Resolution of neurologic signs presumed to be associated with hyperammonemia in 2 endurance horses

Abstract

This case report describes 2 endurance horses with non-hepatic hyperammonemia. The animals were competing in a 160-km endurance competition in extreme heat conditions and were presented for obtundation. One of the horses also had evidence of blindness. The blood ammonia concentration was elevated (196 μmol/L and 249 μmol/L) and both horses improved following treatment with intravenous fluids and supportive care. These are the first documented cases of clinical signs presumed to be associated with hyperammonemia in endurance horses. Despite the severity of the clinical presentation, both horses made a full recovery.

Key clinical message:

Non-hepatic hyperammonemia should be considered as a potential cause of blindness and obtundation in competing endurance horses. Horses appear to respond well to treatment with intravenous fluids.

There is increasing recognition of hyperammonemia and associated clinical signs in horses presented to referral centers for emergency evaluation and treatment (1–5). In human critical care, there is a large number of causes of hyperammonemia including liver failure, inborn errors of metabolism, drug toxicity, and idiopathic causes (6). In horses, hyperammonemia has been described in association with liver failure and gastrointestinal disease, but a genetic disorder has also been suspected in Morgan horses (2,4,5,7,8). Similar to humans, there may be additional causes of hyperammonemia in horses.

Prolonged exercise activity has been associated with transient hyperammonemia in humans (9,10). The severity of exercise-induced hyperammonemia is associated with heat stress in multiple studies (10,11). Plasma ammonia concentrations have been evaluated in exercising horses both during endurance races and at higher speeds (12–15). Reported values suggest that ammonia may increase to over 200 μmol/L during exercise but return to baseline values within 30 to 60 min (16).

Competing endurance horses undergo prolonged exercise under extreme heat stress; as a result, they commonly exhibit exhaustion and decreased mentation. Neurologic signs have been noted in a subset of these exhausted horses, and they have been hypothesized to be due to dehydration or cerebral edema, although the cause has yet to be documented (17,18). Hyperammonemia was suspected in 1 horse with severe neurologic signs at an endurance competition, but the diagnosis was not confirmed (19). To the authors’ knowledge there are no confirmed reports of hyperammonemia in exhausted endurance horses presenting with abnormal neurologic signs. The purpose of this report was to describe 2 horses with severe hyperammonemia and neurologic signs requiring emergency care following endurance competition.

Case descriptions

Case 1

An 11-year-old Arabian mare was presented to the finish line racecourse field treatment center for emergency treatment following successful completion of a 160-km endurance race under high heat conditions (> 32°C for a prolonged period). The owner noted that the horse was stumbling and had an abnormal gait 1 h after completion of the race. The mare was quickly moved to a paddock enclosure at the treatment center due to concerns for the safety of horse and rider. Previously the same year, the mare had attempted two 80-km competitions but was eliminated from 1 of these races.

The mare was found to be ataxic and obtunded. She was poorly responsive to external stimuli and tripped over a metal rail on the ground. When entering the paddock, her head collided with the panel fencing as though not visual or aware of the presence of the panel. Pupillary light responses (PLRs) were absent OU (direct and consensual were lacking) with normal sized to mildly dilated pupils. Menace and dazzle responses were also absent in both eyes. The mare continued to appear obtunded and would hang her head next to the owner unless stimulated. She continued to hit the paddock fencing whenever she moved forward.

A rectal temperature was not taken at the initial examination due to safety concerns, but approximately 1 h after the initiation of treatment it was 38°C. The mare’s heart rate was 48 beats/min (bpm). Her mucous membranes were pink and she had a capillary refill time of 2 s. She had reduced gastrointestinal borborygmi. She did not pass feces during the first 3 to 4 hours of treatment.

