Evaluations of Labrador Retrievers With Exercise-Induced Collapse, Including Response to a Standardized Strenuous Exercise Protocol
Clinical and metabolic variables were evaluated in 14 Labrador retrievers with exerciseinduced collapse (EIC) before, during, and following completion of a standardized strenuous exercise protocol. Findings were compared with previously reported variables from 14 normal Labrador retrievers that participated in the same protocol. Ten of 14 dogs with EIC developed an abnormal gait during evaluation, and these dogs were significantly more tachycardic and had a more severe respiratory alkalosis after exercise compared to the normal dogs. Muscle biopsy characteristics and sequential lactate and pyruvate concentrations were normal. Genetic testing and linkage analysis excluded malignant hyperthermia as the cause of EIC. Common causes of exercise intolerance were eliminated, but the cause of collapse in EIC was not determined.
Introduction
A syndrome of exercise-induced collapse (EIC) is recognized as a common cause of exercise intolerance in otherwise healthy, young adult Labrador retrievers.1–4 Affected dogs can tolerate mild to moderate exercise, but they occasionally become ataxic and collapse after 5 to 15 minutes of participation in trigger activities involving strenuous exercise, especially with concurrent extreme excitement or stress.5 Afamilial basis for the disorder has been established, but the precise mode of inheritance has not been determined.5
Exercise intolerance has been speculated to be due to malignant hyperthermia, hypoglycemia, electrolyte disturbances, cardiac rhythm disturbances, a congenital form of myasthenia gravis, or metabolic myopathy; however, physical and biochemical examinations of affected dogs at rest have not revealed consistent abnormalities.3,6–8 Episodes of collapse typically occur only during strenuous exercise, when immediate veterinary evaluation is difficult.
This report describes the clinical and laboratory evaluations of dogs with EIC at rest and during and following completion of a standardized strenuous exercise protocol. Findings are compared with the findings previously reported from the authors’ laboratory of Labrador retrievers without EIC performing the same exercise.9 Genetic testing and linkage analysis to specifically exclude malignant hyperthermia as the cause of EIC are also described. The purpose of this study was to establish clinical and laboratory variables defining EIC in order to assist veterinary diagnosis of the syndrome and to gain insight into the reason for collapse in affected dogs.
Materials and Methods
Dogs With EIC
Fourteen client-owned, purebred Labrador retrievers were referred to the investigating institution with a presumptive diagnosis of EIC. In this group were five black males, five black females, two yellow females, one yellow male, and one chocolate male. All were sexually intact. Ages at the time of evaluation ranged from 1 to 4 years (mean 1.9 years), and weight ranged from 20 kg to 37.7 kg (mean 29.3 kg). Each dog had experienced at least three observed episodes of exercise-induced weakness or collapse beginning at 5 to 22 months of age (mean 11 months). Veterinary evaluation at rest had determined that systemic causes of collapse other than EIC were unlikely.
Complete physical examination (including orthopedic and neurological examinations), thoracic radiography, electrocardiogram (ECG), echocardiography, complete blood cell count (CBC), serum biochemical profile, and acetylcholine receptor (AChR) antibody titer were performed on each dog. Plasma and urine were collected for determination of carnitine (total, free, and esters). Serum thyroxine was measured in six dogs. Before referral or on the day following exercise, six dogs had serum cortisol measured before and 1 hour after the intravenous (IV) administration of synthetic adrenocorticotropic hormone (ACTH).a
Exercise
A standardized exercise protocol was followed as previously described.9 Briefly, dogs repeatedly retrieved a soft plastic tube (retrieving dummy) thrown 37 to 46 meters on land for 10 minutes or until an abnormal gait was apparent to the investigators. The exercise consisted of short bursts of strenuous exercise interspersed with brief (<5 seconds) pauses as the dummy was rethrown. Dogs typically sprinted at full speed to retrieve the dummy, and they ran back to the handler at a slightly slower pace. Speed was determined by calculating the distance run (number of retrieves × length of retrieve) during time measured with a stopwatch. Dogs were exercised in ambient temperatures ranging from 2°C to 21°C (mean 14.9°C). Exercise and recovery were recorded on videotape for gait and pace analysis. If exercise resulted in an abnormal gait or collapse, dogs were classified as EIC/COLL.
