Postanesthetic brachial triceps myonecrosis in a Spanish-bred horse

Abstract

This report describes a case of postanesthetic brachial triceps myonecrosis affecting only the left forelimb of a horse. A fatal unilateral postanesthetic myonecrosis has not been previously reported in the horse. This article describes the factors in the horse’s history, the anesthetic protocol, and the treatment that may have led to this condition.

Equine anesthesia carries a high risk of unexpected death, and about 1 in 100 horses dies from unforeseen causes within 7 d of anesthesia. Cardiac arrest, fractures during recovery, and postanesthetic myopathy (PAM) are the most common perioperative complications (1). Postanesthetic myopathy is a likely consequence of anesthesia combined with positioning of the horse during anesthesia. This is known to have 2 components: low perfusion pressure secondary to hypotension, and reduced cardiac output combined with high intracompartmental muscle pressure (ICMP) in the dependent limb muscles (2–4). Postanesthetic myopathy was found to be a specific cause of perianesthetic death in 7% of horses that died within a period of 7 d after the induction of anesthesia (1,5).

Postanesthetic myopathy generally affects multiple groups of muscles but a localized presentation may also occur (6). In the present case, degenerative changes typical of myonecrosis were localized to the brachial triceps of the left forelimb while the rest of the muscles appeared to be normal. The observed pathological findings were similar to those described by Friend (7) related to postanesthetic myonecrosis. However, there was also massive myonecrosis affecting muscles of the thorax, back, rump, and limbs. To our knowledge, this is the first case in the equine veterinary literature in which a localized unilateral triceps myonecrosis resulted in a fatal outcome. Dodman et al (3) reported a fatal outcome in a horse with similar muscle lesions in both hind limb adductors.

Case description

A 6-year-old Spanish-bred male horse weighing 554 kg was referred to the University Veterinary Teaching Hospital with a localized ventral mandibular fistula produced by periapical abscessation of the 3rd right lower premolar tooth. Rectal temperature, hydration and color of mucous membranes were normal at admission. Auscultation of the lungs and heart did not reveal any abnormalities. Values from hematological and serum biochemical analyses were normal. Tooth preservation was not possible, and it was decided to extract the infected tooth by tooth repulsion after trephination. Penicillin (Penilevel; Laboratorios ERN, Barcelona, Spain), 22 000 UI/kg bodyweight (BW), gentamicin (Gentavex-10; SP Veterinaria, Tarragona, Spain), 6.6 mg/kg BW, and flunixin-meglumine (Niglumine; Calier, Barcelona, Spain) 1.1 mg/kg BW were administered intravenously (IV), and tetanus antitoxin (Suero Antitetánico; Laboratorios Ovejero, León, Spain), 3000 UI was administered intramuscularly (IM) before surgery.

Preanesthesia was carried out with a combination of romifidine (Sedivet; Boehringer Ingelheim, Barcelona, Spain), 80 μg/kg BW, IV and butorphanol (Torbugesic; Fort Dodge, Madrid, Spain), 0.06 mg/kg BW, IV. Anesthesia was induced with ketamine (Imalgene 1000; Merial, Barcelona, Spain), 2.2 mg/kg BW, IV, and diazepam (Valium; Roche Farma, Barcelona, Spain), 0.05 mg/kg BW, IV, then the patient was intubated with a 30 mm cuffed endotracheal tube. The horse was lifted by a hoist and positioned in left lateral recumbency on a padded operation table. The lower left forelimb was pulled ahead of the right one, and both were separated by a rubber pad. The hind limbs were positioned on leg supports with the upper limb parallel to the lower one and levelled at the point of the hip. Anesthesia was maintained with halothane (Fluothane; Astra Zeneca SA, Madrid, Spain) in 100% oxygen, delivered through a large animal circle breathing system (VLM; Matrx, Montana-Vermala, Switzerland) attached to a large animal ventilator (Racher model 2800; Mallard Medical, California, USA).

