Salmonella Typhimurium Associated with Rodents

Outbreak of Multidrug-Resistant Salmonella Typhimurium Associated with Rodents Purchased at Retail Pet Stores

This submodule provides a case study involving antibiotic use in pocket pets. The module will present an outbreak scenario and highlight the prevalence of antimicrobial use in the pocket pet industry, as well as provide insight into proper antimicrobial use and stewardship in veterinary medicine.

Learning Objectives

This submodule aims discuss the outbreak of multidrug-resistant items purchased at retail pet stores. By the end of the module, you will be able to:

  1. Identify how antibiotics are frequently used in the pocket pet industry and how the use of antibiotics in pocket pets may select for antimicrobial resistance.
  2. Discuss how animal and human health may both be impacted by antimicrobial resistance. 
  3. Describe how veterinarians in agriculture departments work with veterinarians and other health professionals in local, state, and federal health departments to investigate and control zoonotic disease.
  4. Summarize how the use of antibiotics may select for the dissemination of gene cassettes conveying resistance to multiple antibiotics.

Index Case

Based on actual events as reported in the Morbidity and Mortality Weekly Report (MMWR), May 6, 2005 Volume 54 Number (17); pages 429−433.

light bulb icon

Now it's time to Check Your Understanding of Public vs. Private Vet Interactions.

You have recently been hired as the public health veterinarian at the Minnesota Department of Health. It’s only your second week on the job, and you are assigned the following case:

A five-year-old boy named Sam was ill for 14 days with severe diarrhea, stomach cramps, fever, and intermittent vomiting. Occasionally the diarrhea was bloody. When Sam’s mother, Lucy, took him to the doctor, the doctor treated Sam with oral fluids and gave Lucy instructions to give Sam plenty of fluids. The doctor also told her to call back if there was any change in Sam’s condition. Upon returning home from the clinic, Lucy found that Sam’s pet mouse (Melvin) had died. The mouse had been suffering from diarrhea and lethargy since they bought it at a pet store a few days before. Lucy wondered if Sam’s illness and Melvin’s illness were related. She contacted Sam’s pediatrician and the pet store where she bought Melvin. Staff at both locations told her that it sounded like Sam’s illness and Melvin’s illness were a coincidence.

With Lucy’s encouragement, the pediatrician agreed to submit Sam’s stool specimen to a clinical laboratory. A couple days later, the clinical laboratory reported that the stool specimen yielded Salmonella. The clinical laboratory forwarded the Salmonella isolate to the Minnesota State Public Health Laboratory and reported the case to the Minnesota Department of Health. The Minnesota Department of Health attempts to interview all people with a laboratory-confirmed Salmonella infection. As part of this routine surveillance, you contact Sam’s mom to inquire about possible sources of infection. During the course of your conversation, Lucy wonders if Sam’s illness could be related to the recent death of Sam’s pet mouse. You share your knowledge of Salmonella epidemiology with Sam’s mother:

Nontyphoidal Salmonella includes common serotypes such as: S. Typhimurium, S. Enteritidis, and S. Newport. Most mammals, birds, and reptiles are the reservoirs for nontyphoidal Salmonella. This means that most infections can ultimately be traced back to animals. The immediate source for human infections is usually food, but people can also become infected by direct ingestion of animal or human feces, usually by contaminating their hands and then putting their fingers in their mouths. Human salmonellosis usually causes self-limiting gastroenteritis, but can sometimes cause severe infections and even death, especially if the infection becomes systemic.

Initial Questions

light bulb icon

Now it's time to Check Your Understanding of Needed Information.


You learn that the Minnesota State Public Health Laboratory has indeed received the isolate from the clinical laboratory, serotyped the isolate as S. Typhimurium, and found the isolate to be resistant to ampicillin, chloramphenicol, streptomycin, sulfizoxazole, and tetracycline. The Minnesota State Public Health Laboratory will also do PFGE analysis. The PFGE results will be available at the end of the week. What other questions would you like to ask Lucy?


light bulb icon

Now it's time to Check Your Understanding of Outbreak Response and Treatment 1.


Disembowled mouse

Necropsy of a pet mouse showing fecal staining and an enlarged spleen.