Rectal temperature, pulse rate and character, reflexes, arterial blood gas and acid-base status, and plasma lactate and pyruvate were measured at rest and within 2 minutes after terminating exercise. A serum biochemical profile was also performed immediately after exercise. A lead 2 ECG was performed on five dogs immediately following exercise. Rectal temperature, pulse rate and character, and limb reflexes were assessed 5, 10, 15, 30, 60, and 120 minutes after exercise, and venous blood was obtained for measurement of lactate and pyruvate at 15, 30, 60, and 120 minutes after exercise.
Measurement of Variables
Venous blood was collected into plain tubes, and serum was separated within 10 minutes for biochemical analysis using an automated analyzer.b Blood from either the femoral or dorsal metatarsal artery was collected into a heparinized syringe and was analyzed immediately after collection.c Oxygen tension (PaO2), carbon dioxide tension (PaCO2), and pH were measured and corrected for rectal temperature, and bicarbonate concentration was calculated. Venous blood for lactate analysis was collected into sodium fluoride-containing tubes. Plasma was separated within 30 minutes of collection and was frozen and stored at −20°C until analyzed. Lactate concentration was determined by use of an immobilized enzyme membrane system.d For pyruvate analysis, 1 mL of venous blood was collected into a tube containing 1 mL of 10% perchloric acid, mixed, and then centrifuged for 10 minutes. The supernatant was removed and frozen at −20°C until analyzed. Pyruvate concentration was quantified by enzymatic determination, using lactate dehydrogenase, in a spectrophotometric assay.e Plasma and urine concentrations of total, free, and esterified carnitine were determined by radioisotopic enzyme assay using the method of Bieber and Lewin.10
Muscle Biopsy
Twenty-four to 48 hours following the exercise protocol, each dog was routinely anesthetized (acepromazine 0.5 mg/kg intramuscularly [IM]; hydromorphone 0.1 mg/kg IM; propofol 4 mg/kg IV; isoflurane inhalation). A biopsy of the vastus lateralis muscle was obtained surgically, placed in a watertight container, and shipped at 4°C to the Comparative Neuromuscular Laboratory.e Biopsies were flash frozen in isopentane that was precooled in liquid nitrogen and processed by a standard panel of histological and histochemical stains and enzyme reactions (including hematoxylin and eosin, modified Gomori trichrome, periodic acid Schiff, myofibrillar ATPase reactions for fiber typing, succinic dehydrogenase, oil red O, acid and alkaline phosphatase, and esterase).11 Muscle concentrations of total, free, and esterified carnitine were determined by radioisotopic enzyme assay using the method of Bieber and Lewin.10
Examination for Malignant Hyperthermia Mutation
Deoxyribonucleic acid (DNA) was isolated from whole blood of the 14 dogs with EIC using the Gentra isolation protocol,f and it was stored at −20°C. All dogs were genotyped for the V547A malignant hyperthermia ryanodine receptor (RYR1) mutation on canine chromosome 1 using a protocol described by Roberts.12 Deoxyribonucleic acid from dogs with malignant hyperthermia known to contain the RYR1 mutation served as a positive control.
Exclusion Analysis of the RYR1 Gene
Ninety-six dogs (including the 14 dogs with EIC evaluated in this report), 49 additional dogs affected by EIC according to the criteria outlined for this study, and 33 clinically unaffected dogs were selected from five large families in which EIC was segregating. Six canine microsatellite markers (C01.673, C01.246, REN112I02, FH2309, FH2598, FH2294) adjacent to the RYR1 locus on chromosome 1 were amplified by polymerase chain reaction, and their genotypes were obtained using a Beckman CEQ 8000 automated DNA analyzer.12 Genotypes were entered into the FASTLINK13–15 genetic analysis software, and linkage analysis was performed assuming an autosomal recessive mode of inheritance with 80% penetrance. Significant evidence for exclusion would be indicated with logarithm of odds (LOD) scores < −2.0.
