Karsten E. Schober, Dr habil, PD; Corinna Cornand; Babett Kirbach; Heike Aupperle, Dr med vet; Gerhard Oechtering, Prof Dr med vet

Objective—To determine whether serum concentrations of cardiac troponin I (cTnI) and cardiac troponin T (cTnT) are increased in dogs with gastric dilatationvolvulus (GDV) and whether concentrations correlate with severity of ECG abnormalities or outcome. Design—Prospective case series. Animals—85 dogs with GDV. Procedure—Serum cTnI and cTnT concentrations were measured 12 to 24, 48, 72, and 96 hours after surgery. Dogs were grouped on the basis of severity of ECG abnormalities and outcome. Results—cTnI and cTnT were detected in serum from 74 (87%) and 43 (51%) dogs, respectively. Concentrations were significantly different among groups when dogs were grouped on the basis of severity of ECG abnormalities (none or mild vs moderate vs severe). Dogs that died (n = 16) had significantly higher serum cTnI (24.9 ng/ml) and cTnT (0.18 ng/ml) concentrations than did dogs that survived (2.05 and < 0.01 ng/ml, respectively). Myocardial cell injury was confirmed at necropsy in 4 dogs with high serum cardiac troponin concentrations. Conclusions and Clinical Relevance—Results indicate that concentrations of cTnI and cTnT suggestive of myocardial cell injury can commonly be found in serum from dogs with GDV and that serum cardiac troponin concentrations are associated with severity of ECG abnormalities and outcome. (J Am Vet Med Assoc 2002;221:381–388)

G

astric dilatation-volvulus (GDV) is common in large-breed dogs and proves fatal in approximately 20% of affected dogs.1-3,a,b Causes of death and factors reportedly associated with death in dogs with GDV include hypoperfusion and ischemia, reperfusion injury, microinfarction, autonomic dysbalance, acid-base and electrolyte disturbances, myocardial depressant factor, sepsis, endotoxemia, and systemic inflammatory response syndrome with accumulation of pro-inflammatory cytokines leading to circulatory collapse and death.1,4-13 From the Department of Small Animal Medicine (Schober, Cornand, Kirbach, Oechtering) and the Institute of Veterinary Pathology (Aupperle), Faculty of Veterinary Medicine, University of Leipzig, An den Tierkliniken 23, 04103 Leipzig, Germany. Equipment was provided by Dade-Behring Diagnostics, Liederbach, Germany. Presented in part at the 19th Annual Veterinary Medical Forum, Denver, Colo, May, 2001. The authors thank W. Goldmann from Dade-Behring Diagnostics for technical assistance and Dr. S. Kropf from the Coordination Center for Clinical Trials of the University of Leipzig for assistance. Address correspondence to Dr. Schober. JAVMA, Vol 221, No. 3, August 1, 2002

Approximately 40 to 70% of affected dogs develop cardiac arrhythmias that may contribute to death.1,3,6,7,13 Although the etiology of ventricular arrhythmia in dogs with GDV is unknown,14 myocardial necrosis secondary to ischemic, neurohumoral, or toxic cardiac damage has been documented in such dogs and has been assumed to be a major determinant of arrhythmogenesis and myocardial dysfunction.8,10,12 However, a diagnosis of acute myocardial degeneration or necrosis has historically been restricted to dogs undergoing a postmortem examination. Recently, analysis of circulating concentrations of cardiac troponin has improved predictions of the risk of serious cardiac complications in human patients with myocardial ischemia,15,16 myocardial inflammation and necrosis,17 congestive heart failure,18 or septic myocardial injury.19 In addition, serum cardiac troponin concentration has been used to assess myocardial cell injury in dogs,20 cats,21 and rabbits22 with myocardial contusion, dogs undergoing chemotherapy with doxorubicinc or with experimentally induced myocardial infarction,23,24 and mice with autoimmune myocarditis.25 Cardiac troponins are thin filamentassociated proteins of the heart muscle that regulate the calcium-mediated interaction between actin and myosin. They are structurally highly preserved between humans and dogs.26,27 Cardiac troponin I (cTnI) and cardiac troponin T (cTnT) are uniquely expressed in the myocardium and have been widely recognized as highly sensitive and specific serum markers for the noninvasive diagnosis of increased myocyte permeability or necrosis.16,18,28,29 To our knowledge, studies on the importance of circulating concentrations of cardiac troponin in dogs with GDV have not been reported. We hypothesize that measurement of serum cardiac troponin concentration may be useful in predicting severity of cardiac injury in dogs with GDV and that high serum troponin concentrations are associated with a poorer outcome. The purposes of the study reported here, therefore, were to determine the percentage of dogs with GDV that have measurable serum cTnI and cTnT concentrations, determine whether serum cTnI and cTnT concentrations were associated with ECG or histologic abnormalities, evaluate the variability in measurement of cTnI concentration, determine the pattern of release of cardiac troponin after surgical correction of GDV, and evaluate whether serum cardiac troponin concentration can be used to predict outcome of dogs with GDV. Materials and Methods Dogs—Eighty-five dogs examined at the Veterinary Teaching Hospital of the University of Leipzig between October 1997 and May 2001 because of GDV were prospectively enrolled in the study. Dogs were included in the study Scientific Reports: Original Study

381

SMALL ANIMALS

Serum cardiac troponin I and cardiac troponin T concentrations in dogs with gastric dilatation-volvulus

