Warmblood horses, which have long been bred as riding horses in Europe, are becoming increasingly popular in North America as performance horses. Exertional myopathies cause performance limitations in warmblood horses, and researchers have found that PSSM is the most frequently diagnosed myopathy in North American and European warmblood horses.1–3 A diagnosis of PSSM is based on identification of aggregates of amylase-resistant or amylase-sensitive PAS-positive polysaccharide inclusions in skeletal muscle fibers.4–6 Since the identification of a dominant nonsynonymous gain-of-function R309H mutation in GYS1, PSSM has been subdivided into PSSM1, which represents horses with the GYS1 mutation, and PSSM2, which represents horses with abnormal PAS staining but that lack the GYS1 mutation.4,7 It is known that PSSM1 is a glycogen storage disorder with muscle glycogen concentrations that are elevated (typically 1.8-fold as high) as a result of increased and poorly regulated glycogen synthase activity.5,7 Muscle glycogen concentrations have not been reported for PSSM2, and the cause or causes of PSSM2 are not known.
Warmblood horses reportedly have both PSSM1 and PSSM2.8,9 The R309H GYS1 mutation was found in a family of Selle Français with PSSM, and many of these horses developed clinical signs of exertional rhabdomyolysis.8 In warmblood horses, PSSM2 is more common than PSSM1, with only 8% of warmblood horses with PSSM having the GYS1 mutation.9 In contrast, the GYS1 mutation is found in 76% of Quarter Horses with PSSM.9 Investigators of 1 report1 described the clinical signs of PSSM in warmblood horses; however, that report preceded the discovery of the GYS1 mutation. Thus, although PSSM2 appears to be common in warmblood horses, no studies have determined whether the clinical features of PSSM2 in warmblood horses are distinct from those of warmblood horses with PSSM1 or from those of nonwarmblood breeds with PSSM2.
The purpose of the study reported here was to use information in a database to characterize the signalment and primary clinical signs of warmblood horses with PSSM1 and PSSM2 and to determine whether muscle glycogen concentrations are elevated in warmblood horses with PSSM1 and PSSM2.
Materials and Methods
Sample
A database containing results for 3,615 diagnostic equine muscle biopsy specimens submitted to the Neuromuscular Diagnostic Laboratory at the University of Minnesota between December 1993 and February 2015 was searched to identify horses with PSSM. Only horses with muscle biopsy specimens collected for clinical diagnosis were included; specimens collected from horses for research purposes were excluded. A diagnosis of PSSM was based on histologic evidence of abnormal PAS-positive aggregates of sub-sarcolemmal or cytoplasmic amylase-resistant or amylase-sensitive polysaccharide.4,7,10 The diagnosis was made prior to evaluation of the clinical history on the submission sheet such that the investigator was not aware of the reported clinical signs and signalment. Horses with PSSM were subsequently classified by type of PSSM, as determined on the basis of GYS1 test results, and further classified by breed (warmblood or nonwarmblood). Horses with evidence of PSSM (by histologic examination of biopsy specimens) that were heterozygous or homozygous for the R309H GYS1 mutation were classified as PSSM1, and those that did not have the R309H GYS1 mutation were classified as PSSM2; horses that did not have genetic test results were classified as PSSM. Horses were classified as warmblood if the owner or referring veterinarian reported that the horse was from a traditional European warmblood breed registry, was a crossbred of these registries, or was listed as a warmblood without further clarification as to registry. Horses were classified as nonwarmbloods if they did not fit the breed definition of a warmblood.
Signalment and primary clinical signs
Age at time of biopsy, breed, sex, diagnosis determined on the basis of histologic examination of the muscle biopsy specimen, and type of PSSM were summarized. The medical history for each biopsy specimen report was searched to determine whether features of rhabdomyolysis, gait abnormalities, signs of shivers (ie, hyperextension or hyperflexion of hind limbs when backing), fasciculations, muscle atrophy, or lameness were present. On the basis of the version of the submission form used (the submission form changed during the study period), some referring veterinarians made these categorizations via check boxes on the submission form; for all cases, the available history at time of submission was read by laboratory staff and the presence of the clinical signs noted. From December 1993 through October 2007, all medical history was open-ended (186 warmblood and 1,636 nonwarmblood horses), and a laboratory staff member categorized whether signs of shivers, gait abnormalities, rhabdomyolysis, muscle fasciculations, or muscle atrophy were reported. From November 2007 through February 2015, submitting veterinarians selected a check box to indicate whether each of these clinical signs was present (395 warmblood and 1,398 nonwarmblood horses). Information on the forms did not distinguish between rhabdomyolysis and exertional rhabdomyolysis; thus, the term rhabdomyolysis was used for the study. The term rhabdomyolysis represented the diagnosis made by referring veterinarians and encompassed their clinical interpretation of clinical signs and the laboratory data collected prior to muscle biopsy. When reported, muscle enzyme activities provided by referring veterinarians were also analyzed and used to corroborate their diagnosis of rhabdomyolysis. Rhabdomyolysis by definition included horses for which CK or AST values provided by referring veterinarians were greater than the reference range. A more definitive definition of rhabdomyolysis and identification of specific clinical signs of rhabdomyolysis were not possible because of the retrospective nature of the study and the circumspect information provided on submission forms. The number of horses with rhabdomyolysis and these clinical signs were analyzed for the following categories: all warmblood horses, all nonwarmblood horses, warmblood horses with PSSM1, nonwarmblood horses with PSSM1, warmblood horses with PSSM2, nonwarmblood horses with PSSM2, and warmblood horses without PSSM.