Analysis of Data
Clinical and laboratory variables following exercise were compared with 14 historical control dogs that underwent the same exercise protocol using t-tests for two independent samples.g These dogs were purebred, competitive field-trial Labrador retrievers with normal capacity for exercise and no history of weakness or collapse. They were determined to be healthy on the basis of history, physical examination, and laboratory evaluation (CBC, biochemical profile).9 In this group were five black males, three black females, three yellow females, two yellow males, and one chocolate female. One female was spayed, and one male was neutered. Ages ranged from 1 to 7 years (mean 3.4 years), and weight ranged from 23 kg to 38 kg (mean 30.4 kg). Preand postexercise values for these normal dogs were previously reported to establish reference ranges for healthy, trained Labrador retrievers participating in a standardized strenuous exercise protocol.9
Clinical and laboratory findings prior to exercise were compared with established canine reference values.h Normal dogs were exercised in ambient temperatures ranging from 11°C to 28°C (mean 16.7°C),9 while dogs with EIC were exercised in ambient temperatures ranging from 2°C to 21°C (mean 14.9°C). For dogs with rectal temperatures above the upper limit of the thermometer (42.0°C), a value of 42.0°C was used for calculations. When rectal temperatures following exercise were above this limit, temperature values were reported but not statistically compared. The level of significance was P<0.05 for all tests. Analysis was repeated comparing the normal dogs to the smaller group of 10 (EIC/COLL) dogs that developed an abnormal gait during or immediately after exercise. When the Levene’s test for equality of variances indicated that the assumption of equal variances was violated, the results of the t-tests were presented, assuming unequal variances and using separate variances t-tests. Normality was checked graphically to evaluate whether distributions were symmetric and without obvious outliers.
Results
Evaluation of Dogs With EIC When at Rest
Orthopedic and neurological examinations (including assessment of muscle mass, range of motion, strength, muscle tone, and limb reflexes) were normal. Physical examinations, thoracic radiography, ECGs, and echocardiography were normal in all dogs except one. That dog had a grade 2/6 systolic murmur heard loudest over the right heart apex, mild right atrial dilatation, abnormal configuration and mobility of the septal leaflet of the tricuspid valve, and mild tricuspid regurgitation consistent with previously diagnosed tricuspid dysplasia.
All CBC variables were within reference ranges. Serum biochemical variables (including sodium, potassium, calcium, phosphorus, urea, creatinine, bilirubin, creatine kinase [CK], alkaline phosphatase, albumin, and total protein) were within reference ranges for all dogs. Seven dogs were mildly hyperglycemic (range 5.7 to 10.5 mmol/L; reference range 3.3 to 5.6 mmol/L). SerumAChR antibody titers were normal (<0.6 nmol/L) in all dogs. Resting serum thyroxine (10 to 36 nmol/L) and cortisol response to ACTH stimulation (resting cortisol: 20 to 270 nmol/L; post ACTH cortisol: 230 to 570 nmol/L) were normal in the six dogs tested.
Carnitine
Plasma, urine, and muscle carnitine concentrations in the dogs with EIC were compared to reference values previously established by the Comparative Neuromuscular Laboratory.e No consistent abnormalities were seen in plasma concentrations of total, free, or esterified carnitine [Table 1]. Muscle concentrations of total and free carnitine were decreased in three of 14 and in five of 14 dogs, respectively. Urinary excretions of total carnitine, free carnitine, and carnitine esters were increased in six of 14, six of 14, and five of 14 dogs, respectively.
Examination for Malignant Hyperthermia Mutation
All 14 dogs with EIC were negative for the known RYR1 mutation associated with malignant hyperthermia in dogs.
Exclusion of the RYR1 Locus on Chromosome 1
No chromosome 1 markers that were tested showed linkage to EIC. Marker FH2598 excluded linkage to EIC up to 20 centimorgans, which includes the region containing the RYR1 gene.
Exercise
Dogs with EIC retrieved at 137 to 234 meters per minute. The average speed of the four dogs that did not develop an abnormal gait (183±16 meters per minute) was not statistically different from that of the dogs with EIC/COLL (202±27 meters per minute) (P=0.11) or the previously reported normal dogs (P=0.48).9 Ten dogs with EIC retrieved for the full 10 minutes. Four dogs had their exercise halted early when they developed progressively worsening gait abnormalities. Ten dogs exhibited gait abnormalities during or immediately following the retrieving exercise (EIC/COLL group), while four dogs maintained a normal gait.