SMALL ANIMALS

if the diagnosis of GDV was confirmed at the time of surgery and complete clinical, biochemical, and ECG examinations were performed after surgery for correction of GDV. The time interval between the last food intake and surgery was recorded. In all dogs, a physical examination was performed and an electrocardiogram was recorded at the time of admission. Biochemical analyses were performed and follow-up electrocardiograms were recorded between 12 and 24 hours and 48, 72, and 96 hours after surgery. Routine medical treatment was provided to each dog, according to individual needs. A necropsy was performed on dogs that died, if owner consent was given. An experienced pathologist without knowledge of the results of diagnostic tests that had been performed or the treatments that had been given performed histologic examinations of necropsy specimens from all 4 cardiac chambers. Electrocardiography—Six-lead ECG recordingsd of 2 to 4 minutes duration were obtained from each dog at the time of admission and at the predetermined intervals after surgery. Electrocardiograms were evaluated for heart rate and rhythm, signs of conduction disturbances, and alterations in amplitude or duration of the P, QRS, ST, and T segments.30 Dogs were grouped according to severity of ECG abnormalities by a single investigator (KS) blinded to results of biochemical analyses. For group assignment, the most relevant ECG abnormalities in each dog were considered, regardless of when they occurred during the course of disease. Group 1 consisted of dogs with normal ECG findings and dogs with mild ECG abnormalities such as sinus bradycardia, intermittent sinus arrest, sinus tachycardia (heart rate < 180 beats/min), high voltage or splintered R-wave, first- or second-degree atrioventricular block, prolongation of the QT interval, increased Q-wave amplitude, P-pulmonale, and P-mitrale. Group 2 consisted of dogs with moderate ECG abnormalities such as persistent sinus tachycardia (heart rate ≥ 180 beats/min), increased T-wave amplitude, ST elevation (> 0.15 mV) or depression (> 0.20 mV), single monomorphic ventricular or supraventricular premature complexes (VPC or SVPC), fusion beats, slow idioventricular rhythm, single couplets, and transient left or right bundle branch block. Group 3 consisted of dogs with severe ECG abnormalities such as frequent VPC or SVPC (> 15/min), couplets, triplets, polymorphic VPC, paroxysmal atrial fibrillation, ventricular tachycardia, and ventricular flutter or fibrillation. Biochemical analyses—Complete blood counts and biochemical analyses (plasma urea nitrogen, creatinine, total protein, potassium, sodium, magnesium, total calcium, chloride, lactate, and bicarbonate concentrations; alanine aminotransferase [ALT] activity; pH; base excess; PO2; and PCO2) were performed immediately after venous blood collection. Serum was separated from a second venous blood sample, and an aliquot (2 ml) was shipped, at room temperature, to a commercial laboratory for measurement of cTnT concentration (maximum interval between sample collection and analysis, 48 hours). A second aliquot (1 ml) of serum was frozen within 1 hour after blood collection and stored at –20 C until analysis of cTnI concentration (maximum interval between sample collection and analysis, 14 days). Serum cTnI concentration was measured with a commercially available fluorogenic sandwich ELISAe that incorporated 2 polyclonal goat antibodies against human cardiac TnI. The 2 antibodies had independent immunologic epitopes, but epitope sites for both antibodies were located in the stable central region of the molecule (ie, the region that remains after N and C terminal proteolysis in circulation).f The lower limit of detection of cTnI in serum was 0.5 ng/ml. Standard positive controlg,h and calibrationi solutions were used. To determine repeatability of result of the cTnI analysis, 15 randomly selected serum samples were analyzed 5 times at 2-hour intervals (1-day repeatability). To assess the effect of storage at room temperature (22 to 24 C [72 to 75 F]) on 382

Scientific Reports: Original Study

results of the cTnI analysis, 11 randomly selected serum samples that had been stored at room temperature were analyzed 5 times at 24-hour intervals (5-day repeatability). Serum cTnT concentration was measured with an electrochemiluminescence immunoassayj that incorporated 2 mouse monoclonal antibodies against different epitopes of human cardiac troponin T. According to the assay’s manufacturer, the lower limit of detection of cTnT in serum was 0.01 ng/ml. Standard positive controlk and calibrationl solutions were used. Reference values for serum cTnI and cTnT concentration were obtained from a previous study20 of 40 healthy, weight- and sexmatched dogs. For serum cTnI concentration, the reference range was 0 to 0.934 ng/ml; for serum cTnT concentration, the reference range was 0 to 0.010 ng/ml. Routine procedures were used to measure plasma lactate,m potassium,m total calcium,n and magnesiumn concentrations, plasma ALT activity,n and acid-base status.o Reference values generated by the laboratory were used. Statistical analyses—All statistical analyses were performed with standard software.31,p,q Descriptive statistics were calculated for age, body weight, heart rate, sex, and selected biochemical variables. Data were reported as median and range, except when stated otherwise. The Kolmogorov-Smirnov test was used to test whether variables were normally distributed. The Kruskal-Wallis ANOVA on ranks test was used to test for differences among dogs grouped on the basis of severity of ECG abnormalities. The Dunn test was used for pairwise comparisons when significant differences were observed.32 A MannWhitney rank sum test was used to compare values for dogs that survived with values for dogs that died. For these analyses, the maximum value of each variable, irrespective of the time of its determination, was used. Alternatively, serum cTnI and cTnT concentrations recorded at each recording time (12 to 24, 48, 72, and 96 hours after surgery) were used. Multiple logistic regression analysis was used to predict outcome; independent variables such as cTnI, cTnT, lactate, potassium, total calcium, and magnesium concentrations; ALT activity; and pH were incorporated in a stepwise forward procedure. Receiver operating characteristic curve analysis was used to compare the predictive abilities of cTnI and cTnT concentrations and to establish cutoff values at each recording time. A Friedman repeatedmeasures ANOVA on ranks was used to determine the release pattern of cTnI and cTnT during the first 96 hours after surgery, using data from dogs that survived for which complete data sets were available. When significant differences in cardiac troponin concentrations between single time points were found, the nonparametric Student-Newman-Keuls test was used for pairwise comparisons. Repeatability of results of the cTnI assay was evaluated by calculating the coefficient of variation (CV). For all analyses, values of P ≤ 0.05 were considered significant.32