Shortly after arrival at the field treatment center, an intravenous catheter was placed, a heparinized blood sample collected, and 2 L of intravenous hypertonic saline were administered at a rate of approximately 10 L/h. A plasma electrolyte panel was run on-site with a commercial point-of-care (POC) analyzer (i-STAT Handheld analyzer; Abbott Point of Care, Princeton, New Jersey, USA); values are shown in Table 1. An additional blood sample was collected and placed in both EDTA and heparinized tubes that were kept on ice. The sample was transported to a local veterinary hospital (~10 miles away) for a complete blood (cell) count (CBC), plasma biochemical profile, and plasma ammonia concentration (Catalyst DX Chemistry analyzer; IDEXX Laboratories, Westbrook, Maine, USA.). The plasma ammonia concentration was 196 μmol/L [reference interval (RI): 0 to 90 μmol/L]. The creatine kinase (CK) and aspartate aminotransferase (AST) concentrations were above the quantifiable range for the commercial chemistry analyzer. These results are shown in Table 1.

Table 1.

Laboratory values for 2 horses presented for emergency treatment at a 160-km endurance ride.

	Horse 1	Horse 1	Horse 2
Parameter	Admission	Re-check at 36 h	Admission	Ref Range
Na (mmol/L)	129	129	131	128 to 142
K (mmol/L)	3.3	3.6	3.4	1.9 to 4.1
Cl (mmol/L)	97	97	98	100 to 111
HCO3− (mmol/L)	26	27	20.4	24 to 30
NH3+ (μmol/L)	196	1	249	0 to 90
Albumin (g/L)	30		34	22 to 37
AST (U/L)	> 2000		899	175 to 340
BUN (mmol/L)	11.8		11.1	2.5 to 8.9
Calcium (mmol/L)	3.3		3.4	2.9 to 3.6
Creatinine (μmol/L)	150.3		97.2	53.0 to 194.5
GGT (U/L)	14		14	5 to 24
Glucose (mmol/L)	4.3		7.3	3.6 to 6.1
Total bilirubin (μmol/L)	56.4		75.2	8.6 to 39.3
Total protein (g/L)	54		61	57 to 80
Globulin (g/L)	24		27	27 to 50
CK (U/L)	> 15 000		3350	120 to 470
PCV %	38			32 to 48
WBC (cells/μL)	8.4			5.4 to 13.3
Neutrophils (cells/μL)	6.4			2.3 to 9.5
Lymphocytes (cells/μL)	1.6			1.5 to 7.7
Platelets (cells/μL)	56			100 to 400

AST — aspartate aminotransferase; BUN — blood urea nitrogen; GGT — gamma-glutamyl transferase; CK — creatine kinase; PCV — packed cell volume; WBC — white blood cell.

Based on the clinical and biochemical findings, the horse was given a tentative diagnosis of non-hepatic hyperammonemia (NHH). Treatment was continued with a commercially available acetated fluid at approximately 10 L/h for the next 4 h. The fluids were supplemented with 20 mEq KCl/L for the first 20 L of fluids. The first 10 L of intravenous fluids also had 50 ml/L of 23% calcium gluconate added to the fluids. The mare was administered flunixin meglumine, 1 mg/kg body weight (BW), IV, following her first passing of urine, which was grossly normal in color.

Neurologic re-evaluation after 4 h of treatment found the mare to have improved mentation and gradually improving PLRs bilaterally. She was now lethargic as opposed to obtunded. She began to eat small amounts of hay and was producing urine regularly. The intravenous fluids were gradually discontinued over approximately 8 h. The mare’s mentation gradually normalized. The following day, the mare was transported from the racecourse treatment center to a local farm approximately 1 h from the treatment hospital where she continued to appear bright and eat and drink. Approximately 36 h after treatment ended, a physical examination and a neurologic evaluation were performed. At that time the mare had a body temperature of 37.6°C, heart rate of 40 bpm, and respiratory rate of 24 breaths/min. She had good gastrointestinal sounds in all auscultable quadrants and was passing normal feces regularly. She had normal PLRs (direct and indirect), but still had an inconsistent menace response in both eyes. The response was most consistent over the nasal visual fields. A recheck of plasma NH₃⁺ and electrolyte concentrations showed that the plasma NH₃⁺ level had returned to normal (Table 1).

