Abstract
In veterinary medicine, serum creatine-kinase (CK) activity is mostly used to assess skeletal muscle damage. This retrospective study aimed to evaluate the prevalence of increased CK activity in a large, ill-cat population and to characterise associated diseases, clinical and laboratory findings and its prognostic value. Cats with a complete serum biochemistry analysis were consecutively enrolled, divided into two CK activity-based groups (within and above reference interval) and compared. The study included 601 cats. Median serum CK was 402 U/l (range 16–506870). Increased CK (>250 U/l) was observed in 364 (60%) cats, and>30-fold its upper reference limit in 43 (7%). Cats with increased CK had greater (P≤0.05) body weight, and were more likely to have a history of collapse, dyspnoea, abnormal lung sounds, cyanosis, shock and paraplegia, higher median serum alanine aminotransferase, aspartate aminotransferase and lactate dehydrogenase activities and total bilirubin and triglyceride concentrations, but lower, median total protein, albumin, globulin and cholesterol concentrations and proportion of anorexia than cats with normal CK. Cardiac diseases, trauma, bite wounds, systemic bacterial infections, prior anaesthesia and intramuscular injections were more common (P≤0.05) in cats with increased compared to normal CK activity. The hospitalisation period was longer (P=0.007) and treatment cost and mortality were higher (P<0.005) in cats with increased CK activity. However, CK activity was an inaccurate outcome predictor (area under the receiver operator characteristics curve 0.58). Increased CK activity is very common in ill cats.
Creatine kinase (CK) is an enzyme that catalyses the reversible phosphorylation of adenosine diphosphate (ADP) from creatine phosphate resulting in creatine and adenosine triphosphate (ATP). 1,2 The rate and direction of the reaction strongly depends on the cytosolic ATP to ADP ratio. CK is a dimer, and two structurally different subunits, with identical catalytic activity are recognised. Four CK isoenzymes have been identified: CK-1, (Brain-brain, BB), CK-2 (Muscle-brain, MB), CK-3 (Muscle-muscle, MM) are cytosolic enzymes present in high concentrations in the central nervous system (CNS), skeletal and cardiac muscle and skeletal muscle, respectively; mitochondrial CK (CK-Mt), another isoenzyme of CK, is present in many tissues. 2 Serum CK activity has been studied extensively in horses, cattle and dogs. 1,2 The serum half-life of CK in cats has never been published, but in dogs it is less than 2 h. 1,2
In veterinary medicine, serum CK activity is mostly used to assess skeletal muscle damage, and to a lesser extent, myocardial injury. It is also used to interpret serum activities of other enzymes, including aspartate aminotransferase (AST), alanine aminotransferase (ALT) and lactate dehydrogenase (LDH).
Interpretation of increased serum CK activity should be made in caution. First, kittens have a higher reference interval for serum CK activity compared to adult cats. 3 Second, spurious increases in serum CK activity may result from in vitro haemolysis. 1,2,4 Third, trauma, intramuscular injections and restraint have been mentioned to cause increased CK activity in cats. 3
Increased serum CK activity has been reported in several disease conditions in cats, including caudal vena caval and arterial thrombosis, 5,6 cardiac diseases, 7,8 easter lily and chlorpyrifos poisonings, 9,10 several types of myopathies, 5,11–17 hepatozoonosis, 18 and metabolic acidosis. 19 Polymyositis, although uncommon in cats, 20 can lead to high serum CK activity in affected cats. 21 Most of the above reports are single case studies and small case series of specific diseases.
Increased serum CK activity has been reported in 25 hospitalised anorectic cats that received nutritional support via nasoesophageal tubes and compared to 25 non-anorectic cats. 22 Serum CK activity decreased significantly in the anorectic cats 48 h following initiation of nutritional support. The authors concluded that serum CK activity could serve as a useful marker of the nutritional status in ill cats. In that study, significant positive correlations between serum CK activity and LDH as well as AST were observed (r=0.41 and r=0.59, respectively). 22
Despite routine inclusion in commercial biochemical profiles, there are no studies correlating CK with clinical and laboratory abnormalities. To date, there are no available large-scale studies evaluating the diagnostic and prognostic value of serum CK activity in cats, or its correlation with other clinical and laboratory abnormalities. The aims of this observational retrospective study were to evaluate the occurrence of increased serum CK activity in a large population of ill cats admitted to a veterinary hospital and to identify the diseases, clinical history, and physical examination and laboratory findings with increased CK activity. In addition, we aimed to assess serum CK activity as a prognostic indicator.
Materials and methods
Animals, diagnoses and disease categories
The medical records of cats admitted to the Hebrew University Veterinary Teaching Hospital (2006–2008) were reviewed retrospectively. Cats were consecutively enrolled in the study if a complete serum biochemistry analysis, including serum CK activity, was performed during their disease course. Cats with haemolysed sera and those that have undergone prior surgery were excluded. The data retrieved from the medical records included the signalment, history, physical examination and laboratory findings, presence of prior anaesthesia or intramuscular injections, hospitalisation period, treatment cost and 30-day survival post discharge from the hospital. Cats presenting with tachycardia, tachypnoea, weak peripheral pulse, pale mucous membranes, hypothermia and cold extremities were clinically considered in a state of shock. Non-survivors included cats that died naturally as well as euthanased cases.
The number of diagnoses and disease categories exceeds the number of cats, because some cats presented with more than a single diagnosis or diseases category. Sepsis was diagnosed by the attending clinicians if cats presented signs of a systemic inflammatory response, namely severe clinical signs associated with shock (ie, tachycardia, tachypnoea, marked depression, fever or hypothermia), neutropenia or neutrophilia with left shift and severe cytoplasmic toxicity and evidence of organ failure. ‘Systemic bacterial infections’ included, in addition to sepsis, cases of infections accompanied by systemic clinical signs, that did not fulfil the above criteria for sepsis (eg, they did present fever, depression and anorexia but did not have evidence of systemic inflammatory response syndrome (SIRS), laboratory evidence of sepsis or organ failure). Acute gastritis included cats that presented with a history of acute vomiting, with no additional systemic clinical signs or laboratory and imaging abnormalities, and responded to supportive symptomatic therapy within less than 24 h. Cats diagnosed with bite wounds in the study did not include animals with abscesses; the latter were included in a separate group. Cardiac diseases included cardiomyopathy (eg, hypertrophic, restrictive, dilated and hypertrophic-obstructive) and congestive heart failure due to degenerative mitral valve disease (DMVD) and myocarditis. The diagnosis of cardiomyopathy and DMVD was based on an echocardiogram performed by a board-certified cardiologist. Myocarditis was diagnosed based on necropsy and histopatholgy. The category of poisoning included the following toxins: anticoagulants, organophosphates (OPC), carbamates, pyrethroids, ivermectin and metaldehyde. Other disease categories included metabolic, nutritional, neoplastic, immune-mediated, infectious (bacterial, viral, rickettsial, helminthic and protozoal), idiopathic, traumatic and vascular.