Results Dogs—Eighty-five dogs (43 males and 42 females) representing 20 breeds were included in the study. German Shepherd Dogs (n = 22), Great Danes (14), Doberman Pinschers (11), and Bernese Mountain Dogs (9) were the most common. Mean age was 6.4 years (range, 0.8 to 14.0 years) and mean body weight was 44 kg (97 lb; range, 13 to 85 kg [29 to 187 lb]). One dog had concurrent GDV and dilated cardiomyopathy. Repeatability of results of the cTnI assay—The CV for 1- and 5-day repeatability of results of the cTnI assay were 9.1 and 22.5%, respectively. Of the 15 samples used for evaluation of 1-day repeatability, 2 had cTnI concentrations below the lower limit of detection (< 0.5 ng/ml) all 5 times they were analyzed, 10 had slightly or moderately high cTnI concentrations (≤ 40 ng/ml) each JAVMA, Vol 221, No. 3, August 1, 2002

Serum cTnI and cTnT concentrations—Cardiac troponin I was detected in serum from 74 of the 85 (87%) dogs 1 or more times after surgery; serum cTnI concentrations in these dogs ranged from 0.5 to 381.0 ng/ml. Cardiac troponin T was detected in serum from 43 (51%) dogs 1 or more times after surgery; serum cTnT concentrations in these dogs ranged from 0.01 to 8.92 ng/ml. All 43 dogs with detectable concentrations of cTnT had detectable concentrations of cTnI, whereas 31 dogs with detectable concentrations of cTnI did not have detectable concentrations of cTnT. Eleven of the 85 (13%) dogs did not have detectable concentrations of cTnI or cTnT in serum. Of the 31 dogs in which cTnI, but not cTnT, was detected in serum, 13 had serum cTnI concentrations > 2.0 ng/ml. The highest serum cTnI concentrations in these dogs were 24.80, 17.10, and 11.10 ng/ml. Other biochemical analyses—Plasma lactate concentration (1.2 to 18.8 mmol/L; reference range, 1.0 to 3.0 mmol/L) was high in 52 of the 85 (61%) dogs 1 or more times after surgery. Plasma potassium concentration (2.8 to 5.2 mmol/L; reference range, 3.5 to 5.2 mmol/L) was low in 21 (25%) dogs and was high in none. Plasma total calcium concentration (1.98 to 3.34 mmol/L; reference range, 2.30 to 3.00 mmol/L) was low in 20 (24%) dogs and high in 2 (2%). Plasma magnesium concentration (0.46 to 1.45 mmol/L; reference range, 0.80 to 1.30 mmol/L) was low in 58 (68%) dogs and high in 3 (3%). Plasma ALT activity (41 to 17,602 U/L; reference range, 15 to 50 U/L) was high in 82 (97%) dogs. Venous pH (7.21 to 7.45; reference range, 7.32 to 7.38) was low in 31 (36%) dogs and high in 11 (13%). Plasma urea nitrogen concentration (3.8 to 36.4 mmol/l; reference range 3.5 to 8.3 mmol/l) was high in 13 (15%) dogs. Electrocardiographic abnormalities—Electrocardiographic abnormalities were identified in 68 (80%) dogs. Ventricular arrhythmias were identified in 36 (42%) dogs, ST segment depression in 18 (21%), conduction disturbances in 15 (18%), increased Twave amplitude in 9 (11%), increased Q-wave amplitude in 8 (9%), SVPC in 6 (7%), ST segment elevation in 6 (7%), and atrial fibrillation in 3 (4%; 1 of the 3 had concurrent dilated cardiomyopathy). Thirty-seven dogs were classified as having no or only mild ECG abnormalities, 23 were classified as having moderate ECG abnormalities, and 25 were classified as having severe ECG abnormalities (Table 1). Cardiac troponin I was detected in serum from 27 of the 37 (73%) dogs with no or only mild ECG abnormalities, 22 of the 23 (96%) dogs with moderate ECG abnormalities, and all 25 (100%) of the dogs with severe ECG abnormalities. Cardiac troponin T was detected in serum from 6 of the 37 (16%) dogs with no or only mild ECG abnormalities, 10 of the 23 (43%) dogs with moderate ECG abnormalities, and 11 of the 25 (44%) dogs with severe ECG abnormalities. Plasma lactate, potassium, and magnesium concentrations and JAVMA, Vol 221, No. 3, August 1, 2002

venous pH did not differ significantly among groups when dogs were grouped on the basis of severity of ECG abnormalities. For both serum cTnI concentration and serum cTnT concentration, values for dogs with no or only mild ECG abnormalities were significantly lower than values for dogs with moderate ECG abnormalities and values for dogs with severe ECG abnormalities, and values for dogs with moderate ECG abnormalities were significantly lower than values for dogs with severe ECG abnormalities (Fig 1 and 2). Table 1—Selected physical and biochemical variables in 85 dogs with gastric dilatation-volvulus (GDV), grouped on the basis of severity of ECG abnormalities Severity of ECG abnormalities Variable