Hamster lungs

Hamster lungs with gross lesions of hemorrhagic, suppurative pneumonitis. Photo: Arno Wünschmann, DVM, University of Minnesota


These lungs show clear evidence of a raging septicemia

The next day, you learn that the Salmonella isolate from Melvin (the mouse) is serotype Typhimurium—the same as the Salmonella isolate from Sam. However, you know that Typhimurium is a very common serotype and therefore this is not strong evidence linking the boy’s infection with that of his mouse. The antibiotic resistance pattern can also be used to determine if strains may be related. You note that the antibiotic susceptibility results are the same for both S. Typhimurium isolates. The laboratory results on Melvin isolate show resistance to ampicillin, chloramphenicol, streptomycin, sulfizoxazole, and tetracycline (R-type ACSSuT). This was the same resistance pattern in the boy’s isolate, and therefore provides some evidence that links the infections in the boy and his pet mouse.

You are also awaiting a call from the laboratory regarding the PFGE results from Sam’s isolate so that you can compare the PFGE pattern to the isolate from Melvin. The PFGE “fingerprint” is used to determine if the isolates’ DNA are related. Question: What would you do while you are waiting for the PFGE pattern from Melvin’s isolate?


The PFGE information on Sam’s isolate is now available. The PFGE pattern matches an isolate from an ill person in Kentucky and another from South Carolina; both of the patients report recent exposure to rodents. This PFGE pattern was uncommon, representing 23 of 17,737 isolates in the S. Typhimurium database. These three human cases might represent a common source outbreak.

Pulsied-field gel electrophoresis

Pulsed-Field Gel Electrophoresis Patterns of Salmonella Typhimurium Isolates with 3 sources (MN distributor hamster, MN human, a


You still do not know if the isolate from Melvin (the mouse) is related to the isolate from Sam (the child). The antibiotic resistance pattern is the same, but you do not know yet if there is a PFGE match between Sam and Melvin’s Salmonella isolates.

Even more interestingly, the CDC PulseNet search finds a matching PFGE pattern from a Salmonella outbreak that occurred this last month in another state. You call the state health department in this state and are surprised to learn that the outbreak was among workers at a pet store. Furthermore, some of the pet rodents at the pet store had been sick, and specimens from sick rodents had yielded Salmonella Typhimurium with the same antibiotic resistance pattern and PFGE pattern as the workers and as the isolate from Sam. Now you’re getting somewhere!

You call the public health veterinarian who investigated the outbreak at the pet store and who submitted the specimens from the rodents. The veterinarian tells you that the pet store obtained the rodents from a large distributor who had experienced a major die-off of rodents after receiving a shipment of 780 mice from Iowa. Diarrhea was present in the majority of ill mice. Distribution of the animals to pet stores was halted, but not before 243 of the original 780 mice had been sent to various pet stores where they were sold to the public. Of the remaining mice, 60 percent had died by the end of the month and the rest were humanely euthanized.

Source Identification

light bulb icon

Now it's time to Check Your Understanding of Outbreak Response and Treatment 2.

The PFGE results are in! You find that the PFGE pattern from Melvin’s Salmonella is indistinguishable from the pattern from Sam’s isolate. Furthermore, this PFGE pattern turns out to be rare in the PulseNet database, and so it is therefore likely that Melvin and Sam’s cases are related.

Control Measures

It is now time to move backward on the timeline. Where would a mouse get a Salmonella infection that is resistant to multiple antibiotics? What is your next step? Should you search to see if the other human cases with this particular PFGE pattern had rodent exposure prior to getting sick? This type of a study would be a good idea. Maybe there are more than three human cases involved. However, at this point you have gone beyond the bounds of your state and need to be working with other state health departments and with the CDC. The PulseNet search tells your investigative team that there have been 23 human cases of S. Typhimurium with this PFGE pattern since December of the previous year. CDC facilitates a conference call among state health officials from states that have cases with the key PFGE pattern. It is agreed that all cases would be interviewed. Of those interviewed, 13 of the 22 interviewed had pocket pet (small exotic animal) exposure, all within eight days of onset of symptoms. An additional two cases of the 22 interviewed contracted salmonellosis as a result of secondary exposure—infection from exposure to other people who originally got the infection from a pet rodent.

Additional Cases

The remaining seven (32 percent) had no recorded rodent exposure. This means that 68% of the matching human cases of S. Typhimurium in the database could be epidemiologically linked to the outbreak in the pocket pet industry.