Laboratory methods
Samples for complete blood count (CBC) and serum biochemistry analysis were collected in potassium-ethylenediamineteteraacetic acid (EDTA) and plain tubes, respectively. A CBC was performed within 30 min from collection using an automatic impedance haematology analyser (Abacus or Arcus, Diatron, Wien, Austria). The differential white blood cell (WBC) count and the morphological assessment of blood cells were performed manually by examination of modified Wright's-stained blood smears (Hema-Tek 2000 Slide Stainer, model 4488B, Bayer Corporation, Elkhart, IN, USA. Stain: Hema-Tek stain pack; Modified Wright's Stain). Samples for serum biochemistry analysis were allowed to clot at room temperature for 30 min and then centrifuged; sera were separated and either analysed immediately or stored at 4°C pending analysis within 24 h from collection using a wet chemistry autoanalyser (Cobas-Mira, Roche, Rottkreutz, Switzerland, at 37°C). Analysis of CK activity was undertaken using a two part liquid reagent set (Pointe Scientific, Lincoln Park, Michigan, USA). Briefly, the method couples three enzymatic reactions (CK, hexokinase and glucose-6-phosphate dehydrogenase) of which the latter two are auxiliary. The rate of nicotineamide adenine dinucleotide (NADH) formation, measured at 340 nm, is directly proportional to serum CK activity. Electrolyte analysis (Na+, Cl−, Ca2+ and K+) was performed using an ion-selective electrolyte analyser (OmniC, Roche, Germany). Samples for coagulation tests (prothrombin and activated partial thromboplastin times) were collected into citrate tubes, the plasma separated by centrifugation was analysed within 30 min from collection (Automatic coagulometric analyser, ACL200, Instrumentation Laboratories, Warrington, UK or manual coagulometric analyser, KC1A, micro, Amelung, Germany).
Statistical analysis
The distribution (normal versus non-normal) of all continuous parameters was assessed using the Shapiro–Wilk's test. Normally and non-normally distributed continuous parameters were compared between cats with CK activity within and above reference interval (≤250 and >250 U/l, respectively) using Student's t- or Mann–Whitney tests, respectively. Fisher's exact or Pearson's χ2 tests were used to compare the categorical variables and proportion of diagnoses and disease categories between groups. Logistic regression analysis was performed to assess the relationship of CK activity with the outcome. The correlations between continuous laboratory variables were assessed using Pearson's and Spearman correlation tests for normally or non-normally distributed variables, respectively. Correlations were considered weak or moderate if r<0.5 and 0.5–0.75, respectively.
Cats with increased serum CK activity were divided into groups for further analyses. First, groups were constructed based on serum CK activity of five-, 10-, 20- and 30-fold above the upper limit of the reference interval (URL). Secondly, groups were constructed based on a division of the cats with increased serum CK activity into quartiles. Logistic regression analyses were then performed, based on these divisions, using CK activity class as a categorical variable against the outcome, while cats with normal CK activity served as the reference category. Continuous variables were compared between these CK activity groups using one-way analysis of variance (ANOVA) with Bonferonni's correction of α for multiple comparisons. The association of CK activity with the outcome was also assessed by the receiver operating characteristics (ROC) procedure, and its area under the curve (AUC) was calculated, and cutoff points, with their corresponding sensitivity and specificity for prediction of the outcome selected. The following definition was used to assess the accuracy of the ROC curve AUC; low (0.5<AUC<0.7), moderate (0.7<AUC<0.9) and high (0.9<AUC<1). 23 The optimal cutoff point was defined as the point that was associated with the least number of misclassifications. For all tests applied, P≤0.05 was considered statistically significant. All calculations were performed using statistical software (SPSS 15.0 for Windows, SPSS; Chicago, IL, USA).
Results
The study included 601 cats with a median age of 84 months (range 2–264) of which 274 (45.9%) were males (169 castrated) and 325 (53.1%) were females (197 spayed) of the following breeds: domestic shorthair (430, 71.5%), Persian (60, 10%), Siamese (33, 5.5%), domestic longhair (31, 5.2%), Himalayan (20, 3.3%) and several other breeds as well as mixed breed (crosses of pure-breed cats with domestic short- or longhair cats, 4.5%). There were no differences in age, sex or breed distribution between the CK groups. The median body weight was 4.1 kg (range 0.3–11). Cats with increased CK (>250 U/l) had greater (P=0.048) median body weight (4.28 kg, range 0.3–11) compared to cats with normal CK activity (4.00 kg, range 1.1–8.5). The distribution of CK activity was non-normal. Normal serum CK activity (reference interval 0–250 U/l) was documented at presentation in 237 cats (39.4%). Median serum CK activity of all cats was 402 U/l (range 16–506870). Increased CK activity (>250 U/l) was observed in 364 (60.6%) cats. The proportion of cats with serum CK activity greater than five-, 10-, 20- and 30-fold the serum CK URL was 25.9%, 16.3%, 10.3% and 7.2%, respectively (152, 98, 62 and 43 cats, respectively). In this division, cats with serum CK activity >10-fold the CK URL for example, included also cats with CK activity > fivefold the CK URL and so on.
Compared to cats with normal serum CK activity, cats with increased serum CK activity were significantly more likely to have a history and clinical signs of collapse (seven [1.9%] versus zero [0%], P=0.046), dyspnoea (44 [12.1%] versus 10 [4.2%], P=0.001, odds ratio [OR] 3.12, 95% confidence interval [CI95%] 1.54–6.33), abnormal lung sounds (46 [12.6%] versus 15 [6.3%], P=0.013, OR 2.14, CI95% 1.17–3.93), cyanosis (16 [4.4%] versus two [0.8%], P=0.013, OR 5.40, CI95% 1.23–23.71), shock (70 [19.2%] versus 17 [7.2%], P<0.001, OR 3.08, CI95% 1.76–5.38) and paraplegia (27 [7.4%] versus two [0.8%], P<0.001, OR 9.41, CI95% 2.22–39.97). They were less likely to be anorexic (98 [41.4%] versus 120 [33.0%], P=0.037, OR 0.70, CI95% 0.50–0.98). Thirteen cats presented with seizures, of which two (0.3% of all cats) and 11 (1.8% of all cats) were in the normal and increased serum CK groups, respectively (P=0.073). There was no difference (P=0.622) in the proportion of seizures in cats in which serum CK activity was >30- or 20-fold URL (P=0.94 for both of these groups) as compared with cats with serum CK activity ≤30- and ≤20-fold URL, respectively.