Serum CK and AST activities
All versions of the biopsy specimen submission form also requested serum CK and AST activities at the time of biopsy; however, these were not always provided. Values were retrospectively analyzed by comparison of the mean CK and AST activities between groups. Because horses may have exercised prior to collection of blood samples and samples were analyzed on a variety of laboratory equipment with differing reference intervals, an upper value for the reference range of 600 U/L was arbitrarily selected on the basis of that value being approximately twice the approximate mean of the reference range (300 U/L).
Gait abnormalities for warmblood horses with PSSM2
To further characterize the nature of gait abnormalities, the database was searched to identify warmblood horses with PSSM2 that had undergone a thorough lameness evaluation that included nuclear scintigraphy (referring veterinarian not at the University of Minnesota) or a detailed lameness evaluation that included negative results for anesthetic nerve blocks and for which the medical records were available for evaluation (examination performed by a veterinarian at the University of Minnesota). Scintigraphy was used as a proxy for a thorough lameness evaluation for horses examined by referring veterinarians not at the University of Minnesota. Results of the lameness examination performed at the time of the muscle biopsy were reviewed to determine whether an orthopedic cause of lameness was identified and to obtain a description of the gait abnormality. The lameness scale of the American Association of Equine Practitioners was used to standardize lameness grades.
Muscle glycogen concentration
Muscle glycogen concentrations were analyzed in stored muscle samples that had been shipped on ice packs to the Neuromuscular Diagnostic Laboratory and frozen in liquid nitrogen within 24 to 48 hours after biopsy. Samples analyzed were from warmblood horses with PSSM1 (n = 14), warmblood horses with PSSM2 (13), and warmblood horses with no histologic abnormalities (no PSSM [10]). In addition, nonwarmblood horses with PSSM1 (n = 6) and nonwarmblood horses with no histologic abnormalities (no PSSM [6]) were also analyzed for comparison. Samples with no histologic abnormalities had been shipped to the laboratory by referring veterinarians and stored in the same manner as PSSM samples. Wet-weight glycogen concentration was assayed fluorometrically as the glucose residues remaining after approximately 4 mg of muscle tissue was boiled in 1M HCl for 2 hours.11
Statistical analysis
All statistical analyses were performed with commercially available software.a Quantitative variables were analyzed with a Student t test to compare 2 groups or a 1-way ANOVA with Tukey post hoc tests to compare ≥ 3 groups. Results were expressed as mean and SD. Activities of CK and AST were not normally distributed and were logarithmically transformed for analysis. The proportion of horses with CK or AST (or both) activity > 600 U/L was also reported. Categorical variables with > 3 groups (sex and GYS1 mutation status) were analyzed with a χ2 test for independence when n was ≥ 5 and a Fisher exact test when n was < 5. However, because of the lack of stallions for the warmblood horses with PSSM1 and the small sample size, a Fisher exact test was used to compare male and female distribution between warmblood horses with PSSM1 and nonwarmblood horses with PSSM1, irrespective of castration status. For all statistical tests, values of P < 0.05 were considered significant.
Results
Warmblood horses compared with nonwarmblood horses
The database contained 581 muscle biopsy specimens from warmblood horses and 3,034 muscle biopsy specimens from horses of other breeds. The most common warmblood breeds were warmblood horse without further specification as to registration, Hanoverian, Dutch Warmblood, Holsteiner, and warmblood crosses (Table 1). The 4 most common nonwarmblood breeds were Quarter Horse (n = 1,380 [45.5%]), Thoroughbred (328 [10.1%]), Paint (268 [8.8%]), and Belgian Draft (86 [2.8%]); there were 215 (7.1%) mixed-breed horses. Warmblood horses were significantly (P < 0.001) older than nonwarmblood horses, and warmblood horses had a significantly (P < 0.001) larger proportion of geldings than mares or stallions, compared with the sex distribution for nonwarmblood horses (Table 2). Warmblood horses were significantly (P < 0.001) more likely to have gait abnormalities and signs of shivers than were nonwarmblood horses. Warmblood horses were significantly (P < 0.001) less likely to have rhabdomyolysis and muscle fasciculations than were nonwarmblood horses. There was no significant (P = 0.62) difference in the presence of muscle atrophy between warmblood and nonwarmblood horses.