Two days later the horse was transported approximately 1000 miles home. Follow-up reports found the horse to appear clinically normal. The horse returned to endurance competition approximately 12 wk later.

Case 2

An 8-year-old Shagya Arabian mare was presented for emergency treatment to a racecourse field treatment center after completing 107 km of a 160-km endurance race in extreme heat conditions (maximum ambient temperature was ~32°C) similar to Case 1. Previously the same year, the mare had completed five 80-km races finishing as one of the top 10 horses in all of these competitions.

On presentation, the mare’s rectal temperature was 37.8°C and heart rate was persistently increased at 64 bpm. The mare appeared quiet with a decreased responsiveness to her surroundings. Her mucous membranes were pink with a capillary refill time of 3 s. Gut sounds were absent in all quadrants. Based on a clinical examination, the mare was eliminated from competition and treatment was recommended.

Treatment was started with a commercially available acetated intravenous fluid, but the horse showed only minimal change in her demeanor with rehydration. After administration of 20 L of intravenous fluids over 3 h at a field checkpoint, the horse was transported to the finish line of the race where a larger field treatment center was located. The horse exhibited decreased mentation (obtundation) and partially collapsed while standing in the treatment center. A heparinized blood sample was collected and analyzed using a POC analyzer on site (i-STAT Handheld analyzer; Abbott Point of Care). The plasma electrolyte profile is shown in Table 1. The heparinized sample was kept on ice and blood ammonia concentration was measured at a laboratory approximately 10 miles away and was 249 μmol/L (Catalyst DX Chemistry analyzer; IDEXX Laboratories). A plasma biochemical profile was also submitted and the results are shown in Table 1. A CBC was not available for this animal. Based on these results, a tentative diagnosis of NHH was made.

Treatment of the horse continued at the racecourse treatment center with intravenous fluids with a total of 45 L administered over 10 h. Due to the high incidence of ileus and gastrointestinal reflux in endurance horses, a nasogastric tube was passed after the horse was administered 150 mg xylazine intravenously. A total of 6 L of net reflux was obtained, but no additional reflux was obtained thereafter. The mare was also administered flunixin meglumine, 0.5 mg/kg BW, IV.

The horse’s clinical signs gradually improved over the next 12 h. She began to eat and drink normally and her gut sounds returned to normal. She was eventually discharged into the care of the owner. A follow-up blood ammonia concentration was not available for this horse. The owner reported that the mare did not have any recurrence of problems during the 2 y following the event.

Discussion

The 2 horses in this report were presumptively diagnosed with NHH that was potentially associated with endurance exercise and heat stress. Previously, an endurance horse with neurologic signs was identified after elimination from competition and hyperammonemia was speculated as a potential cause, but never proven (19). Both horses in the current report responded well to treatment with intravenous fluids and in Case 1, the ammonia concentration was documented to have returned to normal by the following day. It is interesting to speculate that some of the decreased mentation observed in fatigued endurance horses could be related to other cases of hyperammonemia. At present, these signs of decreased mentation are typically attributed to exhaustion and dehydration.

Non-hepatic hyperammonemia (NHH) has been described in horses and is predominantly associated with gastrointestinal disease (1). It is believed that enteritis and/or ileus may contribute to increased ammonia production within the gastrointestinal system (1). Colic and ileus are common problems in endurance horses and it is possible that the hyperammonemia observed in these 2 horses could be related to ileus (20). They were both noted to have decreased gut sounds which would be consistent with ileus. Horse 2 had a small amount of net gastrointestinal reflux which would also be consistent with ileus. However, neither horse had overt signs of colic that has been reported commonly in cases of hyperammonemia associated with gastrointestinal disease (5). It is important to note that many of the reported cases of hyperammonemia associated with gastrointestinal disturbances had ammonia values considerably higher (370 ± 287 μmol/L) than the values reported herein (5).