PFGE results

The chart above shows that there were 28 cases of human S. Typhimurium isolates with outbreak PFGE patterns, of which 22 were interviewed. Of the 22 interviewed cases, 13 had pet rodent exposure (59%), 2 were secondary cases (9%), and 7 had no reported rodent exposure (32%).

You find that the type of pocket pet exposure was varied. Seven cases were linked to mice or rats used for feeding snakes, four cases were linked to pet mice or rats, and two more cases were linked to pet hamsters. The human cases are mapped out in order to see if there is a geographic cluster.

Map of rodent-associated human exposure cases by state

Number of rodent-associated human cases by state, reporting from 15 cases. There was one case each in Minnesota, Michigan, New Jersey, North Carolina, and Georgia. There were two cases each in Illinois, Missouri, Kentucky, Pennsylvania, and South Carolina. Once you map it, you see that this is a multifocal event, and you look for ways to explain how this same organism caused illness in many different places across the Eastern United States. You are also quite impressed that this outbreak now involves many people across the country, not just one child in Minnesota.

Pocket Pet Industry

In collaboration with CDC, you work with your state department of agriculture to determine the origin of the rodents, to learn more about the pocket pet industry and to learn how animals are raised and distributed. You also want to know how this particular Salmonella came to be resistant to so many antibiotics.

Let’s see what you know about the pocket pet industry.

Rodent Antibiotics

Information on use of antimicrobials was obtained from five rodent breeders/distributors. Routine use of antimicrobials was documented in four facilities in which they were used for prevention of nonspecific rodent enteritis. Delivery of antimicrobial agents in drinking water occurred at weaning, before transport, and/or on arrival at the pet distributor. One pet distributor exclusively used rodent feed containing tetracycline.

An Arkansas small rodent breeder was routinely using both spectinomycin and another unknown antibiotic for animals that appeared stressed or had “wet tail” (nonspecific enteritis). The Iowa distributor linked to the original human outbreak administered spectinomycin for three days to all rodents upon arrival at the facilities; it was also given for three days prior to shipment. The Minnesota distributor did not use antimicrobials prophylactically, and unlike all the other surveyed breeders and distributors, worked closely with a veterinarian. The Georgia distributor fed modified pig feed containing tetracycline to all of their animals.

Salmonella & Antibiotics

Antimicrobial agents are often ineffective at alleviating clinical symptoms of salmonellosis. Antibiotics also can prolong shedding. The dissemination of multidrug-resistant Salmonella in pet rodents might have been facilitated by the use of prophylactic antimicrobials within the pocket-pet (e.g., hamsters, mice, and rats) animal industry. This use might have contributed to disease transmission in colonized animals and increased shedding of Salmonella. As such, the use of antibiotics in raising these animals may have increased transmission among animals, from animals to their human caretakers, and from animals to the people who purchased these animals as pets. Appropriate animal husbandry and hygiene practices, such as those used for the production of antibiotic-free laboratory rodents, could eliminate the need for nontherapeutic antimicrobials to prevent disease in rodents.

Public Awareness

Phage typing (another way to characterize Salmonella) of the S. Typhimurium isolates identified the isolates as Definitive Type 120 (DT120). S. Typhimurium DT120 is often multidrug-resistant and may possess antibiotic-resistance gene cassettes also found in S. Typhimurium DT104. Selective pressure for any one of these five antibiotics in the cassette selects for the entire cassette of resistance genes.

Antibiotic Recap

light bulb icon

Now it's time to Check Your Understanding of Antibiotic Resistance in Pocket Pet Industry 1.

You take stock of the current information on antibiotic use in the rodents:

  1. It is not known whether the infectious agent(s) involved in “nonspecific rodent enteritis” are bacterial, viral, or environmental. Diet and stress could be contributing factors with this syndrome.
  2. The administration of antimicrobials for a Salmonella infection may lead to prolonged shedding of organisms, which would tend to increase transmission to both other rodents and to people.
  3. Every time an antimicrobial agent is administered, there is a selective pressure for the survival of resistant strains. It would be helpful if outbreaks of nonspecific rodent enteritis were cultured and antimicrobial susceptibility tests run on targeted pathogens before administration of an antimicrobial agent.
  4. Salmonella infections may be controlled but not always eliminated when antibiotics are administered to infected rodents. Customers who purchase animals do not continue to administer these antibiotics; therefore, the animal may become clinically ill soon after being purchased. A general rule with Salmonella is that clinically ill animals tend to shed in greater numbers than do animals with subclinical infections.