The status of the patient was assessed clinically by observing the color of mucous membranes, capillary refill time, pulse quality and palpebral reflexes, and by an instrumental standard monitoring with ECG, oxygen saturation of hemoglobin (SpO₂), end expired CO₂, and percentages of inspired and expired halothane. Direct blood pressure measurements were also performed. Polyionic fluids (LacRinger; Braun, Barcelona, Spain) were infused IV at an approximate rate of 10 mL/kg/h. The horse was under anesthesia for 105 min. The heart rate was almost constant at 32 beats/min and no ECG or pulse abnormalities were noticed. The inspired halothane concentrations were maintained in a range of 3% during the first 20 min of anesthesia; the mean arterial blood pressure (MAP) was above 100 mmHg at this time. The inspired halothane concentrations were progressively reduced until 2% was reached. Ocular signs of a light plane of anesthesia and an increase in the mean arterial blood pressure values were evident at 40 min. The inspired percentage of halothane was increased to 2.5% and a bolus of ketamine (Imalgene 1000; Merial), 100 mg and midazolam (Dormicum; Roche, Barcelona, Spain), 5 mg was administered to the patient IV. The inspired percentage of halothane was maintained at 2.5%, while the MAP values remained above 75 mmHg during the rest of the procedure.

Spontaneous ventilation was allowed during the first 30 min of anesthesia. During this period, the end expired CO₂ was below 60 mmHg; however, an arterial blood gas analysis revealed a PaO₂ of 86.7 mmHg, a PaCO₂ of 79.3 mmHg, and a pH of 7.167. Mechanical ventilation was initiated at a rate of 8 breaths per min to improve the oxygenation and to control hypercapnia. Fifteen minutes later, another gas analysis revealed a PaO₂ of 158.3 mmHg, a PaCO₂ of 67.3, and a pH of 7.236; therefore, mechanical ventilation was maintained until the end of anesthesia.

The horse was transported to a padded recovery room and placed in left lateral recumbency. The orotracheal tube was removed and the patency of the nasal passages confirmed. Oxygen was delivered to the patient at a rate of 15 L/min via a nasopharyngeal tube. A dose of xylazine (Xilagesic 20%; Calier, Barcelona, Spain), 0.2 mg/kg BW was given IV to prevent excitement. The horse remained in lateral recumbency for an hour. Then, the horse moved to sternal recumbency, and after 10 min stood up at the first attempt.

At 2 h after recovery from anesthesia the horse was observed to be moderately lame on the left forelimb, with a dropped elbow and flexed carpus. After 4 h, the horse demonstrated increasing discomfort, pain, sweating, and an inability to bear weight on the left forelimb. After 18 h the horse was in sternal recumbency. The pain was refractory to several analgesics. Xylazine (Xilagesic; Calier), 0.2 mg/kg BW, detomidine (Domosedan; Pfizer, Madrid, Spain; 0.015 mg/kg BW) and butorphanol (Torbugesic; Fort Dodge), 0.05 mg/kg BW, and later morphine (Morfina 2%; Braun, Barcelona, Spain), 0.2 mg/kg BW were administered IV with little effect. Physical examination revealed a high rectal temperature (39.6°C), tachycardia (100 beats per min) and tachypnea (80 breaths per min). Intestinal motility in all 4 quadrants was detected by auscultation, and normal urine and feces were observed. Rectal examination and abdominal auscultation did not identify any abnormalities.

A blood sample was taken and the results of hematological and serum biochemical analyses are shown in Table 1. There was increased packed cell volume (PCV) and a left shift neutrophilia. The serum gamma aspartate aminotransferase (ASAT), glutamyltransferase (γGT), alkaline phosphatase (AP), creatine kinase (CK), and lactic dehydrogenase (LDH) activities were increased. There was hyperglycemia and hyperfibrinogenemia. Abnormal findings also included high concentrations of serum total bilirrubin, total proteins, albumin, triglyceride, creatinine, and urea. An arterial blood gas analysis performed at this time showed a PaO₂ of 93.0 mmHg, a decreased PaCO₂ (30.6 mmHg), an increased pH (7.442), decreased HCO₃⁻ (20.6 mmol/L), and a base excess of −2.0 mmol/L.

Table 1.