There were no differences in haematology and coagulation tests results between cats with normal and increased serum CK activities. Several differences in serum biochemistry test results were observed between these groups (Table 1). Cats with increased CK activity (>250 U/l) had significantly higher serum activities of ALT, AST, LDH and concentrations of total bilirubin and triglycerides, but lower concentrations of total protein, albumin, globulin and cholesterol compared to cats with normal serum CK activity (Table 1).
Table 1.
Selected laboratory analytes and abnormalities in cats with normal and increased serum creatine kinase activity*.
| Analyte | All cats n 1 , median, (range) | CK 3 groups n 2 , median, (range) | P value ** | Reference interval | |
|---|---|---|---|---|---|
| CK 3 >250 U/l | CK 3 ≤250 U/l | ||||
| Albumin (g/l) | 598, 30.4, (10.7–67.1) | 364, 29.6, (10.7–56) | 234, 31.0, (16.6–67.1) | 0.002 | 29.0–46.0 |
| Total protein (g/l) | 600, 74, (34–119) | 364, 72, (34–118) | 236, 78, (46–119) | <0.0001 | 54.0–75.0 |
| Globulin (g/l) | 598, 43, (5.0–89.8) | 364, 42, (5.0–89.8) | 234, 45.2, (13.9–87.4) | <0.0001 | 17.0–45.0 |
| Total bilirubin (μmol/l) | 598, 4.19, (0.0–742.14) | 363, 4.96, (0.0–79.86) | 235, 3.59, (0.0–742.14) | 0.004 | 0.00–6.84 |
| Total calcium (mmol/l) | 595, 2.34, (0.98–3.44) | 360, 2.30, (0.98–3.44) | 235, 2.4, (1.50–3.36) | <0.0001 | 2.18–2.95 |
| Free calcium (mmol/l) | 368, 1.19, (0.53–1.85) | 242, 1.18, (0.53–1.85) | 126, 1.23, (0.66–1.80) | 0.025 | 1.10–1.30 |
| Cholesterol (mmol/l) | 600, 3.79, (1.19–15.75) | 364, 3.62, (1.19–11.64) | 236, 3.98, (1.73–15.75) | 0.0098 | 1.55–4.27 |
| Triglycerides (mmol/l) | 567, 0.91, (0.08–55.10) | 342, 1.05, (0.09–26.31) | 225, 0.73, (0.08–55.10) | <0.0001 | 0.17–1.17 |
| ALT 4 (U/l) | 600, 63, (0–2349) | 363, 66, (0–2349) | 237, 54, (8–786) | <0.0001 | 10–120 |
| AST 5 (U/l) | 600, 46, (0–1500) | 363, 57, (0–1500) | 237, 33, (3–1395) | <0.0001 | 10–50 |
| LDH 6 (U/l) | 108, 644, (18–4755) | 57, 1005, (166–4755) | 51, 444, (118–3070) | <0.0001 | 34–360 |
| Abnormality | Selected serum biochemistry abnormalities | P value† | ||
|---|---|---|---|---|
| n 7 (%) 8 | n 9 (%) 10 | n 11 (%) 12 | ||
| Increased ALT 4 | 128 (21.5) | 89 (24.7%) | 39 (16.5) | 0.017 |
| Increased AST 5 | 270 (45.7) | 205 (57.3) | 65 (27.9) | <0.001 |
| Hyperbilirubinemia | 195 (32.5) | 140 (38.6) | 55 (23.4) | <0.001 |
| Hypocalcemia | 174 (29.2) | 126 (35.0) | 48 (20.4) | <0.001 |
| Hypercholesterolemia | 218 (36.3) | 117 (32.5) | 101 (42.8) | 0.01 |
| Hyperproteinemia | 277 (46.2) | 138 (37.9) | 139 (58.9) | <0.001 |
| Hypoalbuminemia | 245 (41.0) | 168 (46.2) | 77 (32.9) | 0.004 |
| Hypertriglyceridemia | 410 (72.3) | 265 (77.5) | 145 (64.4) | 0.003 |
| Hyperglobulinemia | 246 (41.3) | 129 (35.5) | 117 (50.4) | 0.001 |
Only analytes with significant differences between CK groups are included.
Mann-Whitney test.
Pearson's chi-square
number of cats in the study in which an analyte was measured
number of cats in each CK group in which an analyte was measured
creatine kinase
alanine aminotransferase
aspartate aminotransferase
lactate dehydrogenase
number of cats presenting abnormality of all cats
percent of cats presenting abnormality of all cats
number of cats presenting abnormality of cats with increased CK activity
percent of cats presenting abnormality of cats with increased CK activity
number of cats presenting abnormality of cats with normal CK activity
percent of cats presenting abnormality of cats with normal CK activity.
Median serum AST activity (reference interval 0–50 U/l) was 46 (range 0–1500 U/l). Increased serum AST activity (>50 U/l) was present in 270 cats (45.7%). Serum AST activities greater than two-, five- and 10-fold URL activity were present in 17.2%, 4.3% and 1.7% of the cats (103, 26 and 10 cats, respectively). Median serum LDH activity (reference interval 24–360 U/l) was 644 (range 118–4755, measured in 108 cats). Increased serum LDH activity (>360 U/l) was present in 84 cats (77.8%). Serum LDH activity greater than five- and 10-fold URL were present in 6.5% and 1.9% of the cats, respectively (seven and two cats, respectively). There were significant (P<0.001) but weak positive correlations between CK activity and AST, LDH and ALT activities (r, 0.48, 0.41 and 0.21, respectively). There was a significant (P<0.001) moderate positive correlation between AST and ALT activities (r, 0.56). There were no additional significant correlations between serum CK activity and other serum biochemistry analytes.
The percentage of selected diseases and disease categories assigned to the ill cats are listed in Table 2. For each, the median CK activity and the range of activities are listed. Most notably, cardiac diseases in general, and specifically cardiomyopathy (of all kinds) were significantly (P≤0.05) more common in cats with increased compared to cats with normal serum CK activity, as were cases of trauma and bite wounds. Compared to cats with normal CK activity, cats with serum CK activity>7500 U/l (>30-fold URL) had a higher percentage of traumatic diseases (20.9 versus 4.5%, P<0.0001), aortic thromboembolism (ATE, 14 versus 1.6%, P<0.0001), bite wounds (14 versus 2.2%, P=0.01), high rise syndrome (11.0 versus 0.7%, P<0.0001) and sepsis (11.6 versus 0.7%, P<0.0001). Cardiomyopathy was diagnosed in 10/15 cats with ATE, and ATE was present in 10/40 cats with cardiomyopathy.