Mild (n = 37)

Moderate (n = 23)

Severe (n = 25)

Heart rate* (beats/min) 140 (70–200) 160 (94–260) 178 (62–250)c Time (h)† 4 (1–20) 6 (2–24)a 6 (1–24)c cTnI (ng/ml) 0.53 (⬍ 0.5–104) 3.29 (⬍ 0.5–63.4)a 35.0 (1.2–81)b,c cTnT (ng/ml) ⬍ 0.01 (⬍ 0.01–0.98) 0.02 (⬍ 0.01–0.77)a 0.24 (⬍ 0.01–8.92)b,c Total calcium (mmol/L) 2.46 (2.14–2.77) 2.40 (1.98–2.85) 2.27 (2.01–3.34)c ALT (U/L) 131 (41–15,512) 387 (80–9,844)a 289 (43–17,602) Data are given as median (range). *Heart rate at the time of admission. †Time between surgery and last food intake prior to surgery. cTnI = Serum cardiac troponin I concentration. cTnT = Serum cardiac troponin T concentration. Total calcium = Plasma total calcium concentration. ALT = Plasma alanine aminotransferase activity. a Significant (P ⱕ 0.05) difference between dogs with mild and moderate ECG abnormalities. bSignificant (P ⱕ 0.05) difference between dogs with moderate and severe ECG abnormalities. cSignificant (P ⱕ 0.05) difference between dogs with mild and severe ECG abnormalities.

Figure 1—Box-and-whisker plots of maximum serum cardiac troponin I concentration in 85 dogs with gastric dilatation-volvulus (GDV), grouped on the basis of severity of ECG abnormalities. The horizontal line in each box represents the median value. The boxes themselves represent the interquartile range (25th to 75th percentile). Whiskers represent the 5th and 95th percentiles. Filled circles represent outliers. aSignificant (P ≤ 0.05) difference between dogs with mild and moderate ECG abnormalities. b Significant (P ≤ 0.05) difference between dogs with moderate and severe ECG abnormalities. cSignificant (P ≤ 0.05) difference between dogs with mild and severe ECG abnormalities. Scientific Reports: Original Study

383

SMALL ANIMALS

time they were analyzed, and 3 had markedly high cTnI concentrations (> 40 ng/ml) each time they were analyzed. There was no evidence of an increase or decrease in measured cTnI concentration over time. Similar results were found during evaluation of 5-day repeatability.

SMALL ANIMALS

Time course of cardiac troponin release—In dogs that survived at least 4 days, peak serum concentrations of cTnI were observed between 48 and 72 hours after surgery for correction of GDV (Fig 3). Serum concentrations of cTnT, on the other hand, did not differ significantly with time after surgery.

Figure 2—Box-and-whisker plots of maximum serum cardiac troponin T concentration in 85 dogs with GDV, grouped on the basis of severity of ECG abnormalities. See Figure 1 for key.

Figure 3—Box-and-whisker plots of serum cardiac troponin I concentration in 51 dogs 12 to 24, 48, 72, and 96 hours after surgery for correction of GVD. aSignificant (P ≤ 0.05) difference between values 12 to 24 hours and 48 hours after surgery. bSignificant (P ≤ 0.05) difference between values 12 to 24 hours and 72 hours after surgery. cSignificant (P ≤ 0.05) difference between values 48 hours and 72 hours after surgery. dSignificant (P ≤ 0.05) difference between values 48 hours and 96 hours after surgery. e Significant (P ≤ 0.05) difference between values 72 hours and 96 hours after surgery. See Figure 1 for remainder of key. 384

Scientific Reports: Original Study

Outcome—Sixteen (19%) dogs died, including 3 of the 37 (8%) dogs with no or only mild ECG abnormalities, 5 of the 23 (22%) dogs with moderate ECG abnormalities, and 8 of the 25 (32%) dogs with severe ECG abnormalities. Potential causes of death were identified in 7 dogs and included severe gastric necrosis, gastric rupture, peritonitis, and sepsis with DIC. Ten dogs that died had ventricular arrhythmias. All 16 dogs that died had detectable concentrations of cTnI in serum, and 14 of the 16 had detectable concentrations of cTnT. Plasma potassium, magnesium, and total calcium concentrations were not significantly different between dogs that died and dogs that survived (for these analyses, the maximum value of each variable, irrespective of the time of its determination, was used). Mild or moderate azotemia was identified in 7 dogs that survived and 6 dogs that died. Serum cTnI and cTnT concentrations were significantly higher in dogs that died than in dogs that survived at each measurement time (Fig 4 and 5) and when maximum measured value in each dog was analyzed (Table 2). Multiple logistic regression indicated that plasma lactate concentration and plasma ALT activity 12 to 24 hours after surgery were independent predictors of outcome with an accuracy of 88%. Serum cTnI concentration and plasma ALT activity 48 hours after surgery were independent predictors of outcome with an accuracy of 91%. Plasma ALT activity and serum cTnT concentration 72 or 96 hours after surgery were independent predictors of outcome with accuracy of 93 or 95%, respectively. Receiver operating characteristic curves for serum cTnI and cTnT concentrations were similar (Fig 6). The area under the curve was 0.80 (95% confidence interval, 0.73 to 0.88) for cTnI concentration and 0.76 (95% confidence interval, 0.68 to 0.85) for cTnT concentration. When data for all measurements of cardiac troponin concentrations were pooled, the best cutoff point for cTnI concentration for predicting outcome was determined to be 4.04 ng/ml (sensitivity, 76%;