Results of hematological, coagulation, and serum biochemical analyses in a horse with brachial triceps myonecrosis

Parameter	24 h after surgery	4th day after surgery	Normal range
RBC (1012/L)	10.87	9.59	7–13
PCV (%)	51	43	32–47
Hemoglobin (g/dL)	16.8	14.1	11–19
WBC (109/L)	13.6	6.4	6–12
Neutrophils (109/L)	9.65	3.67	2.7–6.7
Lymphocytes (109/L)	1.49	1.67	1.5–5.5
Platelets (109/L)	17a	28a	100–350
Total protein (g/L)	84	68	52–79
Fibrinogen (g/L)	0.006	0.004	0.002–0.004
Albumin (g/L)	0.043	0.036	0.026–0.037
Glucose (mmol/L)	6.43	6.71	4.21–7.04
AP (IU/L)	1119	1032	143–395
Bilirubin (μmol/L)	40.35	80.88	8.55–34.2
AST (IU/L)	942	+	226–366
GGT (IU/L)	52.2	ND	4–13
CK (IU/L)	9310	46 880	86–300
LDH (IU/L)	1417	22 100	< 400
Urea (mmol/L)	19.02	9.28	8.56–17.85
Creatinine (μmol/L)	309.4	234.26	61.88–176.8
Triglyceride (mmol/L)	5.24	6.87	0.04–0.48
Calcium (mmol/L)	2.32	3.4	2.55–3.35
Phosphorus (mmol/L)	0.61	1.21	0.48–1.45
Sodium (mmol/L)	137	138	132–146
Potassium (mmol/L)	2.5	2.8	2.2–4.7
Chlorine (mmol/L)	94	110	99–109

RBC — red blood cells; PCV — packed cell volume; WBC — white blood cells; AP — alkaline phosphatase; AST — aspartate aminotransferase; GGT — gamma-glutamyl transferase; CK — creatine kinase; LDH — lactic dehydrogenase.

^aplatelet count was normal (> 100) by manual quantification; + — out of readable limits; ND — not done.

The affected limb was slightly swollen and warm, but not palpably hard. It was not cyanotic and the horse showed an inability to bear weight on or to extend the limb. Sensory perception of the left limb was assessed as positive. A tentative diagnosis of postanesthetic myopathy and neuropathy of the left forelimb was made based on the abnormal increases in CK and AST activities, related to the presence of substantial muscle damage, and the described clinical signs.

Supportive therapy consisted of antibiotic prophylaxis with penicillin (Penilevel; Laboratorios ERN), 22 000 UI/kg BW, IV, q8h for 7 d and gentamicin (Gentavex-10; SP Veterinaria), 6.6 mg/kg BW, IV, q24h for 5 d; anti-inflammatory agents, phenylbutazone (Butasyl; Fort Dodge, Madrid, Spain) 2.2 mg/kg BW, IV, q12h for 7 d and methylprednisolone, for stabilizing cell membranes (Solumoderin 500; Pharmacia, Madrid, Spain), 0.5 mg/kg IV, q12h for 2 d; anti-endotoxemic drugs, dimethyl sulfoxide (Dimethyl Sulphoxide PRS; Panreac, Barcelona, Spain), 1 g/kg BW, IV, in a 10% solution of a polyionic fluid, q8h for 7 d and flunixin-meglumine (Niglumine; Calier), 0.25 mg/kg IV, q8h for 7 d. Acepromazine maleate (Calmo Neosan; Pfizer, Madrid, Spain), 0.05 mg/kg BW, IM, q8h for 7 d was used to improve muscle perfusion, thiamine (Nervobion; Merck, Barcelona, Spain), 1 g IM, q24h for 7 d to recover traumatic nerve injuries and omeprazole (Omeprazol; Guinama, Valencia, Spain), 1 mg/kg BW orally, q24h for 7 d as a gastroduodenal mucosal protector. Lactated Ringer’s solution (initially, 20 mL/kg BW/h, IV, and for maintenance, 5 mL/kg BW/h) was administered to hydrate and maintain electrolyte balance and renal perfusion, and 5% glucose (G5, Braun, Barcelona, Spain), 3 mL/kg BW/h for nutritional support.

No significant improvement was observed 7 d after surgery. The horse remained permanently in lateral or sternal recumbency. Vital signs remained abnormal, but the owner insisted on continuing therapy. There was no change in the degree of lameness of the left forelimb. Euthanasia was elected on the 7th day after surgery.