Distribution of breeds among warmblood horses without PSSM, warmblood horses with PSSM1, and warmblood horses with PSSM2.
| Breed or registry | Warmbloods without PSSM | Warmbloods with PSSM1 | Warmbloods with PSSM2 |
|---|---|---|---|
| Unspecified warmblood | 80 (28.8) | 7 (43.8) | 53 (28.2) |
| Hanoverian | 46 (16.6) | 1 (6.3) | 26 (13.8) |
| Dutch Warmblood | 32 (11.5) | 0 (0) | 30 (16.0) |
| Holsteiner | 28 (10.1) | 1 (6.3) | 12 (6.4) |
| Warmblood cross | 25 (9.0) | 5 (31.3) | 23 (12.2) |
| Oldenburg | 15 (5.4) | 1 (6.3) | 13 (6.9) |
| Westphalian | 10 (3.6) | 0 (0) | 5 (2.7) |
| Trakehner | 9 (3.2) | 0 (0) | 7 (3.7) |
| Swedish Warmblood | 9 (3.2) | 0 (0) | 7 (3.7) |
| Canadian Warmblood | 7 (2.5) | 0 (0) | 0 (0.0) |
| RPSI or Rheinlander | 4 (1.4) | 1 (6.3) | 3 (1.6) |
| Danish Warmblood | 3 (1.1) | 0 (0) | 3 (1.6) |
| Other warmblood breeds* | 10 (3.6) | 0 (0) | 6 (3.2) |
| Total | 278 (100) | 16 (100.3)† | 188 (100) |
Values reported are number (percentage). For warmblood horses with PSSM1, there was a difference in the proportion of warmblood cross and unspecified warmblood horses, compared with the proportion of those breeds for warmblood horses with or without PSSM2.
Value is combined total for all registries with fewer than 3 horses represented.
Values in column do not sum to 100% because of rounding.
RPSI = Rheinland Pfalz-Saar International.
Distributions (number [percentage]) for age, sex, and clinical signs among warmblood and nonwarmblood horses with and without PSSM.
| Variable | All nonwarmblood horses (n = 3,034) | All warmblood horses (n = 581) | Warmblood horses with PSSM2 (n = 188) | Nonwarmblood horses with PSSM2 (n = 646) | Warmblood horses without PSSM (n = 278) |
|---|---|---|---|---|---|
| Age | |||||
| No. of horses | 2,805 (92.5) | 545 (93.2) | 180 (95.7) | 605 (93.7) | 260 (93.5) |
| Years* | 7.4 ± 5.2 | 8.6 ± 4.0† | 8.9 ± 3.7 | 6.9 ± 4.8‡ | 8.6 ± 4.3 |
| Sex§ | |||||
| Mare | 1,268 (41.8) | 197 (33.9) | 61 (32.1) | 276 (42.7) | 89 (32.0) |
| Gelding | 1,365 (45.0) | 353 (60.8) | 118 (63.1) | 297 (46.0) | 169 (60.8) |
| Stallion | 244 (8.0) | 10 (1.7) | 2 (1.1) | 48 (7.4) | 8 (2.9) |
| Not reported | 157 (5.2) | 21 (3.6) | 7 (3.7) | 25 (3.9) | 12 (4.3) |
| Clinical signs‖ | |||||
| No. of horses | 2,017 (66.5) | 436 (75.0) | 147 (78.2) | 436 (67.5) | 200 (71.9) |
| Rhabdomyolysis | 913 (45.2) | 116 (26.6)† | 39 (26.5) | 221 (50.7)‡ | 44 (23.4) |
| Gait abnormality | 775 (38.4) | 259 (59.4)† | 97 (66.0) | 176 (40.4)‡ | 106 (52.7)¶ |
| Muscle atrophy | 628 (31.1) | 141 (32.3) | 36 (24.5) | 94 (21.6) | 72 (35.0)¶ |
| Fasciculations | 450 (22.3) | 62 (14.2)† | 22 (15.0) | 94 (21.6) | 30 (14.9) |
| Signs of shivers# | 95 (4.7) | 52 (11.9)† | 18 (12.2) | 17 (3.90)‡ | 26 (12.9) |
Value reported is mean ± SD.
Value differs significantly (P < 0.001) from the value for all nonwarmblood horses.
