategy of combining epidemiologic work and environmental health work with trace back of food products and using your results to develop further public health prevention strategies. This was an outbreak of Salmonella with cases in Washington, Oregon, and Idaho. These Salmonella cases were related to consumption of alfalfa sprouts. The cases all had the same DNA fingerprint and there were also positive specimens of sprouts and sprout seeds that were identified, which also had the same DNA fingerprint. [Salmonella Mbandaka - WA, OR, and Idaho - Alfalfa Sprout Outbreak]

In addition to the outbreak investigation, a trace back occurred with regards to where these alfalfa sprout seeds came from. It was subsequently found that a producer in the Imperial Valley in southern California was the source of this seed. Interestingly, it was found that the sprout seed that was sent to Washington State was also sent to Florida and several locations in California. However, there were no subsequent cases of salmonellosis related to sprouts produced in those areas, and we were looking quite carefully for cases in those areas.

First some background information on sprouts. One of the big problems with regards to foodborne diseases and sprouts is that it is very difficult to control bacterial contamination of sprouts. When sprouts are produced, the seed is soaked, and then they are put in an incubator drum where the seeds sprout and grow. It is a warm environment with plenty of nutrients, an ideal place for bacteria to multiply. It has been found that very large numbers of bacteria are produced. If you count all bacteria on finished sprouts, they may have 10⁸ to 10⁹ bacteria per gram. Very few of these are pathogenic bacteria. The problem is that if you have a pathogenic bacteria in the seed to begin with,even with a very small amounts of pathogenic bacteria, those can multiply and produce significant amounts in the finished product. People have tried disinfecting the seeds with various chlorine solutions and chemicals and by doing so, the resulting seeds are not viable. There are many outbreaks that have happened over the past several years, and alfalfa sprouts carry a similar risk to that of consuming raw shellfish: most of the time you are OK, but occasionally there are problems.

Outbreaks with sprouts fall into this area of the new scenario outbreaks. These outbreaks tend to be scattered. Very frequently, sprouts are not a major item on the menu but are a condiment in sandwiches, so it may be difficult for people to identify that they actually ate sprouts. Furthermore, E. coli O157:H7 and Salmonella survive extremely well in seed. They can survive for years in the seed. These seeds are transported widely across international borders, certainly across state lines, so you have dispersed outbreaks.

One of the items discovered during the outbreak was that we had contaminated seed from a particular lot. Some of the sprouters did not have outbreaks, and one sprouter did. An interesting factor with regards to this particular sprouter is that he was not using the chemical solution called calcium hydrochloride. Some sprouters have been soaking their seed in 20,000 parts per million calcium hydrochloride before they sprout them. The sprouters in Florida and California were using this method. This person was not. So that generated considerable excitement among the regulatory agencies and sprouters because perhaps this method may be helpful. Twenty thousand parts per million calcium hydrochloride, I have been told, is a very vile chemical solution, and sprouters do not like using the chemical very much, but that appears to be one of the main methods of intervention. This is a good example of how epidemiology and environmental health, with regards to how these seeds were produced and distributed can be used to have a better understanding of to deal with this problem.

Campylobacter and Drinking Water

The final story is a Campylobacter jejuni outbreak that occurred among Washington teenagers. These teenagers had been at a summer camp in British Columbia. It initially appeared that the camp was the source of the problem. The BC Ministry of Health was able to identify Campylobacter from the drinking water, which was unfiltered, unchlorinated, unboiled, surface water. The case isolates and the water isolates matched with PFGE. [Camylobacter jejuni Outbreak ] Because it took a long time to get the isolates due to border regulations we worked with the Canadians on developing PFGE in British Columbia. It is important to be able to work together on international outbreaks. It would be helpful if we were doing comparable testing which could be transmitted electronically as needed. We are already starting to benefit from those results. This past year there was an outbreak of E. coli O157:H7 related to a contaminated Hungarian sausage. It was not from Hungary, but it was produced in southern Vancouver, British Columbia. We were quite concerned that this product was distributed here in Washington State and whether we had any related cases. People crossed the border and purchased items and also the parent company of this Hungarian sausage shop did distribute products to Washington. It was very helpful to have the pulse field gel electrophoresis and be able to compare the outbreak with our patterns.

I am having preliminary discussions with a variety of other countries – Japan, Korea, Singapore, and Hong Kong – with regards to this. All of these countries are doing PFGE, so it is not an exotic procedure. Comparing results across international lines and getting all of the microbiologists to agree on how to do things is very complicated. In the long run, I think that we will benefit from this and be able to share information in a more timely way.

Student Questions

Q: How were you notified in the 1993 E. coli outbreak?

A: What got our attention was a call from a physician, Dr. Phil Tarr. He called me up said, "John, I have seen about 11 cases of bloody diarrhea or HUS in about 30 hours in the Children’s Hospital emergency room or in the hospital. I have never seen anything like this before." This is not a casual statement with regards to an unusual event. I have known this physician for many years. He had worked on E. coli O157:H7 since the early 1980’s, and for him to say "I have never seen anything like this before," is different from someone who was not so acquainted with O157:H7. I think that this will be another recurring theme that you heard about through the outbreaks that I described. While we have developed all sorts of molecular techniques and so forth with regards to identifying these outbreaks, we still are highly dependent on the clinicians, microbiologists, and others to call us.

Q: I am curious how did they select controls for the outbreak involving Odwalla apple juice?