Postmortem analysis revealed that the left brachial triceps muscle was markedly dry and pale (Figure 1). An extensive granularity was also observed in this muscle. The liver was enlarged and pale. Hematoxylin and eosin (H&E), Periodic acid-Schiff (PAS), and Mallory’s trichromic stains were applied to muscle and liver sections. Severe generalized rhabdomyolisis and myonecrosis were observed microscopically in the triceps. The muscle fibers showed increased acidophilia with homogenized and hyalinized cytoplasm (Figure 2a), while nuclei were pycnotic. Some fibers were fragmented and substantial numbers of macrophages were located between the altered fibers (Figure 2b). Small areas with hemorrhages were also found. A moderate lipomatosis was observed with adipocytes infiltrating the muscle fascicules. Fibrosis and fiber mineralization were absent. Other muscles were also evaluated, but no significant lesions were found. Histopathological analysis of the liver showed a severe and diffuse hydropic vacuolar degeneration of the hepatocytes.

Figure 1.

Macroscopic appearance of the injured muscle in a horse with brachial triceps myonecrosis. The muscle appeared pale and dry. Normal muscle fibers are present in the upper side of the image.

Figure 2a.

Microscopic image of necrosis of muscle fibers. The fibers show increased acidophilia, with homogenization of the cytoplasm and pycnotic nuclei. Some of the fibers show fragmentation of the cytoplasm. Note the presence of abundant adipocytes in the upper side of the image. Hematoxylin and eosin. Bar = 200 μm.

Figure 2b.

Detailed segment of Figure 2a showing necrotic and fragmented muscle fibers. Note the presence of macrophages eliminating the cell debris. Hematoxylin and eosin. Bar = 50 μm.

Discussion

Severe myonecrosis was found in the left brachial triceps and the postmortem analysis showed that the triceps muscle differed significantly from the other muscles. The pathogenesis of muscle fiber necrosis includes ischemia, multiple metabolic factors, autoimmune sensitization and inflammation, toxic insults, trauma, and unknown factors exemplified by the muscular dystrophies (8). Thrombosis is an uncommon but potentially serious ischemic disorder. The physical examination of the patient did not reveal signs of a swollen, cyanotic, cold limb, as was reported in a case of thrombosis of the brachial artery (9). A form of paresthesia clinically similar to PAM and caused by sensory neuropraxis has also been described after general anesthesia (10). Nevertheless, this condition is not commonly associated with increases in the serum CK activity (10).

In a PAM model, both duration of the operation and body position during surgery were associated with an increased risk. Procedures lasting more than 90 min carried an increased risk. The risk for PAM was also increased in horses placed in lateral recumbency (5). In our case, the horse was placed in lateral recumbency for 105 min of surgical anaesthesia and remained in this position for another 60 min during recovery. These facts could explain, in part, the development of the reported complication because lateral recumbency leads to a greater ICMP in muscles, which are essential for the act of standing and walking (11). Moreover, it has been reported that the ICMP values in dependent pelvic limb muscles are considerably lower than in the thoracic limb muscles (12). Therefore, it is possible that ICMP may play a greater role in the occurrence of PAM in the thoracic limb than in the pelvic limb (4). The clinical signs, the high levels of CK and AST detected postoperatively, and the necropsy findings are compatible with an unusual case of localized PAM produced by an increased ICMP.

Anesthesia-related decreases in skeletal muscle blood flow have been quantified, demonstrating that muscle perfusion decreased markedly during halothane or isoflurane anaesthesia (4,13). The traditional approach to prevention of PAM is to avoid intraoperative hypotension. In our case, arterial blood pressure remained above safe levels (MAP > 75 mmHg) and it was not necessary to administer inotropic/vasopressor support during the maintenance of anesthesia. There is evidence that increasing blood pressure by increasing systemic vascular resistance may adversely affect muscle blood flow and cardiac function (4). On the other hand, it has been reported that when a high ICMP or obstructed venous outflow develops, normotension or even hypertension may not provide sufficient driving pressure to ensure adequate muscle perfussion (14) and ischemia and PAM may still develop. In this case, a standard anesthetic protocol was employed and halothane was selected to maintain surgical anesthesia. A lack of differences between halothane and isoflurane has been reported on the overall equine perioperative mortality (5). However, although isoflurane is more hypotensive than halothane, it is able to provide greater cardiac output and femoral blood flow (15). These observations may suggest that cardiac output has a greater effect on muscle blood flow than arterial blood pressure, explaining, in part, the belief that the incidence of PAM is lower in hypotensive horses anesthetized with isoflurane (16). Nevertheless, significant differences have not been reported between halothane and isoflurane regarding the incidence of PAM (5).

