An outbreak of equine botulism type A associated with feeding grass clippings

Abstract

In September 2010, an outbreak of type A botulism involved 4 horses in northern California that were fed grass clippings obtained from a nearby park. All 4 animals developed a progressive flaccid paralysis syndrome clinically consistent with exposure to preformed Clostridium botulinum neurotoxin (BoNT). Within 48 hr of consuming the grass clippings, all 4 horses showed marked cervical weakness (inability to raise their heads to a normal position) and died or were euthanized within 96 hr. One horse was submitted for diagnostic examination and subsequent necropsy. At necropsy, extensive edema was observed in areas of the nuchal ligament and inguinal fascia. A sample of the grass clippings tested positive for preformed BoNT type A by the mouse bioassay test. Emphasis should be placed on early case recognition, rapid initiation of treatment with the trivalent antitoxin product, and preventing exposure to BoNT in spoiled forages.

In September 2010, 3 horses and a pony died over a period of 96 hr after being fed grass clippings obtained from a nearby park in northern California. When seen by the owner’s veterinarian on the morning following ingestion of the grass clippings, horse 1—the dominant mare in the pen and presumed to have had the greatest access to the grass clippings—was observed to have profound generalized weakness and muscle tremors and was unable to raise her head. Horses 2 and 3 were observed with similar cervical weakness, including the inability to lift their heads more than 12 in. (30 cm) above the ground.

The animals were housed on sandy ground in a dry-lot pen and had not been on a consistent program of annual deworming or vaccination during the prior 3 years. Due to the ambiguous clinical presentation, all 4 animals were treated with psyllium, an ivermectin paste wormer, and vaccinated against West Nile virus (WNV). Over the course of the next 24 hr, horse 1 became recumbent and unable to rise, and euthanasia was performed approximately 48 hr after having initial access to the grass clippings. Horses 2 and 3 also became progressively weaker and then recumbent over the course of 48 hr; the 2 horses were euthanized approximately 72 hr after ingestion of the grass clippings. Although the pony appeared normal on the first day following ingestion of the grass, cervical muscle weakness was observed on the second day. Intermittent recumbency was noted by the third day postexposure. At this time, the emergency on-call veterinarian referred the case to the Veterinary Medical Teaching Hospital at the University of California, Davis.

The pony presented to the Veterinary Medical Teaching Hospital in right lateral recumbency. Body condition score was estimated to be 4 out of 9 (normal: 4–7 of 9); rectal temperature was 36.9°C (normal: 36.6°–38.3°C); heart rate was 40 beats/min (normal: 38–40 beats/min); and respiratory rate was 12/min (normal: 8–10/min). The mucous membranes were light pink; capillary refill time was 2.5 sec; and jugular refill was assessed as poor. No cardiac murmurs or arrhythmias were auscultated. Breathing was very labored with increased expiratory abdominal effort; no adventitious sounds were noted. Digital pulse in all 4 legs was described as within normal limits. No abdominal borborygmi were auscultated, and signs of ileus, including distended loops of hypomotile small intestine, were evident via abdominal ultrasound. Diffuse fasciculation affecting the triceps, quadriceps, cutaneous trunci, and gluteal muscle groups was present. The patient was alert and responded to sounds by flicking an ear but was dysphagic, as evidenced by an inability to prehend food and a very weak swallowing reflex. Other clinical signs included decreased eyelid tone (inability to blink), weak or absent tongue tone, mildly decreased tail and anal tone, sluggish pupillary light reflexes, inability to urinate and defecate, and occasional episodes of slow and uncoordinated paddling.

A complete blood count revealed mild leukocytosis with mature neutrophilia, mild lymphopenia, and a mild increase in fibrinogen and total plasma proteins. A serum chemistry panel revealed mildly increased total bilirubin, glucose, triglycerides, creatine kinase, alkaline phosphatase, and gamma-glutamyl transferase. Fecal culture was negative for Salmonella spp. Differential diagnoses included botulism; oleander intoxication; organophosphate intoxication; equine viral encephalitis such as WNV, Eastern equine encephalitis virus, Western equine encephalitis virus, and Rabies virus; central nervous system trauma; equine protozoal myeloencephalitis; aberrant larval migration; and hyperammonemia. The clinical signs were considered to be most compatible with botulism intoxication due to the pony’s severe weakness and relatively alert mental state. The history of exposure to common forage and the rapid progression of similar clinical signs in the pony and 3 horses led to the suspicion of Clostridium botulinum neurotoxin (BoNT) intoxication as the most likely cause of the clinical syndrome.

Supportive care provided for the hospitalized patient consisted of nasogastric intubation for gastric lavage and administration of activated charcoal, O₂ insufflation therapy, intravenous administration of lactated Ringer solution with 1% dextrose, intravenous thiamine and other B vitamins, intravenous nonsteroidal anti-inflammatory drugs, protective ophthalmic ointment, and urinary catheterization. Treatment with hyperimmune multivalent botulism antitoxin equine plasma containing antitoxin against BoNT types A–E was discussed with the owner but declined due to financial limitations. A less expensive antiserum containing types B and C antitoxin was administered. The development of increasingly severe clinical signs was unaffected by treatment, and the patient died on the fourth day following ingestion of the grass clippings.