A: The controls were basically best friend controls. For grant writing research programs this may not be the best controls but what I deal with is "quick and dirty" epidemiology. One of the main needs in an infectious disease outbreak investigation is to identify controls quickly. Although it is important for us to have individuals matched for age, because there is a difference between young versus old people and how susceptible they are to E. coli O157:H7. Also, it is important for these people to have an equal likelihood of exposure. For example, in the case control study with the Jack in the Box outbreak, it was important to match for neighborhoods because not all neighborhoods have a Jack in the Box outlet in them.

Q: In the Jack in the Box case of 1993, there were the deaths of three children, and you said there was a death in the Odwalla outbreak as well. What kind of relationship in terms of cooperating or not cooperating did you get from Jack in the Box or Odwalla juice at the corporate level? There is so much at stake. A product’s reputation might be smeared in the minds of the public forever, or it might be healed over time. At the same time, these companies cannot just say, let’s just keep this quiet, recall the product, and not tell anybody.

A: Our philosophy is not to work with food producers in an adversarial manner. In the Jack in the Box and the Odwalla outbreaks, even before the public announcements were made, when there was some suspicion but not confirmatory evidence that their product was involved, we were speaking with them on the telephone, saying, we think we may have a problem, what do you think? One of the big events with regard to the Jack in the Box outbreak occurred the day before we made the public announcement. We had a big meeting with the local health department staff and food inspectors and two managers from Jack in the Box. We were stewing about how in the world we could explain this association by anything but hamburger. A similar discussion was going on prior to the public announcement with regards to the Odwalla episode. I think that, in general, food producers, have learned that when something bad happens, the best thing to do is to say something bad has happened, we are working on it, we will do whatever we can to make things right. Clearly, it is very important to be right with regards to the offending product, and fortunately we have been right. On the other hand, mistakes have been made. One of the most notable mistakes occurred a few years ago related to a Cyclospora outbreak. The Texas State Health Department implicated strawberries from California. Unfortunately, it was not the strawberries. In that instance, the source was raspberries from Guatemala. It is not only a problem with regards to that particular product, but also the credibility of the public health agencies, the food producers, and the departments of agriculture.

Q: What are some of the regulations involving food production?

A: The regulations are highly varied according to the food product. One of the methods of monitoring quality control is to do microbiological testing, but I think that there is frequently the problem that the volume of the product is so large that it is not possible to use microbiological testing alone. One of the items that are very commonly used in food service and by food processors is something called HACCP, or hazard analysis and critical control points. What HAACP does is recognize the fact that you cannot test every item of food that ever gets produced, and what you need to do is develop certain critical control points that are more important than others in maintaining the safety of food. An example of a critical control point is pasteurization. There are many steps in the process of milk production, but a critical control point would be pasteurization. Therefore, all sorts of work would be done in terms of determining the criteria for pasteurization and monitoring quality of the process and so on. There are a variety of different ways, and microbiological testing is done, but testing is not necessarily a substitute for good food handling practices.

Q: How long had the Salmonella in orange juice outbreak been going on before Children’s Hospital noticed those three cases?

A: It is difficult to know, and I think it is entirely possible that there were other events that were undetected during that time. I think this is a general feature with regards to foodborne outbreaks, and one reason why I bring Salmonella up in an emerging pathogens class even though it is not an emerging pathogen. We have known about Salmonella for years, but the type of foodborne outbreaks that we are seeing are different from those that we have seen in years past. The classical "old scenario" outbreak is the church supper outbreak. A number of individuals would get sick after a church supper, an investigation would occur, and it was the potato salad. Usually someone committed some sort of food handling error while preparing the food. There is a new scenario outbreak, which occurs regularly nowadays, where there are widely dispersed cases and these tend to be associated with some sort of centrally contaminated food that gets widely distributed. I think that there are a lot of reasons why one might imagine seeing more of these. I think that our tools for detecting these outbreaks are much better than they used to be. It is possible that we have had this problem, and it is not emerging in the sense of being new, but it is emerging in the sense of being newly recognized. On the other hand, the nature of food production is also changing. It is more common now to have food items that get produced on a very large scale and get distributed much more widely than they used to. Items that were uncommon in grocery stores ten or fifteen years ago are now everyday items.

Q: Have they thought of using irradiation as a means of controlling bacteria levels in sprouts?

A: Irradiation will work, but basically the chances of having any sprouters who are willing to consider irradiation as one of the control methods is not very good. Most people who market sprouts and many people who like to eat sprouts are eating sprouts because they view them as organic and good for you and it is sort of anathema for them to be considering an irradiated product. It is also possible to irradiate the seed. One wonders though what kind of sprouts you will generate out of irradiated seed. I have been told that a "mere dusting of gamma rays on sprout seed in combination with chlorine will knock the bacteria dead." Irradiation is very difficult even for food products where irradiation is approved, such as chicken. There is no market for irradiated chicken and no one is trying it. The main reason is because of the cost of developing radiation mechanisms that are economical and safe and also the concern is that they will spend large amounts of money developing this and no one would buy it.

Readings:

Armstrong GL et al. "Emerging foodborne pathogens: Escherichia coli 0157:H7 as a model of entry of a new pathogen into the food supply of the developed world." Epidemiologic Reviews, 1996; 18(1): 29-51.

Mead PS et al. "Food-related illness and death in the United States. Emerg Infect Dis, 1999 Sep-Oct; 5(5): 607-25.

Top | Home | Topics | Lecture Intro | Go to Lecture