Value differs significantly (P < 0.001) from the value for all warmblood horses with PSSM2.
Proportions differed significantly (P < 0.001) between all nonwarmblood horses and all warmblood horses and between nonwarmblood horses with PSSM2 and warmblood horses with PSSM2.
Each horse could have multiple clinical signs.
Value differs significantly (P < 0.05) from the value for warmblood horses with PSSM2.
Signs of shivers, such as hyperextension or hyperflexion of hind limb during backing.
Warmblood horses with PSSM1 compared with nonwarmblood horses with PSSM1
Only 16 of 253 tested warmblood horses were identified with PSSM1, as determined on the basis of histologic examination of muscle biopsy specimens and GYS1 testing (Table 3). There was a significantly (P < 0.001) smaller proportion of warmblood horses with PSSM1, compared with the proportion of nonwarmblood horses with PSSM1. All warmblood horses were heterozygous for the GYS1 mutation, whereas some horses of other breeds were homozygous for the GYS1 mutation. Most warmblood horses with PSSM1 were listed as unspecified warmbloods or warmblood crosses (warmblood × Quarter Horse [n = 4] or warmblood × Thoroughbred [1]; Table 1). Of the 430 nonwarmblood horses with PSSM1, 304 (70.7%) were Quarter Horse–related breeds, 54 (12.6%) were draft breeds, and 51 (11.9%) were mixed-breed horses. There were no significant differences between warmblood horses with PSSM1 and nonwarmblood horses without PSSM1 with regard to mean ± SD age (7.5 ± 3.8 years and 7.0 ± 4.6 years, respectively; P = 0.71) or sex distribution (warmblood mares, 10/16 [62.5%]; nonwarmblood mares, 205/416 [49.2%]; warmblood geldings, 6/16 [37.5%]; nonwarmblood geldings, 192/416 [46.2%]; warmblood stallions, 0/16 [0%]; and nonwarmblood stallions, 19/416 [4.6%]; P = 0.45). Rhabdomyolysis was the most commonly reported clinical finding for both warmblood horses with PSSM1 (12/14 [85.7%]) and nonwarmblood horses with PSSM1 (223/303 [73.6%]). There were no significant differences in the proportion of horses with reported gait abnormalities (P = 0.12), rhabdomyolysis (P = 0.53), muscle atrophy (P = 0.99), fasciculations (P = 0.99), and signs of shivers (P = 0.99) between warmblood horses with PSSM1 and nonwarmblood horses with PSSM1.
Genotype for the GYS1 mutation in nonwarmblood and warmblood horses with suspected myopathy.
| Phenotype | Nonwarmbloods | Warmbloods |
|---|---|---|
| P/P | 35 (2.8) | 0 (0) |
| P/N | 408 (32.2) | 16 (6.3) |
| N/N | 825 (65.1) | 237 (93.7)* |
| Total | 1,268 (100.1)† | 253 (100) |
Values reported are number (percentage).
Value differs significantly (P < 0.001) from the value for nonwarmblood horses.
Values in column do not sum to 100% because of rounding.
N/N = Homozygous for normal alleles (ie, no GYS1 mutation). P/N = Heterozygous for the GYS1 mutation. P/P = Homozygous for the GYS1 mutation.
Warmblood horses with PSSM2 compared with warmblood horses without PSSM
There were 188 warmblood horses with PSSM2 and 278 warmblood horses without PSSM (Table 2). Breeds of warmblood horses with PSSM2 and without PSSM were extremely similar (Table 1). There was no significant difference in age at time of biopsy (P = 0.38) or sex distribution (P = 0.41) between warmblood horses with PSSM2 or without PSSM. Similarly, there was no significant difference in the proportion of horses with muscle fasciculations (P = 0.99), rhabdomyolysis (P = 0.37), or signs of shivers (P = 0.87) between warmblood horses with PSSM2 or without PSSM. However, significantly (P = 0.02) more warmblood horses with PSSM2 had gait abnormalities, compared with the number of warmblood horses without PSSM that had gait abnormalities. Additionally, significantly (P = 0.03) fewer warmblood horses with PSSM2 had muscle atrophy, compared with the number of warmblood horses without PSSM that had muscle atrophy.