Gait abnormalities that developed after 4 to 7 minutes of exercise in dogs with EIC/COLL included a rocking or forced gait, carpal hyperextension, a base-wide pelvic-limb stance, excessive pelvic-limb abduction when turning, and occasional stumbling when pausing to pick up the dummy. One dog collapsed after 9 minutes of exercise.
All 10 dogs with EIC/COLL had progression of signs within 2 minutes after halting exercise, to the point that three dogs could not stand. The seven dogs with EIC/COLL that were able to walk did so with flexed (crouched) rear legs [Figure 1] and an incoordinated pelvic-limb gait characterized by long strides, excessive limb abduction during turning, and a base-wide stance [Video 1]. Conscious proprioception was normal as assessed by response to knuckling and hopping. Four dogs were hypermetric in all four limbs. Six dogs exhibited a loss of balance with incoordination, difficulty righting themselves, and falling to the side [Video 2]. None developed a head tilt or abnormal nystagmus. One dog was unable to voluntarily move any limb or trunk muscles for 10 minutes after exercise [Video 3], and this dog exhibited abdominal breathing with apparent loss of intercostal muscle function.
When assessed in lateral recumbency, muscle tone was judged to be decreased in the rear limbs of all 10 of the dogs with EIC/COLL. Patellar reflexes were absent for 5 to 30 minutes (mean 17 minutes) after exercise in all of the dogs with EIC/COLL and in two of the noncollapsing dogs with EIC, but cranial tibial and withdrawal reflexes were not changed from pre-exercise assessment. Seven dogs with EIC/COLL exhibited dramatically increased forelimb extensor tone with normal reflexes while recumbent [Video 4]. All 14 dogs with EIC appeared to be mentally normal. Every dog panted during exercise and recovery. All 10 dogs with EIC/COLL had a gradual return to a normal gait within 10 to 30 minutes (mean 16 minutes). None exhibited muscular pain or stiffness upon recovery, and all were willing to resume retrieving.
In contrast to the dogs reported here, all of the previously reported normal dogs participated in the retrieving drill for 10 minutes with no evidence of weakness, incoordination, or exercise intolerance. They retrieved at an average speed of 179±20 meters per minute. Dogs were noticeably tired and panted heavily at the end of the session, but all indicated eagerness and ability to continue retrieving.9 All dogs were tachycardic and hyperthermic after exercise [Table 2]. Femoral arterial pulses were strong and regular, and in the five dogs that had an ECG performed (including the dog with tricuspid dysplasia), sinus tachycardia was present.When the 14 dogs with EIC were compared with the 14 normal dogs, no difference was found with respect to pulse rate; however, the 10 dogs with EIC/COLL were slightly more tachycardic than the normal dogs immediately following exercise. Body temperature exceeded 42°C immediately after exercise in seven of 14 normal dogs and in nine of 14 dogs with EIC—all from the EIC/COLL group. The rate of temperature decline and the time to achieve normal body temperature after exercise did not differ from those of the normal dogs [Figure 2].
Laboratory Evaluation Following Exercise
Arterial blood gas.
Arterial blood pH was increased, and PaCO2 was decreased in all dogs immediately after exercise [Table 2]. These changes were most pronounced in the 10 dogs with EIC/COLL but were not statistically different between the 14 dogs with EIC and the 14 normal dogs. In all dogs, PaO2 was increased after exercise, but the increase in dogs with EIC was less than that in normal dogs. Decreased serum bicarbonate reflecting hyperlactatemia was documented in all dogs immediately after exercise, but no significant difference was seen between groups.
Serum biochemical values.
Many serum biochemical values were slightly altered after exercise; but with the exception of creatinine, glucose, and CK, values remained within the normal reference range for all dogs [Table 2]. Serum creatinine values following exercise were slightly elevated (≤176 μmol/L; reference range 60 to 140 μmol/L) in three dogs with EIC and in three normal dogs, with no overall differences between groups. Thirteen of the 14 dogs with EIC and nine of 14 normal dogs were mildly hyperglycemic following exercise. The hyperglycemia was most severe in the dogs with EIC/COLL (P=0.02) [Table 2]. Serum CK following exercise was increased over pre-exercise values in all groups, but only the dogs with EIC had increases outside of the reference range. Eight of the dogs with EIC, including five of the 10 dogs with EIC/COLL, had mildly increased (<5× normal) serum CK concentrations. This resulted in a measurably significant difference between normal dogs and dogs with EIC for this variable (P=0.02).