Figure 4—Box-and-whisker plots of serum cardiac troponin I concentrations in 85 dogs 12 to 24, 48, 72, and 96 hours after surgery for correction of GVD. At each time point of analysis, the left column represents dogs that survived and the right column represents dogs that died. *Values for the 2 groups were significantly (P < 0.05) different. See Figure 1 for remainder of key. JAVMA, Vol 221, No. 3, August 1, 2002

SMALL ANIMALS

Figure 5—Box-and-whisker plots of serum cardiac troponin T concentrations in 51 dogs 12 to 24, 48, 72, and 96 hours after surgery for correction of GVD. See Figures 1 and 4 for key. Table 2—Selected physical and biochemical variables in 85 dogs with GDV, grouped on the basis of outcome Figure 6—Receiver operating characteristic curves for use of serum cardiac troponin I (thin line) and serum cardiac troponin T (thick line) concentrations in predicting outcome of 85 dogs with GDV.

Outcome Variable Heart rate (beats/min) Time (h) cTnI (ng/ml) cTnT (ng/ml) Lactate (mmol/L) ALT (U/L) Venous pH

Survived (n = 69)

Died (n = 16)

P value

146 (62–260) 4 (1–24) 2.05 (⬍ 0.5–120.0) ⬍ 0.01 (⬍ 0.01–3.50) 3.10 (1.50–8.10) 150 (41–5,797) 7.32 (7.22–7.38)

188 (144–240) 12 (6–18) 24.9 (0.52–381.0) 0.18 (⬍ 0.01–8.92) 8.95 (3.20–18.80) 1,503 (43–17,602) 7.25 (7.21–7.35)

⬍ 0.001 ⬍ 0.001 ⬍ 0.001 ⬍ 0.001 ⬍ 0.001 ⬍ 0.001 0.003

myocardial edema, mild myocardial hemorrhage, and moderate angiosclerosis. This dog did not have any evidence of myocardial necrosis and had only mild ECG abnormalities prior to death. The fifth dog had severe diffuse acute myocardial necrosis, evident in both ventricles, and mild or moderate myocardial inflammation. Myocardial necrosis was more severe in the subendocardium and the papillary muscles, compared with the epicardium and subepicardium. In addition, mild degeneration of neurons in the sinuatrial ganglion and the atrioventricular ganglion was found.

Lactate = Plasma lactate concentration. For each dog, the highest measured value, regardless of the time of its determination, was used. See Table 1 for remainder of key.

specificity, 70%), and the best cutoff point for cTnT concentration for predicting outcome was determined to be 0.031 ng/ml (sensitivity, 66%; specificity, 74%). These cutoff points were then used to determine sensitivity and specificity of cTnI and cTnT at each measurement time after surgery (Table 3).

Discussion Results of the present study indicate that detectable concentrations of cTnI and cTnT can commonly be found in the serum of dogs with GDV, suggesting that such dogs have cardiomyocyte degeneration and necrosis. Myocardial necrosis has previously been documented in dogs with experimentally induced or naturally occurring GDV,8,10,b dogs with coronary artery disease and myocardial infarction,33-35 and dogs that died suddenly after aggressive behavior.36 However, in all of these previous studies, myocardial necrosis was identified at necropsy, and noninvasive diagnosis of myocyte injury was lacking. We conclude, on the basis of our findings, that analysis of serum car-

Necropsy findings—A necropsy was performed on 5 of the 16 dogs that died. One of these dogs had interstitial myocardial edema, myocardial fibrosis, and lipomatosis. Another had evidence of mild intramyocardial bleeding, areas of acute degeneration and necrosis of cardiomyocytes and Purkinje fibers, and mild myocardial fibrosis. A third dog had mild interstitial bleeding and mild or moderate myocardial degeneration or necrosis. A fourth dog had mild interstitial

Table 3—Sensitivity and specificity of using serum cardiac troponin I (cTnI) and cardiac troponin T (cTnT) concentrations for predicting a fatal outcome in 85 dogs with GDV cTnI Time after surgery (h) 12 to 24 48 72 96

cTnT

Sensitivity (%)

Specificity (%)

75 100 100 75

82 79 73 87

Sensitivity (%) 67 100 100 100

Specificity (%) 78 72 75 72

The cutoff for cTnI concentration was 4.05 ng/ml; the cutoff for cTnT concentration was 0.03 ng/ml.