No gross lesions were identified in the brain or spinal cord at necropsy. Moderate to severe focally extensive areas of edema were noted within the fibers of the nuchal ligament and inguinal area muscles and were more severe in the dorsolateral cervical area. Cervical edema was most prominent surrounding both sides of the lamellar part of the nuchal ligament. Tests were negative for Rabies virus by fluorescent antibody assay (performed at the Napa-Solano-Yolo County Public Health Laboratory, Fairfield, CA) and for the major equine encephalitis viruses by real-time PCR performed on the central nervous system at the CAHFS, Davis Laboratory (WNV) and the National Veterinary Services Laboratory, Ames, Iowa (Western equine encephalitis virus and Eastern equine encephalitis virus). Samples of the grass clippings were examined by the CAHFS Toxicology service for the presence of poisonous plants, including but not limited to the genera Taxus (yew), Aconitum, and Oleandra (oleander); no toxic plants were detected. A mouse bioassay (MBA) was performed at the CAHFS San Bernardino Laboratory per the Centers for Disease Control guidelines ² ; samples tested included grass clippings and both gastrointestinal contents and feces obtained from the patient. Gastrointestinal content and feces were both negative, but a grass clipping sample tested positive by MBA for preformed C. botulinum type A toxin.

In arid portions of the western United States, horses are frequently maintained in dry-lot pens without regular access to fresh grass. Both C. botulinum toxin types A and B have been associated with exposure to spores in the environment and to preformed toxin in forages; type C has been associated with decomposing carcasses of small animals, which can be mechanically incorporated in baled or ensiled forages and in processed feeds such as hay cubes.^4,6,11 Although equids are discriminating in their grazing habits, if chopped green feed is provided, horses are hampered in their ability to sort out and reject toxic materials. When fed as a group, the dominant horses eat first and keep lower-ranking herd members away until they have finished. ⁹ In the current outbreak of forage-related intoxication, the dominant horse developed clinical signs first and showed the most rapid progression of the neurotoxic syndrome; herd history provided important information to narrow the differential diagnostics list.

Typical or pathognomonic necropsy lesions are not considered a feature of equine botulism. However, several reports suggest that edema of the head and neck may be a prominent, although inconsistently observed, necropsy finding associated with cases due to types A and C.^7,11 One third (5 of 15) of horses necropsied during investigation of an outbreak of type C botulism in California were reported to have edema of the cervical fascia along the nuchal and supraspinous ligaments, extending caudally as far as the lumbar region. ⁷ Severe edema of the face and muzzle reported in several other clinical cases of type C botulism was described as dependent edema of postural origin and attributed to BoNT-induced weakness affecting the ability of the horses to raise their heads. ¹¹ In the current type A outbreak, the only case available for pathologic examination had extensive edema of the cervical area along the nuchal ligament and between the muscle fibers of the inguinal area. With the clinical history of profound cervical muscle weakness affecting all 4 horses in the current type A outbreak, the observations suggest that the edema is likely attributable to simple hydrostatic mechanisms, secondary to the unnaturally prolonged static posture. It is also possible that other exotoxins produced by C. botulinum contribute to the pathogenesis of the edema; C. botulinum type C1 can produce a C2 toxin associated with increased vascular permeability.^1,3

Laboratory diagnosis of equine botulism is frequently challenging. Although spores of C. botulinum have been detected in fewer than 30% of fecal samples from horses with clinical signs compatible with botulism, identification of spores in gastrointestinal contents or feces is considered presumptive evidence. ¹¹ Definitive diagnosis requires identification of BoNT in samples of feed, gastrointestinal contents, or serum utilizing the MBA, which is the only test officially recognized by the Centers for Disease Control and Prevention. ² Horses are estimated to be between 1,000 and 10,000 times more sensitive than the mice used in the MBA, ⁷ which renders the MBA a low-sensitivity technique for diagnosis of equine botulism. In addition to the low sensitivity of the MBA to detect BoNT in equine specimens, two explanations have been proposed why BoNT is rarely present at detectable levels in equine tissues or fluids: 1) the small amount of BoNT present in the circulation of affected horses is rapidly and irreversibly bound to motor endplate receptor sites at cholinergic nerve terminals, thereby removing it from circulation,^5,6,10,11 and 2) BoNT in the gastrointestinal tract may be rapidly degraded by microbes and their enzymes.^5,10 The MBA remains the gold standard for official case diagnosis but is often of limited value to equine clinicians due to the low sensitivity of this test for equine cases and the length of time required for results. This outbreak of type A equine botulism was typical in that multiple horses died or were euthanized following exposure to a contaminated forage, and the lack of a rapid clinical assay contributed to selection of an antiserum product that was ineffective against the outbreak strain. Submission of samples for MBA eventually provided a definitive diagnosis; however, clinical case management and prognosis had to be based on clinical judgment supported by case and herd history and by the use of standard clinical tests to exclude other differential diagnoses.

The current report stresses the previously reported fact that cases of equine botulism occurring west of the Mississippi River are most likely to be caused by C. botulinum type A rather than type B.^6,11 In both human beings and equids, type B cases predominate east of the Mississippi, whereas type A cases originate almost exclusively from 7 western states (California, Oregon, Idaho, Montana, Nebraska, Washington, and Wyoming).^2,6,11

For decades, the diagnosis and treatment of type A botulism in adult horses has been made difficult by delays in clinical case recognition, lack of a rapid clinical assay to confirm a botulism diagnosis and distinguish serotypes, lack of awareness of the different geographic distribution of type A versus B, and lack of access to an affordable antitoxin product that included type A. Although significant challenges remain, recent developments include a better understanding of the geographical distribution of types A and B^2,6,8 and the availability of an affordable trivalent (A–C) BoNT antiserum product, ^a which provides therapeutic coverage for all three clinically relevant C. botulinum toxin types.^2,6,8 This information can assist practitioners in the selection of effective antiserum products for treatment of equine botulism cases in the western states. Emphasis should continue to be placed on early case recognition, rapid initiation of treatment, and the education of horse owners regarding practical ways to prevent exposure of their animals to BoNT in spoiled forages.