Eye on ID

Salmonellosis: We are what we eat

Large outbreaks of infections caused by salmonellae are commonly reported in the United States, and smaller outbreaks undoubtedly occur much more often and go undetected or unreported. The largest reported outbreak of salmonellosis in the United States in the past 25 years occurred in 2008. It involved more than 1,400 people in 43 states and was caused by the Saintpaul serotype of Salmonella enterica from salsa containing uncooked tomatoes, peppers and cilantro.

Recently, there was an outbreak of S. heidelberg infection, which caused disease in nearly 400 people in 23 states, with most of the cases occurring in California. The strain was apparently particularly virulent with a hospitalization rate of more than 40%, which is more than double the 20% that occurs in most outbreaks of salmonellosis and the bacteremia rate was 14% vs. the usual 5%. Some of the S. heidelberg isolates were resistant to multiple antibiotics. The antibiotics to which they were resistant tended to be those commonly given to animals to promote growth. The outbreak was traced to chickens processed at Foster Farm facilities in California.

Donald Kaye

Donald Kaye

Salmonellae prevalent in animals

These and other outbreaks bring up many important details about salmonellosis. First, it is important to note that salmonellae are widely distributed in the animal kingdom and have been isolated from virtually every animal species. These include mammals (even whales and Tasmanian devils), fish, reptiles, amphibians, birds (not just poultry and poultry products), crustaceans and even insects. The saying “seek and ye shall find” applies to finding salmonellae in any known member of the animal kingdom.

In fact, the second paper I published was titled, “The parakeet as a source of salmonellosis in man: Report of a case.” It was my parakeet and my infant son and the first report from a parakeet. These animals get infected, just as we do, from what they ingest, probably store-bought bird feed in the parakeet. Poultry also become infected from vertical transmission from infected ovaries.

There are more than 2,500 serotypes, or serovars, of salmonellae distributed in nature and, fortunately, many of them are non-pathogenic or of little pathogenicity for humans. Also fortunately, it generally takes a relatively large inoculum of salmonellae to cause disease (102 to 105). It is when one of the more virulent strains for humans becomes widely disseminated in a food product that large outbreaks of salmonellosis occur. Contributors to the spread of salmonellae and other fecal pathogens in food have been the increasing internationalization of the source of food products; the crowding of food-animals both where they are raised and at slaughter houses; and the increasing use of mass production.

As illustrated by the very large Saintpaul serotype of S. enterica outbreak from salsa, and several large outbreaks from peanut products and other non-animal derived food products, virtually anything that is ingested can serve as a vehicle for transmission of salmonellosis. Vegetables, fruits, nuts, spices, etc, may become contaminated at farms from run-off rain water or irrigation water from wells near food-animal areas. For example, the FDA recently found nearly 7% of spice imports were contaminated with salmonellae. Also, feces from humans, rats, birds or other wild animals may contaminate fields. Contamination also can occur during transit or at the production plant from unsanitary processing or, for example, from animals such as rats and birds.

The bottom line is that any uncooked food product may be contaminated with salmonellae and that incubation during processing or transit or in the store or home can result in growth to a disease-producing inoculum size.

The recent S. heidelberg outbreak also highlights that the increased resistance to multiple antibiotics has become a problem with salmonellae and with other gram-negative bacillary causes of gastroenteritis (eg, fluoroquinolone-resistant Campylobacter). Much of the resistance is probably related to widespread use of antibiotics as growth promoters supplied to food-animals to help livestock and poultry grow faster and to prevent disease among animals living in crowded conditions.

In 1996, the Department of Agriculture implemented the Hazard Analysis and Critical Control Points (HACCP) program to provide quality control and surveillance in processing plants to reduce the amount of contamination with salmonellae. Chicken carcass contamination with salmonellae has declined since the implementation of HACCP from more than 20% to 7.3%. Despite this decrease and the implementation of better hygiene in egg production and the vaccination of laying hens, the incidence of human illnesses associated with salmonellae has actually increased during the past 17 years.

The FDA has published guidelines to help reduce the chance of foodborne salmonellosis. Food must be cooked to 68°C to 72°C (145°F to 160°F), and liquids such as soups or gravies must be boiled. Freezing kills some salmonellae, but it is not sufficient to reliably reduce them below infectious levels. Although strains of salmonellae are usually heat-sensitive, they do acquire heat resistance (eg, to pasteurization) in high-fat environments such as peanut butter.

As it is virtually impossible to prevent animal products from becoming contaminated with salmonellae, the only ways available at present to prevent disease are by using pasteurized products (eg, milk, juices, eggs); adequate cooking; and good hand hygiene and good kitchen hygiene (eg, cleaning surfaces, dishes, utensils, etc, after contact with uncooked poultry and meat products). Washing vegetables and fruits may not remove all salmonellae from the surfaces, and salmonellae can become internalized in tomatoes and sprout seeds.

Intensive surveillance needed

Not much can be done in the home with products such as peanut butter; all we can do is warn people about recalls once issued. The only real protection in these circumstances is to pick up outbreaks early with good surveillance and to identify the food sources with good old shoe-leather epidemiology. In the future, irradiation of all foods may be the answer, but it is expensive, and there is negative public perception about safety. Some studies have suggested that much of the public will not buy foods that have been irradiated.

For more information:

Donald Kaye, MD, is a professor of medicine at Drexel University College of Medicine, associate editor of ProMED-mail, section editor of news for Clinical Infectious Diseases and is an Infectious Disease News Editorial Board member.

Disclosure: Kaye reports no relevant financial disclosures.