This article originally appeared in the Winter 2024 issue of EquiManagement. Sign up here for a FREE subscription to EquiManagement’s quarterly digital or print magazine and any special issues.

Hospital and clinic settings are always at risk of Salmonella spp. outbreaks, despite biosecurity practices. During a recent webinar sponsored by Fluxergy, Ben Sykes, BSc, BVMS, MSc, MBA, DACVIM, PhD, FHEA, and Nicola Pusterla, PhD, DACVIM, AVDC-Equine, discussed testing technologies for Salmonella spp.
Salmonellosis can occur in two forms, said Pusterla:
- An acute or chronic gastrointestinal disease or sepsis.
- A subclinical infection with fecal shedding of Salmonella spp. by carrier horses that show no clinical signs.
Affected horses can shed Salmonella for an average of 50 days after clinical disease resolves. He stressed that hospitals and clinics experiencing an epidemic outbreak not only face economic and legal risks but also have an ethical obligation to look out for all patients in the hospital and to mitigate zoonotic risk to staff and horse owners.
Of horses in isolation, Pusterla said 5-15% have Salmonella spp. but rarely does an outbreak come from an isolation patient due to the many precautions taken with those horses. Two to three percent of a clinic population are infected; 3.5-13% of colic patients have Salmonella; and 0.8% of the general population have Salmonella, he added. The biggest risk comes from an outbreak from routine—or the absence of routine—patient surveillance, Pusterla said. To optimize early detection, he recommends aggressive clinic surveillance of both colitis cases and the general hospitalized horse population.
Sample Collection
Veterinarians can obtain biological samples through various means: feces, rectal-anal swabs, gastric reflux, and mucosal biopsy. While there is no definitive best sample type, Pusterla said feces is considered the standard and can effectively detect clinical and subclinical disease.
Whether to use culture or PCR is the next question. Both Pusterla and Sykes reported that enrichment broth is essential for accurately testing enteric organisms, particularly for Salmonella. Enrichment allows Salmonella spp. to grow in a selected enrichment (selenite) broth while limiting expansion of other enteric organisms. It is necessary to allow two to 24 hours in the enrichment broth before evaluation. Pusterla cautioned against extending this time past 24 hours because it allows more microorganisms to grow, thereby making interpretation more difficult. The presence of more microorganisms might also cause inhibition in a PCR setting. Greater capture of Salmonella occurs at 20-24 hours. If you can’t evaluate the sample before 24 hours, place it in a fridge or freeze it to eliminate bacterial expansion. If time is critical, look at the sample early, perhaps at two hours—a positive result confirms a diagnostic suspicion. If it’s still negative early on, test at a later time within 24 hours.
The Accuracy of Culture vs. PCR
Sykes compared the accuracy of culture results vs. PCR testing. With culture of a single sample, sensitivity is low at 27-55%. For two samples, it is 47-68%, and with five samples it is 79-98%. Specificity of a negative result is as high as 100% with true identification of the type of colony grown. Culture has low sensitivity with false negatives, so cases might be missed. Its high specificity and few false positives minimize unnecessary patient impact.
With PCR analysis, sensitivity approaches 100%—in other words, if disease is present, PCR will find it. Because horses are intermittent shedders, it might be necessary to test multiple times to ensure an accurate diagnosis. Specificity is also very high. While PCR is very sensitive and able to pick up the tiniest genomic fragment, if enrichment grows something, it is live with viable organisms and is a true result. PCR’s higher sensitivity has few false negatives to yield greater protection for patients, facilities, and staff, and its high specificity gives few false positives.
Veterinarians can use PCR to screen for positive results, then perform a culture as a sequential step to provide an antibiogram if treating with antimicrobial drugs. Culture results also give historical knowledge and a database as to what’s going on in the population and the individual hospital. This possibly provides information as to the Salmonella outbreak’s origin. Just a few hours is all that’s necessary for it to become a nosocomial infection.
Sample Pooling
To save money, resources, and time, it might be appropriate to pool samples. One option is to pool up to five fecal samples of 10 grams each that are then homogenized since Salmonella spp. are not randomly distributed in feces. The homogenizing process breaks up the clusters, and then an aliquot is inoculated into enrichment broth.
Another option is to pool up to 10 fecal and environmental samples following selective enrichment steps for PCR detection. Sensitivity is similar to culture but not as good as individual PCR analysis. This method reduces costs by 80%, meaning it’s four times less expensive to pool than to test individual samples. It also reduces hands-on work by 60%. If the general frequency of a positive fecal or environmental culture is low in your practice area, pooling might be an attractive way to reduce testing costs.
Test the pooled enrichment broth after 20-24 hours and, if you get a positive result, perform individual analysis on those five or 10 pooled individual samples.
Environmental Testing
Contamination is predictable in any setting where you have a high density of high-risk horses, said Pusterla. “It is just a matter of when,” he noted. Fecal shedding contaminates the environment to increase potential transmission between comingling healthy horses and those with enteric disease. Horses with diarrhea are tested and put into barrier confinement whereas hospitalized subclinical shedders that show no enteric signs are easy to miss.
Pusterla recommends testing high-risk areas, including isolation and ICU stalls and facilities, treatment rooms and surgical suites, as well as stalls of horses with fever, enteric signs, and leukopenia.
Collect multiple samples from the walls, floor, and drain of each stall after it has been cleaned and disinfected. You can do this using 3M premoistened sponges to which material sticks, Q-tip drag swabs, or Swiffer wipes. A Swiffer’s large surface area likely captures more material but is best used with 20-24 hours of pre-enriched prep in buffered pepsin water. Multiple drag swabs can be put into a bag of selenite broth. Then, pool three to five samples from a single individual stall following selective enrichment. If you identify Salmonella spp., then environmental contamination persists, warranting repeat cleaning and disinfection.
Pusterla concluded by saying environmental testing is elective and not an emergency. At his hospital, staff members obtain 50-60 individual environmental samples weekly. These are individually enriched, then 10 enriched samples are pooled and processed for PCR. This means 60 individual samples are condensed to six PCR analyses. Typically, 98.2% will test negative. If there is a positive test in a PCR analysis, then the 10 samples in that pool must be tested individually.
Related Reading
- ‘Mini-Guts’ Help Researchers Study Infectious Causes of Intestinal Illness in Horses
- Horse Owner Knowledge and Use of Biosecurity
- Short-Term Survival and Laminitis in Horses with Acute Diarrhea
Stay in the know! Sign up for EquiManagement’s FREE weekly newsletters to get the latest equine research, disease alerts, and vet practice updates delivered straight to your inbox.