Hypoxemia is also a contributing factor for PAM (7). In the present case, spontaneous ventilation was allowed until an arterial blood gas analysis showed a value of PaO₂ of 86.7 mmHg. Respiratory acidosis as a result of hypercapnia is a common finding in anesthetized horses. It is due to a ventilatory impairment caused by the large body size and weight of these patients, aggravated by the recumbency and the anesthesia-induced depression of systemic functions, especially during prolonged surgical procedures (17). Therefore, mechanical ventilation was commenced to improve the patient’s oxygenation status. Prevention of PAM is not always possible; it may still occur despite appropriate prophylactic measures. However, reducing the duration of anesthesia, use of adequate padding and limb support, and prevention of intraoperative hypotension and hypoxemia are important factors in reducing its occurrence.

It is probable that in our case, an increased ICMP on the triceps muscle and/or radial nerve may be the primary cause for the limb dysfunction. Damage may have occurred because of increases in pressure within the osteofascial compartments producing a localized ischemia. The compartmental syndrome is defined as a condition in which the circulation and function of tissues within a closed space are compromised by an increased pressure within that space (18). Intracompartmental muscle pressure is increased by a direct compression of the compartment due to the recumbent position of the horse or by an enlargement in the volume of the compartment. A sufficiently high ICMP able to compromise capillary blood flow and possibly affect neural transmission has been reported in anesthetized horses (11). The volume of the muscle compartment may also increase pathologically in cases of edema (18), as the result of lactate accumulation in ischemic muscles or from ischemic damage of the muscle capillary endothelium (12). Postanesthetic compartmental syndrome has been reported only rarely in the Thoroughbred, perhaps because compartmental pressures decrease rapidly when the horse is able to stand (19), and it has not been described previously in the Spanish-bred horse. The triceps is the muscle that is most often involved and this condition produces lameness, consisting of a dropped elbow with an associated inability to bear weight on the affected limb (11). Lack of padding, inappropriate positioning, excessive weight of the animal, and the fitness of the horse are also associated with this syndrome (6,10). The highest ICMP values were recorded in muscles with no padding when the limb in contact with the table was kept perpendicular to the body and the free limb was unsupported (11). However, muscle perfusion disturbances can occur even in horses provided with adequate padding (10). Our patient was a heavy horse and it was in dressage training. These facts could also have contributed to the appearance of the reported condition as earlier described in fit Thoroughbreds (20).

Clinical signs of PAM are usually noticeable during the initial recovery period, but may occasionally be delayed for 1 h or more after recovery from anesthesia (10), as in our case. This fact also suggests that a reperfusion injury may have had a role in the development of the reported condition. Reperfusion generates strong oxidants, possibly initiating membrane lipid peroxidation leading to muscle damage (21). The early clinical signs observed in our case were similar to those of the ICMP syndrome. These signs are related primarily to ischemic damage of the muscle, nerves, and capillary endothelium within the muscle compartment, but compression-induced damage also disrupts nerve conduction (19). Surprisingly, there was only moderate local swelling and no induration, in the affected forelimb. Fever was probably due to increased discomfort of the animal. However, necropsy findings and high elevations in serum LDH, CK, and AST activities confirmed the presence of myonecrosis.

Postoperative increases in serum concentrations of blood urea nitrogen and creatinine may be attributed to decreased blood flow to the kidneys during anaesthesia or posterior dehydration. The nephrotoxic effect of phenylbutazone may also have contributed. The high activities of liver enzymes detected in this case may have been the result of intraoperative hepatic hypoperfusion, which could have caused liver cell damage. The high activities of γGT, AP, and AST may also have been partially induced by the combined use of all the drugs (halothane, diazepam, corticosteroids, NSAID’S) administered to the animal, as well as by the myonecrosis which is potentially a liver degenerative pathogenic factor.

In conclusion, an ICMP syndrome could be the most probable cause of this unusual case of localized PAM, which resulted in a fatal outcome. This syndrome was probably initiated during surgery when the horse was placed in lateral recumbency in spite of adequate padding. Then, compressive forces may have occluded the blood supply to the affected triceps muscle, despite the fact that the mean blood pressure values were above the safe limits. Direct compression may also have damaged nerves in the affected forelimb, contributing to the observed signs. It would be important to develop quantitative techniques for assessing muscle perfusion and ICMP, suitable in clinical practice, to reduce the morbimortality of equine anesthesia due to this condition.