Warmblood horses with PSSM2 compared with nonwarmblood horses with PSSM2
There were 646 nonwarmblood horses with PSSM2. Warmblood horses with PSSM2 were significantly (P < 0.001) older and comprised more geldings and fewer mares and stallions, compared with results for nonwarmblood horses with PSSM2 (Table 2). Significantly (P < 0.001) fewer warmblood horses with PSSM2 had rhabdomyolysis, compared with the number of nonwarmblood horses with PSSM2 that had rhabdomyolysis. A significantly (P < 0.001) higher proportion of warmblood horses with PSSM2 had a gait abnormality, compared with the proportion of nonwarmblood horses with PSSM2 that had a gait abnormality. Signs of shivers were significantly (P = 0.001) more common in warmblood horses with PSSM2 than in nonwarmblood horses with PSSM2. There was no significant difference in muscle fasciculations (P = 0.09) or muscle atrophy (P = 0.49) between warmblood horses with PSSM2 and nonwarmblood horses with PSSM2.
Gait abnormalities and PSSM2
Nine warmblood horses (4 mares, 4 geldings, and 1 stallion; mean ± SD age, 7.9 ± 3.0 years) with PSSM2 underwent a detailed lameness examination performed by a veterinarian at the University of Minnesota (n = 2) or a lameness examination that included nuclear scintigraphy performed by a referring veterinarian (7). Results of nuclear scintigraphy were unremarkable for 5 of the 7 horses; for the other 2 horses, there was no response to treatment for the affected area (tarsalmetatarsal joint in one horse and the third trochanter of the other horse) identified on the bone scan. The predominant clinical sign for the 9 horses was a grade 1 or 2 lameness of the hind limbs in 4 horses, short strides or refusal to move forward in 3 horses, mild lameness in multiple limbs in 1 horse, and lameness in the hind limbs and a shivers-like gait in 1 horse. Six of these horses also had generalized stiffness. Muscle atrophy or weakness was reported for 4 horses, and 2 horses had signs of pain on palpation of the dorsal musculature. None of these 9 horses had episodes of rhabdomyolysis. A clear orthopedic cause of lameness that would account for the clinical signs was not identified for 8 horses; a chronic injury of the suspensory apparatus of the left hind limb was identified in the remaining horse.
Serum CK and AST activities
Serum CK activities were reported for 9 of 16 warmblood horses with PSSM1, 241 of 430 nonwarmblood horses with PSSM1, 61 of 188 warmblood horses with PSSM2, 336 of 646 nonwarmblood horses with PSSM2, and 95 of 278 warmblood horses without PSSM. There was not a significant (P = 0.63) difference in the proportion of warmblood horses with PSSM1 and nonwarmblood horses with PSSM1 that had CK activity > 600 U/L (7/9 [78%] and 170/241 [71%], respectively). There was a significantly (P < 0.001) lower proportion of warmblood horses with PSSM2 that had CK activity > 600 U/L (22/61 [36%]), compared with the proportion of nonwarmblood horses with PSSM2 that had CK activity > 600 U/L (203/336 [60%]). There was no significant (P = 0.60) difference in the proportion of horses with CK activity > 600 U/L between warmblood horses with PSSM2 and warmblood horses without PSSM (22/61 [36%] and 29/95 [31%], respectively).
Serum AST activity was reported for 8 of 16 warmblood horses with PSSM1, 213 of 430 nonwarmblood horses with PSSM1, 63 of 188 warmblood horses with PSSM2, 308 of 646 nonwarmblood horses with PSSM2, and 91 of 278 warmblood horses without PSSM. There was not a significant difference (P = 0.68) in the proportion of warmblood horses with PSSM1 and nonwarmblood horses with PSSM1 that had AST activity > 600 U/L (7/8 [88%] and 156/213 [73%], respectively). Similar to results for CK activity, there was a significantly (P = 0.003) lower proportion of warmblood horses with PSSM2 that had AST activity > 600 U/L, compared with the proportion of nonwarmblood horses with PSSM2 that had AST activity > 600 U/L (21/63 [33%] and 165/308 [54%], respectively). There was no significant (P = 0.28) difference in the proportion of horses with AST activity > 600 U/L between warmblood horses with PSSM2 and warmblood horses without PSSM (21/63 [33%] and 23/91 [25%], respectively).
Mean CK and AST activities in the small number of warmblood horses with PSSM1 were not significantly different from values for any other group. Mean CK and AST activities in warmblood horses with PSSM2 were similar to values for warmblood horses without PSSM and significantly (P < 0.001) lower than values for nonwarmblood horses with PSSM1 and nonwarmblood horses with PSSM2 (Figure 1).
Mean ± SD serum CK (A) and AST (B) activities in horse classified on the basis of breed and diagnosis. *†Value differs significantly (*P = 0.01; ‡P < 0.001) from the value for the warmblood horses with PSSM2 and warmblood horses with no PSSM. nonWB = Nonwarmblood horse. WB = Warmblood horse.