JAVMA, Vol 221, No. 3, August 1, 2002

Scientific Reports: Original Study

385

SMALL ANIMALS

diac troponin concentrations is useful for detecting myocardial cell injury in dogs with GDV and that serum cardiac troponin concentrations may provide important prognostic information. Serum cTnI concentration appeared to be more sensitive than serum cTnT concentration for detection of myocardial cell injury; however, their usefulness in predicting outcome was comparable. In the present study, serum cTnI and cTnT concentrations, particularly among dogs with severe ECG abnormalities and dogs that died, were similar to those reported for dogs with experimentally induced myocardial infarction.23,24 Cummins and Cummins23 found that serum cTnI concentration ranged from 44 to 217 ng/ml in dogs after complete ligation of the left anterior descending coronary artery, leading to myocardial infarct sizes of 2.3 to 27.5% of the total heart mass. In the present study, 12 dogs had serum cTnI concentrations > 44 ng/ml, and 7 of the 12 died. In 1 of these dogs, myocardial damage comparable to that associated with severe myocardial infarction was confirmed at necropsy. In another study37 of 6 dogs with experimentally induced ischemic heart disease, infarct size was correlated with serum cTnT concentration; maximum cTnT concentrations ranged from 5 to 10 ng/ml. It was also found that increases in serum troponin concentrations after infarction parallel decreases in cardiac tissue concentrations of troponin in dogs.24 In the present study, 4 dogs with high serum cTnI and cTnT concentrations had histologic evidence of myocardial necrosis, whereas 1 dog with concentrations close to the detection limits of the assay had unremarkable histologic findings. Minor increases in serum cardiac troponin concentrations may occur without histologic evidence of myocardial cell injury secondary to increased myocyte membrane permeability with release of cytosolically dissolved cardiac troponin. The free intramyocardial troponin pool accounts for about 2 to 8% of total cardiomyocyte troponin content in dogs.37 Structurally bound cardiac troponin is released only after major injury with cell disruption and necrosis.24,28,37 Comparative histologic and serum biochemical studies in rats with isoprenalineinduced cardiac muscle damage were used to determine cutoff values of serum cardiac troponin concentrations associated with histologically detectable myocardial abnormalities.38 It was found that both cTnI and cTnT are highly sensitive markers for myocardial damage. In addition, minimum serum concentrations of 1.3 ng/ml for cTnI and 0.35 ng/ml for cTnT were necessary for myocardial cell injury to be detected histologically.38 The time course of cTnI release among dogs with GDV in the present study was similar to that reported for dogs23,37 and humans15,16,39 with myocardial infarction, with serum cTnI concentration peaking 48 to 72 hours after surgery for GDV. In some, but not all, dogs in the present study, a continuous increase in serum cardiac troponin concentrations was observed. Most of these animals died, suggesting that continuously increasing serum cardiac troponin concentrations may have prognostic value. In dogs with experimentally induced myocardial infarction,23 circulating cTnI and cTnT concentrations peaked between the first and 386

Scientific Reports: Original Study

third day after arterial occlusion. Circulating cTnI concentration decreased thereafter, reaching baseline values by about 200 hours after arterial occlusion.23 On the other hand, a second peak in cTnT concentration was observed between the fourth and sixth day after arterial occlusion and was associated with continuous troponin release secondary to ongoing myocardial cell injury associated with reperfusion. This suggests, therefore, that serum cTnI and cTnT concentrations should be measured daily for the first 3 days after surgery in dogs with GDV to detect patients at a high risk for cardiovascular compromise. In the present study, serum cardiac troponin concentrations were associated with severity of ECG abnormalities, with dogs with more severe ECG abnormalities having higher serum cardiac troponin concentrations. Although malignant ventricular arrhythmias may be evident at the time of admission in dogs with GDV, they most often are first detected approximately 24 to 72 hours after surgery,2,3,7,8 which parallels the time course of cTnI release in the present study. Thus, ventricular arrhythmias seem to be associated with acute myocyte degeneration and necrosis in dogs with GDV. Similar observations were made by Harris40 in 1950, who reported that in dogs with experimental coronary artery occlusion, ectopic ventricular activity developed after a latency of 4.5 to 8 hours and persisted for 48 to 96 hours. Ventricular arrhythmias were closely related to duration of ischemia and histologic signs of myocardial necrosis; however, additional excitatory factors, such as histamines and catecholamines, were not studied. Electrolyte disturbances, particularly hypokalemia and hypomagnesemia, metabolic acidosis, cytokines, tissue hypoperfusion with gastric and hepatic necrosis and hyperlactatemia, reperfusion, autonomic dysbalance, and endotoxin accumulation may have contributed to arrhythmogenesis and outcome in the present study,3,8,19,41 and logistic regression analysis indicated that factors other than severity of myocardial injury were associated with outcome. Plasma lactate concentration, a marker of tissue hypoperfusion,12,41 and plasma ALT activity, an indicator of acute hepatocyte ischemia and necrosis, seemed to be particularly important. Thus, despite the association found between serum cardiac troponin concentrations, severity of ECG abnormalities, and outcome in the present study, it is uncertain whether and to what degree ventricular arrhythmias contributed to outcome in these dogs. In a study2 of 295 dogs with GDV, no obvious association between occurrence of ventricular arrhythmias and outcome was found. However, dogs with acute GDV have a tendency to have ventricular fibrillation.a Regardless, results of the present study do suggest that severity of myocardial cell injury, development of malignant ventricular arrhythmias, and outcome are closely related in dogs with GDV and that repeated analysis of serum cardiac troponin concentrations may be useful in detecting dogs prone to develop ventricular electrical instability. Repeatability of assay results was determined only for the cTnI assay in the present study. Therefore, conclusions cannot be drawn regarding accuracy of the cTnT assay. In addition, results regarding accuracy of JAVMA, Vol 221, No. 3, August 1, 2002