Traceback Diagram

The investigative team of agriculture and health department veterinarians provide the following information regarding the breeders, the distributors, and the particular animals that were owned by the human cases.

Traceback diagram - rodents assoc w Salmonella Typhimurium outbreak

The tracebacks of all infected human cases led to the original outbreak occurring from three distributors: GA, AR, and IA. Multiple small rodent breeders supplied the GA and AR distributors, whereas an AR breeder specifically supplied the AR and IA distributors. That IA distributor shipped rodents to an IL distributor that supplied a MI distributor, which distributed to MI pet stores and led to one MI human case. It also shipped to a MN distributor that supplied MN pet stores and led to one MN human case. The AR distributor shipped to a pet store in IL-1 that led to human cases IL-1 and IL-2, a pet store in NJ that led to one NJ human case, a pet store in SC-2 that led to one human case called SC-2, and a pet store in MO that led to one human case and a second human case due to infection from the first human case. The GA distributor shipped to a pet store in KY-1 that led to one human case called KY-1, a pet store in KY-2 that led to one human case called KY-2, a pet store in GA-1 that led to one human case, a pet store in SC-1 that led to one human case called SC-1, and a pet store in GA-2 that led to one human case and a second human case due to infection from the first human case. There was also a NC distributor that had no contact with breeders related to the other distributor, but that distributed to a pet store in NC and led to one human case.

Veterinarians from the Minnesota Department of Agriculture were able to visit two of the implicated distributors. The Minnesota distributor had thoroughly cleaned and disinfected the facilities since the outbreak. The environmental samples were negative for Salmonella. At the Georgia facilities, environmental swabs yielded isolates that were indistinguishable from each other and closely related to the PFGE results from Sam, Melvin, and the Minnesota distributor. All of the isolates showed the same antibiotic-resistant cassette (susceptibility profile).

The facilities had no standard cleaning or sanitation procedures. The managers claimed that it is hard to convince employees to thoroughly clean so many cages, primarily due to time constraints. However, you recall that laboratory animals destined for research laboratories can be maintained free of Salmonella without the need for antibiotic treatments.

You also find that at least one retail pet store has a corporate policy stating that their pocket pet providers cannot use antimicrobials in their animals. Despite this policy, they still purchased animals from several producers that routinely used antimicrobials. This particular chain reportedly told customers that the rodents they purchased were likely to have diarrhea when taken home, and that the customers should use spectinomycin to treat the animals at home.

light bulb icon

Now it's time to Check Your Understanding of Antibiotic Resistance in Pocket Pet Industry 2.

Module Summary

  • There is clear evidence that pocket pets can serve as vehicles for transmission of Salmonella.
  • Antimicrobial use in the pocket-pet industry is common. Antibiotic usage contributes to the selective pressure for the development and spread of multidrug-resistant organisms. Close contact between pocket pets and people—especially children that are generally more susceptible to enteric diseases and less likely to wash their hands—increases the risk that pocket pets will transmit their infections to people. Subsequent to this outbreak investigation, pocket pets must be considered a potential source for multidrug-resistant Salmonella infections.
  • Producers/breeders of pocket pets should use diagnostic testing for nonspecific enteritis rather than using antibiotics just in case bacterial agents may be involved.
  • Good husbandry and sanitation practices could help prevent transmission and eliminate the need for prophylactic antimicrobial treatment within the pet rodent industry.
  • The prophylactic use of antimicrobials most likely contributed to multidrug-resistant Salmonella spread among the pet rodent breeding populations. Antimicrobials may prolong and mask the clinical signs of Salmonella infections and cause extended shedding of a higher number of organisms. This could result in an increased dissemination of the infection through a facility.
  • Salmonella is a common enteric pathogen that is easily spread among most animals species, including humans, rodents, primates, birds, reptiles, and ruminants. It can readily develop resistance, and therefore inappropriate and unnecessary use of antimicrobials in raising animals is discouraged.


CDC: Outbreak of Multidrug-Resistant Salmonella Typhimurium Associated with Rodents Purchased at Retail Pet Stores

New England Journal of Medicine: Multidrug-Resistant Salmonella enterica Serotype Typhimurium Associated with Pet Rodents