Citation: American Journal of Veterinary Research 78, 11; 10.2460/ajvr.78.11.1305
Muscle glycogen concentration
Muscle glycogen concentration was significantly (P < 0.001) higher in both warmblood horses with PSSM1 and nonwarmblood horses with PSSM1, compared with concentrations for all other groups (Figure 2). Post hoc tests revealed no difference in muscle glycogen concentration between warmblood horses with PSSM1 and nonwarmblood horses with PSSM1. Glycogen concentrations did not differ significantly among warmblood horses with PSSM2, warmblood horses without PSSM, and nonwarmblood horses without PSSM.
Mean ± SD muscle glycogen (wet-weight) concentration in horses classified on the basis of breed and diagnosis. *Value differs significantly (P < 0.001) from the value for the warmblood horses with PSSM2, warmblood horses with no PSSM, and nonwarmblood horses with no PSSM. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 78, 11; 10.2460/ajvr.78.11.1305
Discussion
The present study included data for > 500 warmblood horses. Most of the warmblood horses with PSSM had PSSM2, and only 15 of 253 [6%] warmblood horses that underwent muscle biopsy and GYS1 testing were identified with PSSM1. The much higher prevalence of PSSM2 than PSSM1 found in warmblood horses is consistent with results of an earlier study9 that included fewer horses (n = 72) from the same database. In that study,9 92% of the warmblood horses with PSSM diagnosed by histologic examination of muscle specimens had PSSM2, with only 8% having the GYS1 mutation responsible for PSSM1. These numbers do not reflect the prevalence of PSSM within warmblood horses, but merely indicate the prevalence within a population of horses from which muscle biopsy specimens were obtained because of suspected muscle disease.
Rhabdomyolysis was a consistent clinical feature of PSSM1 across breeds (14/16 [85.7%] warmblood horses with PSSM1 and 223/303 [73.6%] nonwarmblood horses with PSSM1). This is in agreement with results for several other studies2,3,8,12,13 that confirmed the presence of rhabdomyolysis through elevations in serum CK activity. It should be noted that the definition of rhabdomyolysis in the present study was the clinical interpretation made by the referring veterinarian and, when available, corroboration through abnormal elevations in serum muscle enzyme activities. Mean CK and AST activities did not differ significantly between warmblood horses with PSSM1 and other groups in the present study, which was likely attributable to the small sample size of warmblood horses with PSSM1. For horses with PSSM1, the age at muscle biopsy and sex distribution were similar between warmblood and nonwarmblood breeds. No sex predilection has been reported for Quarter Horse–related breeds with PSSM1.14 However, Selle Français mares heterozygous for the GYS1 mutation were found to have a 4.8-fold increase in the risk of developing exertional rhabdomyolysis, compared with the risk for geldings to develop exertional rhabdomyolysis, in a study8 of 22 related warmblood horses with PSSM1. In the present study, a disproportionate number of warmblood horses with PSSM1 were crosses with nonwarmblood breeds (5/16 [31.3%]), compared with the number of crosses for other groups of horses (13/189 [12.2%] warmblood horses with PSSM2 and 25/278 [9%] warmblood horses without PSSM), which suggested that outcrossing to nonwarmblood breeds may have been a source of the relatively rare GYS1 mutation in warmblood horses.
Although signalment and clinical signs were similar among horses with PSSM1, there were marked differences in age, sex, rhabdomyolysis, and primary clinical signs between warmblood horses with PSSM2 and nonwarmblood horses with PSSM2. Warmblood horses with PSSM2 were older at the time of muscle biopsy and comprised a higher proportion of geldings, compared with results for nonwarmblood horses with PSSM2. Older age and a higher proportion of geldings were also a feature of all warmblood horses in the Neuromuscular Diagnostic Laboratory database, compared with results for all other breeds in the database. Thus, the difference in signalment between warmblood horses with PSSM2 and nonwarmblood horses with PSSM2 may not have been a specific feature of warmblood horses with PSSM2. Rather, this may have reflected the fact that warmblood horses are typically used in disciplines that begin regular exercise at a later age than for other breeds, geldings may be preferred more strongly in typical warmblood disciplines, and fewer warmblood horses may be kept as breeding stallions because of the stallion approval process required by many warmblood stud-books. The distribution of warmblood breeds within the 188 warmblood horses with PSSM2 was similar to the entire population of 581 warmblood horses in the database, which indicated that there was not a specific warmblood breed with a predilection for PSSM2.