a

Miller T, Nakayama T, Schwartz D, et al. Effects of acute gastric distension and recovery on tendency for ventricular arrhythmia (abstr), in Proceedings. 17th Annu Vet Med Forum 1999;743. b Van Sluijs FJ. Gastric dilatation-volvulus in the dog. Doctoral Thesis, Department of Small Animal Medicine and Surgery, University of Utrecht, Utrecht, The Netherlands, 1987. c DeFrancesco TC, Atkins CE, Keene BW, et al. Evaluation of cardiac troponin T as a potential predictor of doxorubicin cardiotoxicity in dogs (abstr), in Proceedings. 18th Annu Vet Med Forum 2000;712. d CARDIOVIT AT-6, Schiller, Baar, Switzerland. e OPUS Immunoassay System, OPUS Troponin I, Dade-Behring Diagnostics Inc, Westwood, Mass. f Jacobs-Gebauer K, Dade-Behring Diagnostics Inc, Liederbach, Germany: Personal communication, 2001. g OPUS Troponin I Control Sera I to III, Dade-Behring Diagnostics Inc, Westwood, Mass. h Liquecheck Cardiac Markers Control, BIORAD Laboratories, Diagnostics Group, Irvine, Calif. i Opus Troponin I Calibrators A to F, Dade-Behring Diagnostics Inc, Westwood, Mass. j Elecsys Troponin T STAT Immunoassay, Elecsys 2010 analyzer, Boehringer Mannheim, Mannheim, Germany. k PreciControl Cardiac 1 and 2 Elecsys, Boehringer Mannheim, Mannheim, Germany. l Troponin T STAT CalCheck Elecsys 1 to 3, Boehringer Mannheim, Mannheim, Germany. m Kodak Ektachem DT Systems, Johnson & Johnson Clinical Diagnostics Inc, Rochester, NY. n Hitachi 704, Nissei Sangyo GmbH, Ratingen, Germany. o ABL5 Blood Gas System, Radiometer Medical A/S, Kopenhagen, Denmark. p Sigma Stat, version 2.0 for Windows 95, Jandel Scientific, San Rafael, Calif. q SPSS for Windows, version 10.0, SPSS Inc, Chicago, Ill.

References 1. Muir WW, Bonagura JD. Treatment of cardiac arrhythmias in dogs with gastric distension-volvulus. J Am Vet Med Assoc 1984; 184:1166–1171. JAVMA, Vol 221, No. 3, August 1, 2002

2. Brockman DJ, Washabau RJ, Drobatz KJ. Canine gastric dilatation/volvulus syndrome in a veterinary critical care unit: 295 cases (1986–1992). J Am Vet Med Assoc 1995;207:460–464. 3. Bourman JD, Schertel ER, Allen DA, et al. Factors associated with perioperative mortality in dogs with surgically managed gastric dilatation-volvulus: 137 cases (1988–1993). J Am Vet Med Assoc 1996;208:1855–1858. 4. Lefer AM. Myocardial depressant factor and circulatory shock. Klin Wochenschr 1974;52:358–370. 5. Fisher WD, Heimbach DW, McArdle CS, et al. A circulating depressant effect following canine hemorrhagic shock. Br J Surg 1973;60:392–393. 6. Muir WW. Gastric dilatation-volvulus in the dog, with emphasis on cardiac arrhythmias. J Am Vet Med Assoc 1982;180: 739–742. 7. Muir WW, Lipowitz AJ. Cardiac dysrhythmias associated with gastric dilatation-volvulus in the dog. J Am Vet Med Assoc 1978;172:683–689. 8. Horne WA, Gilmore DR, Dietze AE, et al. Effects of gastric distension-volvulus on coronary blood flow and myocardial oxygen consumption in the dog. Am J Vet Res 1985;46:98–104. 9. Muir WW. Acid-base and electrolyte disturbances in dogs with gastric dilatation-volvulus. J Am Vet Med Assoc 1982;181: 229–231. 10. Muir WW, Weisbrode SE. Myocardial ischemia in dogs with gastric dilatation volvulus. J Am Vet Med Assoc 1982;181:363–366. 11. Orton CE, Muir WW. Isovolumetric indices and humeral cardioactive substance bioassay during clinical and experimentally induced gastric dilatation-volvulus in dogs. Am J Vet Res 1983;44: 1516–1520. 12. Muir WW. Shock. Compend Contin Educ Pract Vet 1998;20: 549–566. 13. Brockman DJ, Holt DE. Management protocol for acute gastric dilatation-volvulus syndrome in dogs. Compend Contin Educ Pract Vet 2000;22:1025–1034. 14. Benchuk TN, Hauptman JG, Braselton WE, et al. Intracellular magnesium concentrations in dogs with gastric dilatation-volvulus. Am J Vet Res 2000;61:1415–1417. 15. Olatidoye AG, Alan HB, Feng YJ, et al. Prognostic role of troponin T versus troponin I in unstable angina pectoris for cardiac events with meta-analysis comparing published studies. Am J Cardiol 1998;81:1405–1410. 16. Ottani F, Galvani M, Ferrini D, et al. Direct comparison of early elevations of cardiac troponin T and I in patients with clinical unstable angina. Am Heart J 1999;137:284–291. 17. Lauer B, Niederau C, Schannwell M, et al. Cardiac troponin T in patients with clinically suspected myocarditis. J Am Coll Cardiol 1997;30:1354–1359. 18. Missov E, Calzolari C, Pau B. Circulating cardiac troponin I in severe congestive heart failure. Circulation 1997;96:2953–2958. 19. Ver Elst KM, Spapen HD, Nguyen DN, et al. Cardiac troponins I and T are biological markers of left ventricular dysfunction in septic shock. Clin Chem 2000;46:650–657. 20. Schober KE, Kirbach B, Oechtering G. Noninvasive assessment of myocardial cell injury in dogs with suspected cardiac contusion. J Vet Cardiol 1999;1:17–25. 21. Kirbach B, Schober K, Oechtering G, et al. Diagnostik von Herzmuskelschäden bei Katzen mit stumpfen Thoraxtraumen über biochemische Parameter im Blut. Tierarztl Prax 2000;28:25–33. 22. Bertinchant JP, Robert E, Polge A, et al. Release kinetics of cardiac troponin I and cardiac troponin T in effluents from isolated perfused rabbit hearts after graded experimental myocardial contusion. J Trauma 1999;47:474–480. 23. Cummins B, Cummins P. Cardiac specific troponin-I release in canine experimental myocardial infarction: development of a sensitive enzyme-linked immunoassay. J Mol Cell Cardiol 1987;19: 999–1010. 24. Ricchiuti V, Sharkey SW, Murakami MM, et al. Cardiac troponin I and T alterations in dog hearts with myocardial infarction: correlation with infarct size. Am J Clin Pathol 1998;110:241–247. 25. Smith SC, Ladenson JH, Mason JW, et al. Elevations of cardiac troponin I associated with myocarditis. Experimental and clinical correlates. Circulation 1997;95:163–168. Scientific Reports: Original Study

