Salmonellosis in Horses
Salmonellosis (see Salmonellosis) is one of the most commonly diagnosed infectious causes of diarrhea in adult horses. Clinical manifestations range from no abnormal clinical signs (subclinical carrier) to acute, severe diarrhea and even death. The disease is seen sporadically but may become an epidemic, depending on the virulence of the organism, level of exposure, and host factors. Infection can occur via contamination of the environment, feed, or water or by contact with animals actively shedding the bacteria. Stress appears to play an important role in the pathogenesis—a history of surgery, transportation, or change in feed; concurrent disease, particularly GI disorders (colic); or treatment with broad-spectrum antimicrobial drugs often precedes the diarrhea.
Salmonella enterica of the serogroup B includes S enterica serovar Typhimurium and S enterica Agona, two of the most frequently isolated serovars from horses with clinical disease. Knowing the serovar and antibiogram can help track or monitor the type of serovar of salmonellae affecting any given group or population of horses (eg, tracking nosocomial spread within a veterinary hospital). The emergence of multidrug-resistant S enterica isolates is concerning both in dealing with nosocomial infections and zoonosis.
Three forms of salmonellosis have been recognized in adult horses. One is the subclinical carrier, which may or may not be actively shedding the organism but has the potential to transmit the bacteria to susceptible animals either by direct contact or by contamination of the environment, water, or feed sources. Multiple fecal cultures may be necessary to identify carriers, because the organism is shed in the feces intermittently and in small numbers. If stressed, the carrier may develop clinical disease. The national prevalence of fecal shedding of S enterica by normal horses in the USA is estimated to be <2%; however, the proportion of hospitalized horses shedding is much higher (~8%). The most common serovars identified among the general population of horses were S enterica Muenchen and S enterica Newport (both serogroup C2).
The second form of the disease is characterized by a mild clinical course, with signs of depression, fever, anorexia, and soft but not watery feces. Affected horses may have an absolute neutropenia. Clinical disease may last 4–5 days and usually is self-limiting, and S enterica can be isolated from the feces. Recovered horses may continue to excrete the organism in their feces for days to months; therefore, isolation of the shedding horse and thorough cleaning and subsequent disinfection of the contaminated area are recommended.
The third form of salmonellosis is characterized by an acute onset of severe depression, anorexia, profound neutropenia, and frequently abdominal pain. Diarrhea develops 6–24 hr after the onset of fever; feces are fluid and foul smelling. Affected horses dehydrate rapidly, and metabolic acidosis and electrolyte losses occur as the horse deteriorates. Clinical signs of sepsis and hypovolemic shock can progress rapidly. There may be signs of abdominal discomfort, straining, or severe colic secondary to ileus, gas distension, and colonic inflammation and possible infarction. Protein-losing enterocolopathy can occur with plasma protein concentrations becoming dangerously low (albumin <2 g/dL) after a few days of diarrhea. These horses can become bacteremic because of bacterial translocation of enteric organisms, and coagulation abnormalities resulting in disseminated intravascular coagulation can occur. If untreated, this form of salmonellosis is often fatal.
Salmonella bacteremia can occur in neonatal foals, especially from farms with endemic salmonellosis (see Bacterial Diarrhea in Foals).
Diagnosis is based on clinical signs, severe neutropenia, and isolation of salmonellae from feces, blood, or tissues. Submission of 10–30 g of feces for culture has been more successful in identifying salmonellae than has culturing fecal swabs. It is important to collect and submit feces based on the recommendations of the laboratory performing the culture. Working with a diagnostic laboratory that uses enrichment techniques along with agars specifically selected to optimize recovery of S enterica is advisable. Because salmonellae cannot be consistently cultured from feces, multiple daily samples (generally 3–5) should be collected from each horse. Culturing of rectal mucosal biopsies increases the probability of isolating the organism; however, the technique is not without risk to the horse. Fecal samples that must be mailed should be placed in transport media suitable for enteric pathogens at the time of collection and shipped on ice. A PCR test is available and, depending on the primers used, appears to be more sensitive than routine bacterial culture for detection of salmonellae.
Treatment of the severe form of salmonellosis is based on IV fluid and electrolyte replacement and efforts to control the host’s responses initiated by the systemic inflammatory response. A polyionic isotonic fluid is used for volume replacement. Because of active secretion of fluid and electrolytes into the lumen of the intestine, IV fluid volumes of 40–80 L/day may be necessary. Electrolyte and acid-base deficiencies are common and are corrected by use of oral and/or IV fluids supplemented with electrolytes. It is difficult to predict the electrolyte status of affected horses. Deficits should be determined by serum biochemical analysis; supplementation with sodium chloride, potassium chloride, calcium gluconate, magnesium sulfate, and occasionally sodium bicarbonate may be indicated.
Antimicrobial treatment in adult horses with salmonellosis is controversial and does not appear to alter the course of the colitis or decrease shedding of salmonellae; however, it may reduce the likelihood of bacteremia. Selection of an antimicrobial is not easy and should ideally be based on the sensitivity of the organism isolated. Resistance patterns vary among Salmonella isolates and can change over the course of an outbreak. There is potential nephrotoxicosis from aminoglycoside antibiotics in volume-depleted horses; therefore, the hydration status of the horse should be considered when selecting an antimicrobial. The ideal antibiotic should also be lipid soluble.
The use of GI protectants (eg, biosponge, bismuth subsalicylate, activated charcoal) may be beneficial. These substances may bind bacterial toxins. NSAIDs, such as flunixin meglumine, help counteract the effect of endotoxin, control pain, and possibly help prevent laminitis. The dosage of NSAID used has been quite variable. Serious adverse effects, such as gastric and colonic ulceration and renal nephrotoxicosis, can result from NSAID treatment, so the minimum effective dosage should be used. Equine plasma may be administered to correct hypoproteinemia and to supply coagulation factors and, depending on the source of the plasma, specific antibodies to endotoxin and Salmonella. Colloidal plasma substitutes such as hetastarch may be necessary to maintain oncotic pressure in horses with substantial protein loss into the GI tract. These colloidal plasma substitutes may be less expensive and better tolerated than equine plasma in some horses. Often, equine plasma and colloidal plasma substitutes are both used in horses with hypoproteinemia due to colitis.
Low-dose polymyxin B (6,000 units/kg, bid) has also been advocated to bind circulating endotoxin. In controlled trials, polymyxin B ameliorated some of the known effects of endotoxemia in horses. Antimicrobial doses of polymyxin B are substantially higher than the dose used to bind endotoxin and may be nephrotoxic. Low-dose polymyxin B therapy is unlikely to be nephrotoxic in adequately hydrated horses receiving IV fluids.
Prevention of salmonellosis is difficult, because the organism is present in the environment as well as in the feces of some healthy animals. In a hospital environment where horses are stressed, may be off feed, and are often receiving antimicrobial treatment, aggressive identification and strict isolation of salmonellae-infected horses is indicated. Biosecurity practices to minimize cross-contamination between hospitalized horses are also advisable. Serotyping, antimicrobial susceptibility profiles and genotyping by pulse field electrophoresis, plasmid profile analysis, and phage typing can be used to determine whether isolates are genetically related and help determine whether infection is nosocomial.
Owners should be made aware of the zoonotic risk of S enterica infection. People working with infected animals should practice strict hygiene.