Lactate and pyruvate.
The resting and post-exercise lactate and pyruvate concentrations and the lactate to pyruvate ratios in the dogs with EIC and the dogs with EIC/COLL were not different from the values in normal dogs [Table 3]. Plasma lactate and pyruvate concentrations at 15, 30, 60, and 120 minutes after exercise were also not different between groups.
Muscle biopsy.
No abnormalities were noted during monitoring of temperature, blood pressure, and ventilation during general anesthesia. Abnormalities were not detected in muscle biopsy specimens with any of the histological and histochemical stains and reactions performed; therefore, inflammatory and noninflammatory (breed-associated and dystrophic) myopathies and neuropathies were eliminated as causes of exercise intolerance and collapse.
Discussion
When dogs with EIC were at rest, no abnormalities were seen in the clinical or laboratory evaluations that distinguished them from normal dogs. Initial evaluation was sufficient to eliminate many causes of exercise intolerance, such as severe anemia, cardiac disease, severe pulmonary disease, glucocorticoid deficiency, hypothyroidism, and acquired myasthenia gravis.8,16 Normal muscle mass, normal pre-exercise patellar reflexes, normal muscle biopsy findings, and the ability to participate in moderate exercise made it unlikely that the dogs with EIC were affected by any of the most common inherited myopathies occurring in this breed. Common myopathies affecting Labrador retrievers include centronuclearmyopathy (CNM) (previously known as hereditary myopathy of Labrador retrievers) and dystrophindeficient muscular dystrophy, as well as acquired inflammatory myopathies such as infectious or immune-mediated myositis.7,17–21 Abnormalities of glycolytic or oxidative metabolism that could result in poor exercise tolerance were also excluded by the absence of storage products in muscle fibers, the normal carnitine concentrations, and normal lactate and pyruvate concentrations.
All dogs with EIC appeared eager to participate in the standardized strenuous exercise protocol. The four dogs that did not develop a gait abnormality during this retrieving drill subjectively appeared to be less enthusiastic about the drill and may have been pacing themselves, although their rate of retrieving was not different from other dogs. The authors subsequently observed at least one episode of collapse in each of these four dogs during retrieving drills at the investigating institution or in videotape provided by the owners.
The earliest gait abnormalities seen in exercising dogs with EIC usually included a rocking or forced gait and a wide-based, pelvic-limb stance. Those dogs that could still walk after the exercise protocol did so with a very characteristic, crouched rear-limb gait, with long strides and a base-wide posture when turning. Loss of the patellar reflex after exercise was a consistent finding in all dogs with EIC/COLL, and this persisted even beyond recovery to a normal gait. Perhaps most dramatic was the balance problem exhibited by six of the dogs.