Table 2.
Percentage of diagnoses and disease categories (proportion > 1%) in decreasing order and serum creatine-kinase (CK) activity in 601 cats with increased or normal serum CK activity.
| Disease or disease category | Total n (% of all cats) | Median (range) CK 1 activity (U/L) | n (%) cats in each CK 1 group | P value | Odds ratio (95% confidence interval) | |
|---|---|---|---|---|---|---|
| >250 U/L | ≤250 U/L | |||||
| Infectious diseases (all) | 174 (28.9) | 403 (16–291730) | 104 (28.6) | 70 (29.5) | 0.80 | 0.95 (0.67–1.37) |
| Renal failure 2 | 99 (16.5) | 298 (52–41136) | 55 (15.1) | 44 (18.6) | 0.26 | 0.78 (0.51–1.21) |
| Inflammatory 3 | 88 (14.6) | 218 (51–9905) | 39 (10.7) | 49 (20.7) | 0.001 | 0.46 (0.29–0.73) |
| Infectious bacterial diseases | 86 (14.3) | 533 (18–97829) | 60 (16.5) | 26 (11.0) | 0.073 | 1.6 (0.98–2.62) |
| Hepatic lipidosis | 57 (9.5) | 368 (16–97829) | 33 (9.1) | 24 (10.1) | 0.67 | 0.89 (0.51–1.54) |
| Neoplasia 4 | 57 (9.5) | 300 (36–6321) | 26 (8.0) | 21 (8.9) | 0.116 | 0.65 (0.37–1.12) |
| Pancreatitis* | 51 (8.5) | 203 (53–17212) | 23 (6.3) | 28 (11.8) | 0.018 | 0.50 (0.28–0.90) |
| Cardiac diseases 5 * | 46 (7.6) | 657 (70–506870) | 39 (10.7) | 7 (3.0) | <0.0001 | 3.94 (1.73–8.97) |
| FIV 6 | 41 (6.8) | 209 (39–4474) | 19 (5.2) | 22 (9.3) | 0.054 | 0.54 (0.29–1.02) |
| Cardiomyopathy 7 * | 40 (6.7) | 803 (70–506870) | 37 (10.2) | 3 (0.5) | <0.0001 | 6.59 (2.32–18.74) |
| Gastrointestinal diseases* | 38 (6.3) | 223 (47–9427) | 17 (4.7) | 21 (8.9) | 0.039 | 0.50 (0.26–0.98) |
| FLUTD 8 | 34 (5.6) | 543 (85–61524) | 22 (6.0) | 12 (5.1) | 0.611 | 1.21 (0.59–2.49) |
| Trauma 9 | 34 (5.6) | 891 (188–97829) | 27 (7.4) | 7 (3.0) | 0.029 | 2.63 (1.13–6.15) |
| Gingivitis* | 34 (5.6) | 231 (33–16453) | 15 (4.1) | 19 (8.0) | 0.043 | 0.50 (0.25–0.99) |
| Hepatitis 10 | 32 (5.3) | 279 (57–4012) | 18 (4.9) | 14 (5.9) | 0.63 | 0.83 (0.40–1.70) |
| Neurologic diseases 11 | 30 (5.0) | 569 (71–12106) | 20 (5.5) | 10 (4.2) | 0.48 | 1.32 (0.61–2.87) |
| Systemic bacterial diseases | 29 (4.8) | 901 (114–97829) | 23 (6.3) | 6 (2.5) | 0.05 | 2.60 (1.04–6.48) |
| Upper respiratory diseases | 26 (4.3) | 238 (16–3382) | 12 (3.3) | 14 (5.9) | 0.12 | 0.54 (0.25–1.20) |
| Urinary tract infection | 25 (4.1) | 565 (52–14062) | 18 (4.9) | 7 (3.0) | 0.22 | 1.72 (0.71–4.2) |
| Pneumonia | 24 (4.0) | 404 (16–3076) | 14 (3.8) | 10 (4.2) | 0.82 | 0.91 (0.40–2.08) |
| Diabetes mellitus | 24 (4.0) | 251 (39–66744) | 12 (3.3) | 12 (5.1) | 0.28 | 0.64 (0.28–1.45) |
| Dermatologic diseases 12 * | 19 (3.2) | 206 (72–1832) | 9 (2.5) | 10 (4.2) | 0.23 | 0.58 (0.23–1.44) |
| Hemoplasmosis | 18 (3.0) | 451 (60–71545) | 11 (3) | 7 (3.0) | 1.00 | 1.02 (0.39–2.68) |
| Urinary tract obstruction | 18 (3.0) | 517 (65–4004) | 12 (3.3) | 6 (2.5) | 0.64 | 1.31 (0.49–3.55) |
| Bite wounds* | 18 (3.0) | 1735 (171–291730) | 16 (4.4) | 2 (0.8) | 0.013 | 5.40 (1.23–23.71) |
| Lymphoma | 18 (3.0) | 218 (44–3860) | 8 (2.2) | 10 (4.2) | 0.16 | 0.51 (0.20–1.31) |
| Arterial thromboembolism* | 15 (2.5) | 1852 (118–506870) | 15 (4.1) | 0 (0.0) | 0.002 | NA |
| Poisoning 13 * | 15 (2.5) | 1351 (150–15329) | 14 (3.8) | 1 (0.4) | 0.007 | 9.44 (1.23–72.27) |
| Mammary ACA 14 | 15 (2.5) | 219 (36–2783) | 7 (1.9) | 8 (3.4) | 0.27 | 0.56 (0.20–1.57) |
| Abscess | 12 (2.0) | 1065 (33–14062) | 9 (2.5) | 3 (1.3) | 0.3 | 1.98 (0.53–7.38) |
| FIP 15 | 11 (1.8) | 580 (84–30000) | 6 (1.6) | 5 (2.1) | 0.76 | 0.78 (0.24–2.58) |
| FPLV 16 | 9 (1.5) | 386 (92–1706) | 6 (1.6) | 3 (1.3) | 1.00 | 1.31 (0.32–5.28) |
| Sepsis* | 9 (1.5) | 14062 (278–97829) | 9 (2.5) | 0 (0.0) | 0.014 | NA |
| High rise syndrome* | 9 (1.5) | 12450 (435–32920) | 9 (2.5) | 0 (0.0) | 0.014 | NA |
| Pyothorax | 9 (1.5) | 972 (200–36100) | 8 (2.2) | 1 (0.4) | 0.096 | 5.30 (0.66–42.68) |
| Urolithiasis | 8 (1.3) | 782 (184–6747) | 6 (1.6) | 2 (0.8) | 0.49 | 1.97 (0.39–9.84) |
| Organophosphate poisoning | 7 (1.2) | 1351 (151–2905) | 6 (1.6) | 1 (0.4) | 0.25 | 3.96 (0.43–33.06) |
| Acute gastritis | 7 (1.2) | 466 (138–862) | 5 (1.4) | 2 (0.8) | 0.71 | 1.64 (0.32–8.51) |
Significant (P ≤ 0.05) difference between groups
creatine kinase activity (reference interval 0–250 U/l)
acute or chronic renal failure
inflammatory non-infectious diseases (e.g., pancreatitis)
all kinds of neoplasia combined
all types of cardiac diseases combined, including cardiomyopathy, congestive heart failure due to degenerative valve disease, myocarditis and congenital cardiac defects
feline immunodeficiency virus infection
all types of cardiomyopathy combined, including hypertrophic, restrictive, dilated and hypertrophic-obstructive cardiomyopathy
feline lower urinary tract disease
all types of trauma combined including hit by car, high rise syndrome and bite wounds
including cholangiohepatitis and cholangitis. Hepatic lipidosis and hepatic neoplasms are excluded
including cranial neoplasia, meningitis, meningoencephlitis, toxoplasmosis, epilepsy and undiagnosed neurologic disorders
including pyoderma, food allergy, atopy, pemphigus, and parasitic skin diseases
all poisonings combined, see text for details
adenocarcinoma
feline infectious peritonitis virus infection
feline paleukopenia virus infection; NA – not applicable.