Rhabdomyolysis was not the primary clinical sign for warmblood horses with PSSM2. Rhabdomyolysis was reported in 39 of 147 (26.5%) warmblood horses with PSSM2, whereas it was reported in 12 of 14 (85.7%) warmblood horses with PSSM1 and, notably, 221 of 436 (50.7%) nonwarmblood horses with PSSM2. Furthermore, mean CK and AST activities were lower and the proportion of horses with activities > 600 U/L was lower for warmblood horses with PSSM2 than for nonwarmblood horses with PSSM1 or PSSM2. This is in agreement with a previous report13 of lower serum CK activity in warmblood horses with PSSM2 than in warmblood horses with PSSM1. In 97 of 147 (66%) warmblood horses with PSSM2, a gait abnormality was the primary clinical sign reported, which was significantly greater than the percentage for nonwarmblood horses with PSSM2 (176/436 [40%]) and warmblood horses with PSSM1 (1/16 [7%]). The specific gait abnormality did not appear to be related to shivers because 12% to 13% of warmblood horses in the database from which muscle biopsy specimens were collected had clinical signs of shivers, regardless of whether or not PSSM2 was diagnosed. On the basis of a review of in-depth lameness examinations of a small number of horses, the gait abnormality of warmblood horses with PSSM2 was often described as a stiff gait, shortened stride, reluctance to move forward, and mild hind limb lameness without a clear orthopedic cause. The retrospective nature of the study reported here and the variable, often circumspect, clinical descriptions on submission forms precluded a more thorough description of this gait abnormality. We propose that the gait abnormality of warmblood horses with PSSM2 could be attributable to primary muscle pain; however, a prospective study with standardized lameness evaluations and histologic examination of muscle biopsy specimens will be needed to provide a better description of the altered gait of warmblood horses with PSSM2.
Results of the present study of chilled muscle biopsy specimens shipped to the laboratory confirmed that PSSM1 was a glycogen storage disorder, with muscle glycogen concentrations approximately 1.4-fold as high in PSSM1 muscle as in muscle with no evidence of PSSM or other histopathologic abnormalities. Increased glycogen storage with PSSM1 arises from a gain-of-function mutation in GYS1 and increased glycogen synthase activity.7 In contrast to results for PSSM1 muscle, mean glycogen concentrations for 13 warmblood horses with PSSM2 in the present study were similar to concentrations in muscle without evidence of PSSM. This finding is important and worthy of further study because current dietary recommendations for horses with PSSM2 are the same as for horses with PSSM1, which are intended to decrease muscle glycogen concentrations by limiting dietary nonstructural carbohydrate content and insulin stimulation of glycogen synthase.13 Muscle glycogen concentrations likely decreased during shipping of chilled samples; however, all samples were handled in a similar manner, which mitigated any potential bias between diagnostic categories created because we did not analyze snap-frozen samples. If PSSM2 is a heterogeneous group of muscle diseases, analysis of glycogen content in a larger number of horses would be needed to determine whether excessive glycogen storage occurs in individual horses with PSSM2.
In contrast to the characteristic feature of amylase-resistant polysaccharide for PSSM1, a diagnosis of PSSM2 is based on subjective quantification of the aggregates of amylase-sensitive polysaccharide in muscle fibers.4–6 Controversy has existed as to whether these criteria accurately describe a specific myopathy.2,14 A study10 of muscle biopsy specimens collected from horses necropsied for any cause revealed that 50 of 225 (22%) horses had increased amylase-sensitive glycogen consistent with a diagnosis of PSSM. However, for 17 of those horses, the increase in PAS staining was attributable to cytoplasmic masses, rather than glycogen, which made the prevalence of increased amylase-sensitive glycogen 15%. False-positive and -negative inclusion of horses identified with PSSM2 is possible and affected by the histologist's experience, tissue handling artifacts, glycogen depletion during delays in chilling or shipping of samples, and elevated glycogen concentrations in trained horses. On the basis of the differences in clinical signs between other breeds with PSSM2 and warmblood horses with PSSM2, it is possible that a diagnosis of PSSM2 represents a histopathologic description of glycogen aggregates, rather than a specific myopathy. However, clinical signs of a gait abnormality did appear to be closely associated with PSSM2 in warmblood horses in the present study. Research into potential causes of PSSM2 would be strengthened by further defining the frequency of amylase-sensitive aggregates in healthy fit horses, using improved specific histopathologic diagnostic criteria, and performing studies within (rather than across) breeds. Investigators of a recent study15 that involved the use of electron microscopic and immunohistochemical evaluation of muscle biopsy specimens obtained from Arabian horses with and without a history of exertional rhabdomyolysis found that a subset of Arabians with a previous histopathologic diagnosis of PSSM2 or recurrent exertional rhabdomyolysis had abnormal aggregates of desmin within myofibers. Aggregates of the cytoskeletal intermediate filament desmin occur in a number of muscle diseases, including disorders affecting the Z-disc (termed myofibrillar myopathies).16 Because Arabian bloodlines have been incorporated into warmblood breeds, immunohistochemical and electron microscopic evaluation of muscle biopsy specimens collected from warmblood horses with gait abnormalities could be of value in better defining PSSM2 in warmblood horses.