The gait abnormalities in the dogs with EIC/COLL were more suggestive of neurological dysfunction as opposed to isolated muscular weakness or exhaustion.22,23 Dogs with disorders affecting muscle, peripheral nerve, or the neuromuscular junction typically experience weakness and a short-strided, stiff gait rather than the incoordinated, longstrided, base-wide gait and loss of balance seen in these dogs.7,17,23–26 Also, most dogs with muscle or neuromuscular junction disease experience muscle weakness with minimal exercise, not exclusively with strenuous exercise (as seen in dogs with EIC).7,17,26
The collapse in EIC may be a manifestation of combined muscular and nervous system dysfunction. A wide range of metabolic disorders of glycogen, fatty acid, or mitochondrial oxidative metabolism are associated with neurological and muscular dysfunction in humans, but they are infrequently described in small animals.27–29 Early reports of increased plasma lactate concentrations in dogs with EIC after exercise led to speculation that EIC might be a metabolic myopathy.8,30 Evaluation of plasma lactate and pyruvate concentrations and determination of the lactate to pyruvate ratio after exercise are useful for assessing the mitochondrial electron chain and other aspects of oxidative metabolism. Disorders affecting oxidative metabolism may also be associated with primary or secondary alterations in carnitine status, as this peptide acts as a carrier for longchain fatty acids into mitochondria.29,32
Dogs with EIC in this study developed similar increases in lactate and pyruvate as healthy dogs undergoing the same strenuous activity protocol; also, muscle biopsy findings were normal, making a disorder of oxidative metabolism unlikely.8,27–33 No consistent abnormalities were seen in plasma or muscle concentrations of carnitine. The mildly decreased muscle carnitine and increased urinary excretion of carnitine seen in some of the dogs with EIC have been reported as nonspecific findings in a wide range of muscular and neuromuscular disorders.29
All of the dogs with EIC/COLL had progressive deterioration of their condition during approximately the first 2 minutes after exercise was terminated. One dog—still running at the end of exercise—became unable to voluntarily move any limbs or maintain a sternal position. This dog lost intercostal respiratory muscle function within 2 minutes of halting exercise. Deterioration after stopping exercise is commonly reported by owners of dogs with EIC.4 It is important for veterinarians evaluating these dogs to inform owners that collapse can be fatal and to recommend that exercise be halted as soon as any gait abnormality is recognized.5
Serum biochemical findings following exercise in dogs with EIC and EIC/COLL did not provide an explanation for their collapse. Hypoglycemia and abnormalities of sodium, potassium, and calcium were not observed. Although a few of the dogs with EIC did have mild (<5× normal) elevations in serum CK after exercise, the increases were not in the range characteristic of a dystrophic myopathy or acute myonecrosis secondary to exertional rhabdomyolysis, malignant hyperthermia, or heat stroke.17,30,34–37 Serum CK was only measured immediately following exercise, and possibly elevations were more pronounced 4 to 24 hours later. The increases following exercise do suggest that some degree of muscle damage occurred in the dogs with EIC, even though clinical examination and evaluation of muscle biopsies obtained 24 hours later were unremarkable.
Exercise-induced collapse has been speculated to be a heat-related disease or even a manifestation of heat stroke or exercise-induced malignant hyperthermia.6 Body temperatures after exercise were extremely elevated in the dogs with EIC, but they were not obviously different from the reported values of normal Labrador retrievers participating in the same exercise drill. Rectal temperatures were higher than 42.0°C immediately after exercise in nine of 10 dogs with EIC/COLL and in only seven of 14 of the reported normal dogs.9 Labrador retrievers participating in retrieving activities have been previously shown to develop body temperatures greater than dogs doing other types of exercise.9,37–39 Ambient temperature ranges were different for the two groups of dogs in this analysis, but previous studies have suggested that rectal temperature following exercise is not related to ambient temperature.9 Rates of body temperature decline did not differ between the normal dogs and dogs with EIC/COLL.
Heat stroke in dogs is typically associated with altered mentation and a slow recovery taking hours to days. Mortality is high, with many dogs succumbing to the effects of endothelial injury, microvascular thrombosis, disseminated intravascular coagulation (DIC), and acute renal failure. 36,40,41 Collapse in dogs with EIC is not likely a manifestation of heat stroke, as affected dogs are mentally normal and recover quickly, without veterinary intervention and without significant clinical or laboratory consequences. Rarely is death during an episode of EIC accompanied by postmortem changes suggesting heat-related endothelial injury and DIC; more commonly, complete postmortem evaluation of all tissues is normal.i