Data regarding prior general anaesthesia were available in 301 cats. Cats with increased CK were more likely to have a history of prior general anesthesia than cats with normal CK (56/199 [28%] versus 14/102 [13.7%] cats, P=0.005, OR-2.46, CI95% 1.29–4.68). The effect of prior intramuscular injections (within 24 h prior to presentation) was assessed in 303 cats (201 and 102 cats with CK activity above and within reference interval, respectively) of which 93 cats received prior intramuscular injections. Median serum CK activity was significantly (P<0.0001) higher in cats that received prior intramuscular injections compared to those that did not (947, [range, 48–291730] and 351, [range, 24–506870] U/l, respectively,). Additionally, prior intramuscular injection was also more common in cats with serum CK activity above reference interval compared to cat with normal CK activity (37.5% versus 16.7%, respectively, P<0.0001) and was also more common in cats with CK activity >7500 U/l compared to cats with normal CK activity (45.5% versus 28.9%, P=0.051).
The median hospitalisation time for cats with both normal and increased serum CK activities was 2 days (range 0–18 and 0–43), however, it was significantly (P=0.008) longer in the latter. The median cost of treatment was significantly (P<0.0001) higher in cats with increased serum CK activity compared to those with normal serum CK activity (334 € [range, 58–1625] and 273 € [range, 40–3224], respectively). Of 601 cats, 453 (75.4%) survived while 148 (24.6%) did not, of which 88 (14.6%) died and 60 (9.9%) were euthanased. Mortality was significantly (P=0.002, OR 1.91, CI95% 1.28–2.85) more likely in cats with increased CK activity than in cats with normal CK activity (29.1% and 17.7%, respectively). In cats with serum CK activity >7500 U/l (43 cats) the mortality was 39.5% (17 cats) and was significantly (P=0.019, OR 2.13, CI95% 1.12–4.05) more likely than in those with CK activity ≤7500 U/l (23.5% of 558 cats). Logistic regression analysis using serum CK activity quartiles as a categorical variable, showed that cats in the high (fourth) quartile (CK>1264 U/l) had a significantly (P=0.011, OR 2.06, CI95% 1.18–3.60) greater mortality compared to cats with normal serum CK activity. Based on a ROC analysis, serum CK activity was an inaccurate predictor of death (AUC 0.58, CI95% 0.53–0.64). 22 The following cutoff points with their corresponding sensitivity and specificity of an accurate prediction of death were obtained from the ROC analysis; CK activity 550 U/l corresponded to sensitivity of 50% and specificity of 61%; CK activity 1000 U/l corresponded to sensitivity of 30% and specificity of 74%; CK activity 1800 U/l corresponded to sensitivity of 21% and specificity of 81%. Similar, although slightly better, results were observed for the other muscle enzymes, AST (AUC 0.59, CI95% 0.54–0.65) and LDH (AUC 0.69, CI95% 0.55–0.81).
Discussion
Serum CK activity has been measured routinely in cats for a long time, however, the frequency of its increased activity in ill cats, its overall diagnostic value and association with specific diagnoses has never been assessed in a large-scale study. The present study of over 600 cats showed that increased serum CK activity is very common, and its proportion was 61%. This finding is unexpected, as the primary diagnoses in most cats were not skeletal, muscular, or cardiac diseases. However, as CK has a high muscle-specificity, it has to be assumed that muscle injury probably existed in these cats. 1,2 Presence of muscle tissue damage is partially supported by the observed increases of other, although less muscle-specific enzyme activities, such as LDH and AST that were also present in many of the cats. This concurrent increase in CK and LDH activities and the significant correlation between LDH and CK activities is in agreement with previous findings in hospitalised cats. 22 Serum AST activity was also increased in many cats, however, this abnormality was less prevalent, and its magnitude in relation to its URL was smaller compared to CK. Leakage of AST from injured cells probably occurs in more severe cellular damage compared to leakage of CK and LDH. The latter two are mostly cytosolic enzymes, while AST has two isoenzymes, cytosolic and mitochondrial. Induction of significant leakage of mitochondrial AST in damaged cells occurs when cellular injury is more severe compared to the severity of injuries that induce leakage of cytosolic enzymes. 1,2 The activity of CK was positively correlated with AST and LDH activities, although these correlations were weak, probably reflecting lower muscle tissue specificity and longer half-lives of the latter. 1,2 The activity of AST was correlated, although only moderately, with ALT activity. This correlation possibly reflects liver damage, because AST also can leak when hepatocelullar injury occurs. Thus, AST is not a skeletal muscle-specific enzyme such as CK. In contrast, ALT is considered more hepato-specific than AST, and increased serum ALT activity usually reflects hepatocyte injury. Skeletal muscles of small animals contain some ALT, and when severe muscle injury occurs, serum ALT is expected to increase, although to a lesser extent compared to CK, AST and LDH activity. 1,2 This was probably reflected by the significant, though very weak, positive correlation between CK and ALT activities in the study. However, the possibility that liver damage was present in a considerable number of our cats and contributed to increased serum AST activity should not be overlooked, because inflammatory liver diseases, hepatic lipidosis and pancreatitis were common in our cats.