In the study reported here, warmblood horses with a suspected myopathy much more commonly had PSSM2 than PSSM1. In contrast to primary clinical signs for nonwarmblood horses with PSSM1 and PSSM2, rhabdomyolysis was not the primary clinical sign for warmblood horses with PSSM2. Rather, a gait abnormality characterized by stiffness and mild hind limb lameness was the most common clinical sign for warmblood horses with PSSM2. In contrast to results for horses with PSSM1, excessive glycogen storage was not found in warmblood horses with PSSM2, as determined on the basis of measured muscle glycogen concentrations.
Acknowledgments
Supported by the Endowment of the Mary Anne McPhail Equine Dressage Chair.
Dr. Valberg is a member of the group that licenses the genetic test for PSSM1 mentioned in this report and receives royalties from sale of the test.
The authors thank Keri Gardner for technical assistance.
ABBREVIATIONS
| AST | Aspartate aminotransferase |
| CK | Creatine kinase |
| GYS1 | Glycogen synthase 1 gene |
| PAS | Periodic acid–Schiff |
| PSSM | Polysaccharide storage myopathy |
| PSSM1 | Type 1 polysaccharide storage myopathy |
| PSSM2 | Type 2 polysaccharide storage myopathy |
Footnotes
Prism, version 5.0, GraphPad Software Inc, La Jolla, Calif.
References
1. Hunt LM, Valberg SJ, Steffenhagen K, et al. An epidemiological study of myopathies in warmblood horses. Equine Vet J 2008; 40: 171–177.
2. McGowan CM, McGowan TW, Patterson-Kane JC. Prevalence of equine polysaccharide storage myopathy and other myopathies in two equine populations in the United Kingdom. Vet J 2009; 180: 330–336.
3. Stanley RL, McCue ME, Valberg SJ, et al. A glycogen synthase 1 mutation associated with equine polysaccharide storage myopathy and exertional rhabdomyolysis occurs in a variety of UK breeds. Equine Vet J 2009; 41: 597–601.
4. McCue ME, Armien AG, Lucio M, et al. Comparative skeletal muscle histopathologic and ultrastructural features in two forms of polysaccharide storage myopathy in horses. Vet Pathol 2009; 46: 1281–1291.
5. Valberg SJ, Cardinet GH III, Carlson GP, et al. Polysaccharide storage myopathy associated with recurrent exertional rhabdomyolysis in horses. Neuromuscul Disord 1992; 2: 351–359.
6. Valentine BA, McDonough SP, Chang YF, et al. Polysaccharide storage myopathy in Morgan, Arabian, and Standardbred related horses and Welsh-cross ponies. Vet Pathol 2000; 37: 193–196.
7. McCue ME, Valberg SJ, Miller MB, et al. Glycogen synthase (GYS1) mutation causes a novel skeletal muscle glycogenosis. Genomics 2008; 91: 458–466.
8. Johlig L, Valberg SJ, Mickelson JR, et al. Epidemiological and genetic study of exertional rhabdomyolysis in a warmblood horse family in Switzerland. Equine Vet J 2011; 43: 240–245.
9. McCue ME, Valberg SJ, Lucio M, et al. Glycogen synthase 1 (GYS1) mutation in diverse breeds with polysaccharide storage myopathy. J Vet Intern Med 2008; 22: 1228–1233.
10. Valentine BA, Cooper BJ. Incidence of polysaccharide storage myopathy: necropsy study of 225 horses. Vet Pathol 2005; 42: 823–827.
11. Passonneau JV, Gatfield PD, Schulz DW, et al. An enzymic method for measurement of glycogen. Anal Biochem 1967; 19: 315–326.
12. Ribeiro WP, Valberg SJ, Pagan JD, et al. The effect of varying dietary starch and fat content on serum creatine kinase activity and substrate availability in equine polysaccharide storage myopathy. J Vet Intern Med 2004; 18: 887–894.
13. Valberg SJ. Muscling in on the cause of tying-up, in Proceedings. 58th Annu Conv Am Assoc Equine Pract 2012; 85–123.
14. Firshman AM, Valberg SJ, Bender JB, et al. Epidemiologic characteristics and management of polysaccharide storage myopathy in Quarter Horses. Am J Vet Res 2003; 64: 1319–1327.
15. Valberg SJ, McKenzie EC, Eyrich LV, et al. Suspected myofibrillar myopathy in Arabian horses with a history of exertional rhabdomyolysis. Equine Vet J 2016; 48: 548–556.
16. Selcen D, Ohno K, Engel AG. Myofibrillar myopathy: clinical, morphological and genetic studies in 63 patients. Brain 2004; 127: 439–451.