Early reports that dogs with EIC were hyperthermic at the time of collapse led some to conclude that EIC might be a manifestation of malignant hyperthermia or canine stress syndrome.6,7 Malignant hyperthermia is a hereditary disorder of skeletal muscle, with collapse episodes triggered by exposure to halothane anesthesia, depolarizing muscle relaxants, and occasionally exercise or stress.12 Typical manifestations of collapse associated with malignant hyperthermia include a rapidly progressive elevation in body temperature, tachycardia, hypercarbia, and rhabdomyolysis.35,42 Most variants of malignant hyperthermia are known to be caused by a mutation in the gene on chromosome 1 encoding the skeletal muscle calcium-release channel (ryanodine receptor 1: RYR1).12 The DNA isolated from dogs with EIC did not contain the V547A mutation shown to cause some forms of canine malignant hyperthermia.12 Exclusion analysis was also performed with microsatellite markers near the RYR1 locus using DNA from 96 dogs with EIC; other mutations at this site were ruled out, conclusively proving that dogs with EIC do not suffer from classical malignant hyperthermia.43
Dogs with EIC and normal Labrador retrievers participating in this strenuous exercise drill developed significant respiratory alkalosis and metabolic acidemia reflective of hyperventilation and strenuous anaerobic activity. Similar findings have been reported in normal Labrador retrievers participating in field trial training and competition.9,37 Hypocarbia associated with hyperventilation has been proposed as a possible mechanism for collapse in dogs with EIC.37 Experimental hyperventilation (PaCO2 <20 mm Hg) of anesthetized dogs causes nervous system vasoconstriction and substantially decreases cerebral and spinal cord blood flow.44,45 Hyperthermia and catecholamine release during exercise may magnify this response.46 Extreme hyperventilation may alter somatosensory input from the lower limbs, cause a loss of balance, and lower seizure threshold.47,48 Normal dogs and dogs with EIC had PaCO2 concentrations well below those reported to cause clinical signs experimentally, but the 10 dogs with EIC/COLL did have slightly lower PaCO2 than the normal dogs.
Conclusion
This clinical and metabolic evaluation of dogs with EIC was unsuccessful in determining the cause for collapse in affected dogs. This study does, however, provide strong evidence to discount many of the previously proposed explanations for collapse in EIC. Also, a detailed clinical and clinicopathological description of collapse episodes is provided.At the present time, diagnosis requires recognition of the syndrome by veterinarians and systematic elimination of all other disorders causing exercise intolerance.
Authors’ Note
Five years have passed since the 14 dogs with EIC were evaluated. One dog was euthanized immediately following evaluation. Six dogs were adopted out to pet homes where they no longer participate in the trigger activities associated with collapse, and five of the six dogs have not collapsed since relocation. Three dogs remained with their owners, and episodes of collapse reportedly occur if they are allowed to engage in trigger activities. One dog has not had activity limited, but episodes of collapse have become very infrequent. Three dogs were lost to follow-up. No dogs have developed progressive systemic or neurological disorders, and all are considered by their owners to be healthy aside from the EIC.
Addendum
A mutation in the dynamin 1 gene (DNM1) was recently described as the causal mutation for EIC in Labrador retrievers.49 The dynamin 1 protein has an essential role in brain and spinal cord neurotransmission and synaptic vesicle endocytosis.Agenetic test for the mutation is now available through the Veterinary Diagnostic Laboratory at the University of Minnesota.j
Cortrosyn, 250 μg per dog; Amphastar Pharmaceuticals, Rancho Cucamonga, CA 91730
Abbott Spectrum II; Abbott Park, Chicago, IL 60064-3500
ABL 330; Radiometer, Copenhagen, Denmark
YSI Inc.; Yellow Springs, OH 45387
Comparative Neuromuscular Laboratory, Department of Pathology, University of California - San Diego, La Jolla, CA 92093-0709
The Gentra Puregene DNA Isolation kit (Qiagen, www1.qiagen.com) was used to extract genomic DNA from 3 to 6 mL of EDTA whole blood per manufacturer’s instructions
Statistical Package for the Social Sciences, V.13, Chicago, IL 2004
Approximate normal ranges for common measurements in dogs and cats. In: Tilley LP, Smith FW, eds. The Five Minute Veterinary Consult. 3rd ed. Philadelphia: Lippincott Williams and Wilkins, 2004:1380.
Taylor SM, unpublished data, 2007
Citation: Journal of the American Animal Hospital Association 45, 1; 10.5326/0450003
Citation: Journal of the American Animal Hospital Association 45, 1; 10.5326/0450003
A Labrador retriever affected by exerciseinduced collapse; the dog has a crouched pelvic-limb posture immediately following completion of 10 minutes of strenuous exercise.
Temporal changes in rectal body temperature following exercise in 14 normal Labrador retrievers, 14 Labrador retrievers with exercise-induced collapse (EIC), and 10 Labrador retrievers showing gait abnormalities due to exercise-induced collapse (EIC/COLL).