387

SMALL ANIMALS

the cTnI assay are applicable only to the assay used in this study. In contrast to assays for measuring cTnT concentration, assays for measuring cTnI concentration are not yet standardized. Differences in the immunologic epitopes detected by the specific antibodies used in different assays, variations in heterophilia and cross-reactivity of the antibodies used, alterations in antibody recognition secondary to degradation of the epitopes, and differences in assay sensitivities, substrates, and thresholds may result in conflicting results when studies using different assays for cTnI concentration are compared.42 In addition, azotemia or hyperbilirubinemia may affect assay results.43 An important limitation of the present study is that continuous ECG recordings were not analyzed. Evaluation of 2- to 4-minute ECG recordings may have resulted in important arrhythmias being missing, leading to an underestimation of ECG severity in some dogs.b Additionally, a necropsy was performed on only a limited number of dogs, limiting our ability to verify results of biochemical analyses. Therefore, a comparative analysis of serum concentrations of cardiac troponin and severity of cardiomyocyte injury could not be performed. In addition, no effort was made to quantify the extent and range of myocardial lesions, as could be done with morphometry or a more systematic approach involving multiple cross-sectional samples of each cardiac chamber. Therefore, the histologic findings may not have been representative in each dog.

SMALL ANIMALS

26. O’Brien P, Landt Y, Ladenson JH. Differential reactivity of cardiac and skeletal muscle from various species in a cardiac troponin I immunoassay. Clin Chem 1997;43:2333–2338. 27. O’Brien PJ, Dameron GW, Beck ML. Differential reactivity of cardiac and skeletal muscle from various species in two generations of cardiac troponin-T immunoassays. Res Vet Sci 1998;65:135–137. 28. O’Brien PJ, Dameron GW, Beck ML, et al. Cardiac troponin T is a sensitive, specific biomarker of cardiac injury in laboratory animals. Lab Anim Sci 1997;47:486–495. 29. Fredericks S, Merton GK, Lerena MJ, et al. Cardiac troponins and creatine kinase content of striated muscle in common laboratory animals. Clin Chim Acta 2001;304:65–74. 30. Tilley LP. Beurteilung der P, QRS und T-Ausschläge und Herzrhythmusstörungen. In: Tilley LP, ed. EKG bei Hund und Katze. Grundlagen, Auswertung und Therapie. Hannover, Germany: Schlütersche, 1989;55–203. 31. Fox E, Shorton K, Ulrich C. Sigma Stat user’s manual. San Rafael, Calif: Jandel Scientific, 1995;9-1–12-187. 32. Glatz SA. Primer of bio-statistics. New York: McGraw-Hill Health Professions Division, 1997. 33. Kidd L, Stepien RL, Amrheiw DP. Clinical findings and coronary artery disease in dogs and cats with acute and subacute myocardial necrosis: 28 cases. J Am Anim Hosp Assoc 2000; 36:199–208. 34. Falk T, Jönnson L. Ischemic heart disease in the dog: a review of 65 cases. J Small Anim Pract 2000;41:97–103.

388

Scientific Reports: Original Study

35. Kelly DF. Myocardial infarction (MI) in dogs. Vet Pathol 1995;32:550–554. 36. Pinson DM. Myocardial necrosis and sudden death after an episode of aggressive behavior in a dog. J Am Vet Med Assoc 1997; 211:1371–1372. 37. Voss EM, Sharkey SW, Gernert AE, et al. Human and canine cardiac troponin T and creatine kinase-MB distribution in normal and diseased myocardium. Infarct sizing using serum profiles. Arch Pathol Lab Med 1995;119:799–806. 38. Bertsch T, Bleuel H, Aufenanger J, et al. Comparison of cardiac troponin I and cardiac troponin T concentrations in peripheral blood during orciprenaline induced tachycardia in rats. Exp Toxic Pathol 1997;49:467–468. 39. Mair J, Wagner I, Morass B, et al. Cardiac troponin I release correlates with myocardial infarction size. Eur J Clin Chem Clin Biochem 1995;33:869–872. 40. Harris AS. Delayed development of ventricular ectopic rhythms following experimental coronary occlusion. Circulation 1950;1:1318–1328. 41. Hughes D. Lactate measurement: diagnostic, therapeutic, and prognostic implications. In: Bonagura JD, ed. Kirk’s current veterinary therapy XIII. Philadelphia: WB Saunders Co, 2000;112–116. 42. Jaffe AS, Ravkilde J, Roberts R, et al. It’s time for a change to a troponin standard. Circulation 2000;102:1216–1225. 43. Donaldson A, Cove-Smith R. Cardiac troponin levels in patients with impaired renal function. Hosp Med 2001;62:86–89.

JAVMA, Vol 221, No. 3, August 1, 2002