Reptile-associated salmonellosis: A preventable pediatric infection

Medicine and Health Rhode Island, Jan 2003 by Bandy, Utpala, McCarthy, Helen, Hannafin, Christopher

EPIDEMIOLOGY

Salmonellosis, a major cause of diarrheal illness in the United States, causes an estimated 14 million illnesses and 600 deaths annually. There are 2,449 known Salmonella serotypes of which approximately 200 are detected in the United States in any given year.' Between 1987 and 1997, the rate of reported Salmonella infections (average of 40,169 reports per year) decreased from 19 to 13 per 100,000 persons. However, while some serotypes are reported to have decreased in frequency, serotypes associated with reptiles are showing an increase in frequency. The serotypes with the greatest average annual increase in the number of isolates reported from 1987 to 1997 are Salmonella Stanley, and Salmonella Marina; both serotypes are reptile-associated. Of the top 20 increasing Salmonella serotypes, seven, or 35%, are common to reptile-associated Salmonella serotypes (Stanley, Marina, Flint, Kintambo, Wassenar, Ealing, Carrau, and Abaetuba). Also, data from the National Salmonella Surveillance System, representing 441,863 Salmonella isolates from humans from 1987-1999, demonstrate that the rate of overall isolation of Salmonella is highest in the New England region; that overall, children have the highest Salmonella isolation rate; and that infants have a 4-- 13 fold higher rate of invasive disease than other age groups.2

In Rhode Island from 1997 to 2001 (Figure 1), the total number of cases of Salmonella infections (all ages) decreased from a high of 172 total cases in 1997 to 161 cases reported in 2001. The average number of Salmonella cases reported for this period in children younger than 5 years of age is 44, which represents an attack rate four to five times higher than for older children and adults. Of note is the fact that about a third of the cases (range 28.6 to 35.6%) in the under five age group had isolates known to be reptile-associated strains. Since cases are not interviewed, a history of a reptile in the home could not be confirmed. Of the isolates in children under five, there was an average of 2.8 invasive infections (bacteremias) per year (range 0-7). Whether the bacteremia isolates were reptile associated strains or not is unknown as serotyping blood isolates for Salmonella bacteremia was not routinely performed until recently. It needs to be pointed out that the numbers reported here are underestimates and that the actual number of cases is probably significantly higher. Many cases of Salmonella illness are not reported, not every ill person seeks medical attention, and health-care providers may not always obtain a specimen for diagnosis.

An estimated 3% of American households own close to 7.3 million reptiles, and the number of reptiles imported into the U.S. has increased dramatically from 27,806 in 1986 to 798,405 in 1993.(3) The majority of the imported reptiles are iguanas; snakes and turtles are also included in this category. Approximately 93,000 cases per year of Salmonella infections are estimated to be attributable to pet reptile contact.' Reptile-associated Salmonellosis is not a new phenomenon. In the early 1970s, pet turtles were responsible for an estimated 280,000 cases of salmonellosis each year. The Centers for Disease Control and Prevention (CDC) estimated that in 1973 pet turtles accounted for 14% of the Salmonella-caused illnesses in the United States. In an attempt to control this number of Salmonella infections, a ban was enforced on all interstate shipment of pet turtles with shells of 4 inches or less in length. At that time, the U.S. Food and Drug Administration (FDA) reasoned that turtles larger than 4 inches did not pose as much of a threat for Salmonella infection because children would not likely fit them into their mouths. Since that time, studies have shown that direct contact is not necessary for transmission of Salmonella bacteria. Nevertheless, the ban on turtles proved to be very effective, because an estimated 100,000 fewer annual cases of turtle-associated salmonellosis occurred among children following the ban.5

TRANSMISSION

The principal habitat of Salmonella bacteria is the intestinal tract of humans and animals. Salmonella is endemic in reptiles; iguanas have long been known to harbor Salmonella bacteria. Because iguanas will not breed if closely confined, most of the pet iguanas sold in the U.S. have either been captured in the wild or have been bred in farming/ranching operations in Central America. Reptiles are infected from before birth, obtaining infection as live newborn or shelled embryos passing through the cloaca of the mother. Being captive bred, incubated and born, is no guarantee that any reptile is Salmonella-free; upon entry into the U.S. a large number of these animals (up to 90%) are already asymptomatic carriers of Salmonella. High rates of Salmonella in the feces of iguanas may be related to the eating of feces by hatchlings, a typical behavior for iguanas, and other vegetarian lizards, that establishes normal intestinal flora for hindgut fermentation. Reptiles can also become infected through contact with other reptiles. Infected reptiles periodically shed Salmonella bacteria in their feces. While treating animals with antibiotics may seem a logical first line of defense in preventing transmission, attempts to eliminate Salmonella in reptiles (as is true for humans as well) with antibiotics have been unsuccessful and can promote the selection of resistant strains.