In this study, 10% of the cats had marked increases in serum CK activity that was >20-fold its URL, suggesting that severe muscle injuries in ill cats is not uncommon and is often severe. In human medicine, rhabdomyolysis, which is lysis of skeletal muscle, is characterised biochemically by myoglobinaemia, myoglobinuria and an increase of serum levels of muscle enzymes (in absence of myocardial injury), particularly CK activity. The latter should be increased at least fivefold URL along with the presence of myoglobinaemia and myoglobinuria. 24,25 This kind of definition remains to be established in cats. Serum or urinary myoglobin concentrations were not measured in this study. However, if CK activity is used as the sole measure for assessment of muscle damage, by applying the human criteria (ie, increased CK activity>fivefold its URL), roughly 25% of our cats would have satisfied the human criteria of rhabdomyolysis although they suffered from a wide variety of diseases, some of which were not primary muscular or myocardial diseases, and in some cats, presence of true rhabdomyolysis is questionable. Thus, the human criteria of rhabdomyolysis probably cannot be applied to cats. Additionally, in human patients, rhabdomyolysis is considered a severe systemic disease associated with high mortality rate. 25 In contrast, the mortality rate of cats with over a fivefold increase in the URL of CK activity was not significantly higher compared to cats with normal CK activity, suggesting that this similar biochemical abnormality of humans and cats does not represent a similar disease severity. Future investigations, including serum muscle enzymes activity levels, serum and urine myoglobin concentrations, serum cardiac troponin concentrations and possibly histological and ultrastructural studies of skeletal muscle sections should be performed to define rhabdomyolysis in cats, although the latter are mostly impractical ante-mortem in a clinical setting. At present, because of the high proportion of increased serum CK activity in ill cats, a diagnosis of true rhabdomyolysis, such that results from severe muscle necrosis, should probably reserved for cats in which CK activity is markedly (eg, 20–30-fold URL) increased.
In cats with serum CK activity above 30-fold its URL in which presence of severe skeletal muscle injury was highly probable, primary diseases could be divided into two types: obvious muscular lesions (ie, trauma and saddle thromboembolism) and systemic bacterial infections, including sepsis. In the former type of diseases, the presence of a markedly increased serum CK activity is not surprising due to crushing and ischaemia, respectively. 25 The pathogenesis of muscle damage in systemic bacterial infections is not completely clear and could be related to presence of fever, shock or direct skeletal muscle injury by infectious agents or their toxins. 25 Indeed, cats with shock were more likely to have increased serum CK activity, thus shock and its consequences might have played a role in skeletal muscle injury. Alternatively, it may have been due to a direct effect of bacterial toxins. A case of rhabdomyolysis associated with Escherichia coli infection has been previously reported in a cat and was attributed to endotoxaemia. 26 Similar cases in human patients have been reported with Gram-positive and -negative bacteria. 27–30 Endotoxaemia may lead to rhabdomyolysis through induction of hypovolaemia, shock and hypoperfusion of muscles resulting from vasoactive mediators (ie, histamine and kinins) release, decreased cardiac output and presence of disseminated intravascular coagulation with resultant multiple microthrombi in muscle tissue. 25,29 Some of the clinical signs that were more prevalent in cats with increased serum CK activity were suggestive of hypoxaemia (eg, dyspnoea, cyanosis and abnormal lung sounds) and shock, all of which can result in decreased oxygen delivery to skeletal muscles. Measurement of arterial and venous blood gases, pH and total CO2 could improve our understanding of the role of hypoxaemia and hypoperfusion in the pathogenesis of increased serum CK in cats and should be considered in future studies.
Hypokalaemia is listed as a potential cause of skeletal muscle damage and rhabdomyolysis in humans, dogs and cats, 25,31–33 however, in the present study there were no potassium concentration differences between groups, nor was hypokalaemia more prevalent in cats with increased serum CK activity. Potassium has a potential role in development of increased serum CK in specific diseases such as hypokalaemic myopathy and hyperaldosteronism, although rhabdomyolysis has not been reported in these conditions. 5,11,13,14,32 Hyperkalaemia, hypocalcaemia, and hyperphosphataemia are potential complications of skeletal muscle damage, 25,32,33 however, these abnormalities were not observed in our cats, even when the increase of serum CK was >7500 U/l. The possibility that rhabdomyolysis did not occur in our cats cannot be overlooked, however, alternatively, these muscle damage-related biochemical abnormalities may be uncommon in feline rhabdomyolysis or occur at later stages of the disease.
We have no obvious single explanation to account for the high proportion of mild to moderately increased serum CK activity in our general ill-cat population. Possibly, muscle damage occurs through multiple mechanisms that could include hypoxaemia, fever, inflammation, infection, adverse drug reactions and shock or hypovolaemia leading to skeletal muscle hypoperfusion and ischaemia. 25,29 Alternatively, excessive restraint and previous intramuscular injections and anaesthesia might have accounted for some of the observed increased CK activity, and all three might have been underestimated due to the nature of this retrospective study. Intramuscular injections have been previously documented with increased serum CK activity in dogs, sheep, horses and cattle in several studies. 1,2,34–38 However, in cats, although these were listed in a general text as causes of increased CK activity, this statement was not supported by previous studies. 1 The results of the present study support this statement, because serum CK activity was significantly increased in cats with a history of intramuscular injections of several medications compared to cats that were not injected prior to presentation. Historical evidence of intramuscular injection should, therefore, be included in the interpretation of serum CK activity in cats. It cannot be excluded that in some cats that were injected intramuscular, restraint played a role in muscle injury, because intramuscular injections are sometimes used to administer sedatives in aggressive cats, or alternatively, restraint was needed to administer the injection or collect blood samples.
Increased serum CK activity has previously been positively associated with anorexia in ill cats in a hospital setting 22 and is thought to result from muscle catabolism due to a negative energy balance. 1 However, in the present study, anorexia at presentation was significantly more common in cats with normal CK compared to cats with increased CK activity. Possibly, cats with increased serum CK activity had more acute conditions and did not have the time to develop anorexia, or for their owners to observe its presence. This hypothesis cannot be supported by our present results, because the duration of illness prior to presentation was not recorded in all cats. Anorexic cats were not followed in order to examine whether their serum CK activity had increased later during the disease course, nor were cats with increased CK activity followed in order to examine whether anorexia had developed later on. Thus, conclusions cannot be made with regards to serum CK activity as a marker for anorexia in cats.