Treatment of NHH primarily focuses on decreasing the blood concentration of NH₃⁺. In both cases, continuous isotonic intravenous fluids were used to restore hydration and create diuresis. Additionally, in Case 1 the fluids were supplemented with potassium in order to help mitigate the clinical signs of hyperammonemia (21). As the horses were improving with treatment, additional medications such as lactulose and neostigmine were not used. Lactulose helps to acidify the colon making it less friendly to urease-producing bacteria (22). Lactulose also promotes the formation of NH₄⁺ thereby decreasing passage through the blood-brain barrier (22). Another treatment that has been described for NHH in horses and humans is sodium benzoate; this could be considered for cases not responsive to intravenous fluids with potassium supplementation (22.23). Sodium benzoate binds with glycine to form hippurate, which is then excreted by the renal system (22).

It is not possible to completely exclude hepatic failure in these 2 horses, but the clinical findings and biochemical profiles were not supportive of this diagnosis. Both patients had an increased total bilirubin concentration; however, both animals were anorexic at the time of presentation. Conjugated bilirubin values were not available from the machine used for analysis. The gamma-glutamyl transferase (GGT) values for both animals were within the normal range. The aspartate aminotransferase (AST) values were increased; however, the creatine kinase (CK) values were also increased suggesting that the AST changes could be consistent with muscle damage. Increases in CK and AST are common in endurance horses particularly following difficult rides (24). Due to the field conditions, an ultrasound of the liver and biopsy was not possible in either case.

The horse in Case 1 had more marked increases in muscle enzyme values. However, she did not show typical clinical signs of rhabdomyolysis (muscle pain and swelling) and her urine was not grossly discolored. Significantly increased values of CK and AST previously have been noted in competing endurance horses without obvious clinical signs (24). A follow-up biochemical profile the following day would have been ideal in this case, but was not performed due to the improving clinical signs and the owner’s financial limitations. Exercise and muscle damage have been associated with increased ammonia concentrations and it is possible that this may have played a role in the hyperammonemia in these horses (25). However, the authors have previously tested both healthy and clinically compromised endurance horses with muscle damage, but without neurologic signs, and blood ammonia concentrations were normal (CL Fielding and KG Magdesian, personal communication, 2006).

The clinical signs observed in the 2 horses in this report could have been caused by other factors besides the hyperammonemia. Other potential causes of decreased mentation in humans and endurance horses could include dehydration, electrolyte derangements, hypoglycemia, and hyperthermia (26). Total protein concentrations were not increased above the normal range in either horse, suggesting that severe dehydration was unlikely. Electrolyte concentrations were mildly abnormal but the severity would have been unlikely to cause the severe neurologic signs. There was no evidence of hypoglycemia or hyperthermia. Hyperammonemia was the major abnormality identified bio-chemically, and the clinical presentation was similar to that of horses with hyperammonemia described in other studies (4,5).

In humans, blood ammonia concentrations increased from 31 ± 2 μmol/L at rest to 151 ± 15 μmol/L with exercise in hot conditions (10). In humans, plasma ammonia has a half-life of 2.3 ± 1.0 min (27,28). Plasma ammonia values in endurance horses (n = 23) exercising in heat without clinical signs after 160 km had a normal plasma ammonia concentration (median 8.0 μmol/L, range: 7.0 to 10.0 μmol/L) within 1 h of ending exercise (CL Fielding and KG Magdesian, personal communication, 2006). However, it is possible that another transient condition could have caused the neurologic signs and that the hyperammonemia was unrelated.

In conclusion, these case reports represent another differential diagnosis or etiology for NHH in horses presenting for emergency care. A history of endurance competition or prolonged exercise, particularly in severe heat, should warrant testing for hyperammonemia in horses exhibiting neurologic signs. In these 2 cases, treatment with intravenous fluids and time appeared to resolve the clinical signs. CVJ