Myocardial injury in heart diseases could also contribute to an increase in serum CK activity, however, it has been mentioned only occasionally as a cause for an increased serum CK activity in cats. 2 Cardiac diseases in general, and specifically cardiomyopathy were significantly more common in cats with increased compared to cats with normal CK in this study. Arterial thromboembolism was present in 25% of the cats diagnosed with cardiomyopathy and likely contributed significantly to the increase in serum CK activity that was observed in this group. In addition, myocardial ischaemia might have been also present in systemic bacterial infections, due to thrombosis and infarction. Measurement of serum cardiac troponins should help to determine if myocardial injury plays a significant role in the increase in serum CK activity.
Involuntary skeletal muscle contractions due to neurological disorders have been previously associated with increased total serum CK activity, due to increased CK-MM activity 1 and seizures were listed among the traumatic conditions that cause increased serum CK activity in animals in general. 2 In this study, seizures, observed in only 13 cats, tended to be more common in cats with increased serum CK activity, but the power of this statistical analysis is limited by this small number of cats. Presentation of a cat within a relatively short time after the seizure could lead to a relatively normal CK activity. Conversely, negative results can occur due to the short half-life of CK. 1 Due to the small number of cats with seizures in this study it was not possible to analyse the data with regards to the time lag from the seizure episode to presentation, and future investigations are warranted.
The results of this study clearly demonstrate that cats with increased CK activity present a higher proportion of systemic clinical signs, suggesting a more severe disease compared to cats with normal CK activity. This suggestion is also supported by a higher proportion of serum biochemistry abnormalities and differences in serum biochemistry values such as higher bilirubin and triglyceride and lower albumin concentrations. Thus, when serum CK is increased in cats, this should probably be a marker of a more severe disease. Other factors, such as intramuscular injections and excessive restraint, could also contribute to increased CK activity.
Development of acute tubular necrosis and renal failure is a significant complication in human rhabdomyolysis and was observed in 46% of 475 human hospitalised patients. 24,25,32 This study did not follow the serum biochemistry of cats with increased serum CK after presentation, nor did it record post mortem histopathology of the kidneys, and thus cannot shed light on the renal consequences of skeletal muscle damage and rhabdomyolysis in cats. Nevertheless, at presentation, there were no differences in the proportion of azotaemia or serum concentrations of creatinine, urea, phosphorus and potassium between cats with increased and those with normal serum CK activity.
This study has several limitations. First, this is a retrospective study, and some of the medical records were missing, limiting the statistical analysis. For example, data concerning presence of prior intramuscular injection or anaesthesia were available for only half of our cats and thus, these variables could only be analysed in these cats. Second, the medical records did not contain data concerning the extent of trauma induced by venepuncture during sampling of blood. A traumatic venepuncture could potentially lead to an increase in CK activity if muscle damage was induced although this has never been specifically reported in the veterinary literature as a potential cause for such an increase.
In summary, increased serum CK activity is very common in ill cats, even when no clear evidence of muscle damage is apparent, and was found as a marker of a more severe disease, longer hospitalisation and significantly higher treatment costs. Thus, the results of this study support inclusion of CK activity measurement in routine serum biochemistry analyses in ill cats. Serum CK activity was not a good predictor of the outcome, and its accuracy as an outcome predictor was not superior to that of other muscle enzymes (eg, AST and LDH). Nevertheless, when serum CK activity is extremely increased (>7500 U/l or 30-fold its URL), it is associated with a higher mortality and can serve as a negative prognostic indicator.
References
- 1.Hoffmann W.E., Solter P.F. Diagnostic enzymology of domestic animals. Kaneko J.J., Harvey J.W., Bruss M.L. Clinical biochemistry of domestic animals, 6th edn, 2008, Academic Press: San Diego, 351–378. [Google Scholar]
- 2.Stockham S.L., Scott M.A. Enzymes. Stockham S.L., Scott M.A. Fundamentals of veterinary clinical pathology, 2nd edn, 2008, Wiley-Blackwell: Ames, 640–674. [Google Scholar]
- 3.Levy J.K., Crawford P.C., Werner L.L. Effect of age on reference intervals of serum biochemical values in kittens, J Am Vet Med Assoc 228, 2006, 1033–1037. [DOI] [PubMed] [Google Scholar]
- 4.Jacobs R.M., Lumsden J.H., Grift E. Effects of bilirubinemia, hemolysis, and lipemia on clinical chemistry analytes in bovine, canine, equine, and feline sera, Can Vet J 33, 1992, 605–608. [PMC free article] [PubMed] [Google Scholar]
- 5.Rose S.A., Kyles A.E., Labelle P., et al. Adrenalectomy and caval thrombectomy in a cat with primary hyperaldosteronism, J Am Anim Hosp Assoc 43, 2007, 209–214. [DOI] [PubMed] [Google Scholar]
- 6.Smith S.A., Tobias A.H., Jacob K.A., Fine D.M., Grumbles P.L. Arterial thromboembolism in cats: acute crisis in 127 cases (1992–2001) and long-term management with low-dose aspirin in 24 cases, J Vet Intern Med 17, 2003, 73–83. [DOI] [PubMed] [Google Scholar]
- 7.Kidd L., Stepien R.L., Amrheiw D.P. Clinical findings and coronary artery disease in dogs and cats with acute and subacute myocardial necrosis: 28 cases, J Am Anim Hosp Assoc 36, 2000, 199–208. [DOI] [PubMed] [Google Scholar]
- 8.Wray J.D., Gajanayake I., Smith S.H. Congestive heart failure associated with a large transverse left ventricular moderator band in a cat, J Feline Med Surg 9, 2007, 56–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Jaggy A., Oliver J.E. Chlorpyrifos toxicosis in two cats, J Vet Intern Med 4, 1990, 135–139. [DOI] [PubMed] [Google Scholar]
- 10.Rumbeiha W.K., Francis J.A., Fitzgerald S.D., et al. A comprehensive study of Easter lily poisoning in cats, J Vet Diagn Invest 16, 2004, 527–541. [DOI] [PubMed] [Google Scholar]
- 11.Moore L.E., Biller D.S., Smith T.A. Use of abdominal ultrasonography in the diagnosis of primary hyperaldosteronism in a cat, J Am Vet Med Assoc 217, 2000, 213–215, 197. [DOI] [PubMed] [Google Scholar]
- 12.Gaschen F., Gaschen L., Seiler G., et al. Lethal peracute rhabdomyolysis associated with stress and general anesthesia in three dystrophin-deficient cats, Vet Pathol 35, 1998, 117–123. [DOI] [PubMed] [Google Scholar]
- 13.Dow S.W., LeCouteur R.A., Fettman M.J., Spurgeon T.L. Potassium depletion in cats: hypokalemic polymyopathy, J Am Vet Med Assoc 191, 1987, 1563–1568. [PubMed] [Google Scholar]
- 14.Leon A., Bain S.A., Levick W.R. Hypokalaemic episodic polymyopathy in cats fed a vegetarian diet, Aust Vet J 69, 1992, 249–254. [DOI] [PubMed] [Google Scholar]
- 15.Bellah J.R., Robertson S.A., Buergelt C.D., McGavin A.D. Suspected malignant hyperthermia after halothane anesthesia in a cat, Vet Surg 18, 1989, 483–488. [DOI] [PubMed] [Google Scholar]
- 16.Cooper B.J., De Lahunta A., Gallagher E.A., Valentine B.A. Nemaline myopathy of cats, Muscle Nerve 9, 1986, 618–625. [DOI] [PubMed] [Google Scholar]
- 17.Waldron D., Pettigrew V., Turk M., Turk J., Gibson R. Progressive ossifying myositis in a cat, J Am Vet Med Assoc 187, 1985, 64–65. [PubMed] [Google Scholar]
- 18.Baneth G., Aroch I., Tal N., Harrus S. Hepatozoon species infection in domestic cats: a retrospective study, Vet Parasitol 79, 1998, 123–133. [DOI] [PubMed] [Google Scholar]
- 19.Neumann S. Serum creatine kinase activity in dogs and cats with metabolic diseases, Dtsch Tierarztl Wochenschr 112, 2005, 343–347. [PubMed] [Google Scholar]
- 20.DeLahunta A., Glass E.N. Lower motor neuron: spinal nerve, general somatic efferent system. DeLahunta A., Glass E.N. Veterinary neuroanatomy and clinical neurology, 3rd edn, 2009, Elsevier-Saunders: Saint Louis, 95–96. [Google Scholar]
- 21.Ginman A.A., Kline K.L., Shelton G.D. Severe polymyositis and neuritis in a cat, J Am Vet Med Assoc 235, 2009, 172–175. [DOI] [PubMed] [Google Scholar]
- 22.Fascetti A.J., Mauldin G.E., Mauldin G.N. Correlation between serum creatine kinase activities and anorexia in cats, J Vet Intern Med 11, 1997, 9–13. [DOI] [PubMed] [Google Scholar]
- 23.Swets J.A. Measuring the accuracy of diagnostic systems, Science 240, 1988, 1285–1293. [DOI] [PubMed] [Google Scholar]
- 24.Melli G., Chaudhry V., Cornblath D.R. Rhabdomyolysis: an evaluation of 475 hospitalized patients, Medicine (Baltimore) 84, 2005, 377–385. [DOI] [PubMed] [Google Scholar]
- 25.Poels P.J., Gabreels F.J. Rhabdomyolysis: a review of the literature, Clin Neurol Neurosurg 95, 1993, 175–192. [DOI] [PubMed] [Google Scholar]
- 26.Mazaki-Tovi M., Aroch I. Rhabdomyolysis associated with Escherichia coli gastroenteritis in a cat suffering from chronic renal failure, Vet Rec 147, 2000, 137–138. [DOI] [PubMed] [Google Scholar]
- 27.Betrosian A., Thireos E., Kofinas G., Balla M., Papanikolaou M., Georgiadis G. Bacterial sepsis-induced rhabdomyolysis, Intensive Care Med 25, 1999, 469–474. [DOI] [PubMed] [Google Scholar]
- 28.Fisk D.T., Bradley S.F. Rhabdomyolysis induced by Salmonella enterica serovar Typhi bacteraemia, Clin Microbiol Infect 10, 2004, 595–597. [DOI] [PubMed] [Google Scholar]
- 29.Henrich W.L., Prophet D., Knochel J.P. Rhabdomyolysis associated with Escherichia coli septicemia, South Med J 73, 1980, 936–937. [DOI] [PubMed] [Google Scholar]
- 30.Yilmaz S., Demircioglu F., Oren H., Güneş B., Irken G. Rhabdomyolysis due to Escherichia coli sepsis in three pediatric patients with acute lymphoblastic leukemia, Pediatr Hematol Oncol 26, 2009, 57–62. [DOI] [PubMed] [Google Scholar]
- 31.Goto A., Takahashi Y., Kishimoto M., et al. Primary aldosteronism associated with severe rhabdomyolysis due to profound hypokalemia, Intern Med 48, 2009, 219–223. [DOI] [PubMed] [Google Scholar]
- 32.Shelton G.D. Rhabdomyolysis, myoglobinuria, and necrotizing myopathies, Vet Clin North Am Small Anim Pract 34, 2004, 1469–1482. [DOI] [PubMed] [Google Scholar]
- 33.Warren J.D., Blumbergs P.C., Thompson P.D. Rhabdomyolysis: a review, Muscle Nerve 25, 2002, 332–347. [DOI] [PubMed] [Google Scholar]
- 34.Aktas M., Auguste D., Concordet D., et al. Creatine kinase in dog plasma: preanalytical factors of variation, reference values and diagnostic significance, Res Vet Sci 56, 1994, 30–36. [DOI] [PubMed] [Google Scholar]
- 35.Aktas M., Lefebvre H.P., Toutain P.L., Braun J.P. Disposition of creatine kinase activity in dog plasma following intravenous and intramuscular injection of skeletal muscle homogenates, J Vet Pharmacol Ther 18, 1995, 1–6. [DOI] [PubMed] [Google Scholar]
- 36.Houpert P., Serthelon J.P., Lefebvre H.P., Toutain P.L., Braun J.P. In vivo non-invasive quantification of muscle damage following a single intramuscular injection of phenylbutazone in sheep, Vet Hum Toxicol 37, 1995, 105–110. [PubMed] [Google Scholar]
- 37.Lefebvre H.P., Laroute V., Braun J.P., Lassourd V., Toutain P.L. Non-invasive and quantitative evaluation of post-injection muscle damage by pharmacokinetic analysis of creatine kinase release, Vet Res 27, 1996, 343–361. [PubMed] [Google Scholar]
- 38.Toutain P.L., Lassourd V., Costes G., et al. A non-invasive and quantitative method for the study of tissue injury caused by intramuscular injection of drugs in horses, J Vet Pharmacol Ther 18, 1995, 226–235. [DOI] [PubMed] [Google Scholar]