Cardiac arrhythmias have been reported in horses with gastrointestinal tract disorders.1–4 These arrhythmias can be caused by metabolic or electrolyte disturbances, hypovolemia, toxemia, and drugs such as inhalation anesthetics or can be associated with primary myocardial injury.4 Horses with acute colic often have endotoxemia, particularly those with strangulating gastrointestinal lesions or enterocolitis.5,6 Such horses may be more likely to develop cardiovascular complications associated with endotoxemia, including cardiac arrhythmias and myocardial injury.3 The prevalence of cardiac arrhythmias or cardiac damage in these horses is unknown and is probably underestimated. Ventricular arrhythmias have been reported in horses with strangulating and nonstrangulating small or large intestinal disorders, proximal enteritis, and salmonellosis.1–4 Myocardial degeneration and fibrosis have been documented in horses with duodenitis-proximal jejunitis and ventricular arrhythmias.4 Myocardial necrosis has been seen in horses with endotoxemia and large colon torsion.3,7
Primary myocardial disease is a diagnosis of exclusion that is usually made in patients with cardiac arrhythmias without other apparent causes. Echocardiograms in horses with myocarditis are usually normal other than abnormalities caused by the arrhythmia, unless there is severe myocardial injury.7,8 Serum concentration of CK-MB (the cardiac isoenzyme of CK) and serum lactate dehydrogenase activity (hydroxybutyrate dehydrogenase or lactate dehydrogenase 1 and 2) have been used in the past for diagnosis of myocardial cell injury; however, their sensitivity and specificity for diagnosis of myocardial cell damage in horses have been questioned.5,9,10 Cardiac troponin I is a sensitive and specific marker for myocardial injury in humans and dogs and more recently has been used with success in horses to detect myocardial damage.11–18
The primary objectives of the study reported here were to determine prevalence of myocardial injury in horses with colic on the basis of high concentrations of cTnI, frequency of cardiac arrhythmias within the first 24 to 48 hours after hospital admission, and associations between high cTnI concentrations and cardiac arrhythmias, clinical course, and outcome (survival to discharge from hospital vs nonsurvival [death or euthanasia]). Secondary objectives were to determine whether high concentrations of cTnI were associated with other clinical, laboratory, or physical examination findings typically associated with endotoxemia and whether there was a significant difference in cTnI concentration depending on the type or severity of disease. Our hypothesis was that cTnI concentration would be high (> 0.1 ng/mL) in horses with severe colic such as strangulating gastrointestinal tract lesions or enterocolitis and that cTnI concentration would be superior to CK-MB concentration for the detection of associated myocardial damage.
Materials and Methods
This study was performed prospectively. The study protocol was reviewed and approved by the University of Pennsylvania clinical investigation review committee, and owner consent was obtained. Horses with signs of colic were selected to enter the study. Other criteria to enter the study included collection of a blood sample at admission and an initial diagnosis of gastrointestinal tract disease. A 24-hour continuous ECG was recorded unless the horse died or was euthanized before the recording could be obtained. Cardiac troponin I and CK-MB concentrations were measured from a control group of 22 clinically normal horses. In addition, in 4 other horses that underwent surgical procedures for reasons other than colic, cTnI and CK-MB concentrations were measured before surgery and 12 and 24 hours after surgery, and a continuous 24-hour ECG was recorded after surgery.
Variables recorded at admission included age, breed, sex, heart rate, respiratory rate, rectal temperature, duration of colic signs (peracute [< 4 hours], acute [4 to ≤ 12 hours], and chronic [> 12 hours]), attitude (alert or signs of depression), degree of signs of pain (mild, moderate, or severe), presence of diarrhea, PCV, total protein concentration, WBC count, fibrinogen concentration, pH, Pco2, Po2, plasma lactate concentration, peritoneal WBC count and total protein concentration (when available), and plasma concentrations of sodium, potassium, chloride, calcium, and magnesium. For statistical analysis, a low magnesium concentration was defined as < 0.46 mmol/L and a high lactate concentration as > 2 mmol/L.
Blood for cTnI and CK-MB analysis was drawn via jugular venipuncture at admission in all horses and at 12 and 24 hours after admission in medically treated horses or at 12 and 24 hours after surgery in surgically treated horses. Samples were collected into lithium heparin tubes, and plasma was separated and either analyzed immediately or frozen at −70°C and analyzed within 1 month after collection. Cardiac troponin I and CK-MB concentrations were measured with a commercially available enzyme immunoassay systema that has a range of detection for cTnI concentration of 0 to 50 ng/mL and an analytic sensitivity of 0.03 ng/mL. This system uses antibodies against human cTnI that have been evaluated for cardiac tissue reactivity in horses, and validation of this measurement of cTnI concentration has been performed in horses.19 A cTnI concentration > 0.1 ng/mL and a CK-MB concentration > 10 ng/mL were considered abnormal for the statistical analysis. The upper reference limit for cTnI was calculated as < 2 SDs greater than the mean cTnI of the control group, similar to the reference values used in our laboratory. The upper reference limit for CK-MB was the highest value obtained from the control group of 22 horses.
A 24-hour Holter (ambulatory) ECG monitor was placed the morning after admission in medically treated cases or within 12 hours after surgery to record cardiac rhythm. The 24-hour Holter ECG was evaluated for the presence, type (supraventricular or ventricular), and frequency of arrhythmias. Isolated or continuous premature supraventricular or ventricular complexes, uniform or multiform ventricular arrhythmias, R on T phenomenon, isolated or sequential premature ventricular complexes, and ventricular tachycardia were all noted. More than 1 premature supraventricular complex or premature ventricular complex per hour, > 2 sequential premature ventricular complexes, ventricular tachycardia, R on T phenomenon, and multiform premature ventricular complexes were all considered abnormal for the statistical analysis.
Horses were treated medically or surgically. The outcome of the case was determined (survival to discharge from the hospital vs nonsurvival [death or euthanasia]). A necropsy was performed in all horses with a nonsurvival outcome. The lesions were categorized as strangulating, obstructive, inflammatory, and other (surgically treated nonstrangulated or obstructive lesions and medically treated noninflammatory lesions). The disease was considered severe if it was a strangulating surgical lesion or enterocolitis. Nonstrangulating surgical lesions were classified as being of moderate severity. Uncomplicated medically treated diseases and obstructive lesions that resolved with medical treatment were classified as mild.
Statistical analysis—Clinical and clinicopathologic data and information on the outcome were obtained. Descriptive statistics were calculated for continuous variables, with values reported as mean ± SD. Relationships between heart rate, absolute cTnI and CK-MB concentrations, treatment (medical or surgical), outcome, lesion type, disease severity, and occurrence of abnormal cardiac rhythms were examined by means of a nonparametric approach (Kruskal-Wallis analysis). Spearman rank correlation analysis was used to examine relationships between absolute cTnI and CK-MB concentrations and plasma lactate concentration and heart rate at admission.
Additionally, cases were grouped by whether plasma cTnI, CK-MB, or lactate concentrations were within the reference range for our laboratory (abnormal cTnI concentration, > 0.1 ng/mL; abnormal CK-MB concentration, > 10 ng/mL; and abnormal lactate concentration, > 2 mmol/L). The Fisher exact test was used to examine relationships between these grouped data and categorical clinical data. Changes in cTnI concentration from admission to 12 and 24 hours were analyzed by repeated-measures ANOVA. Statistical testing was performed with a commercial software package.b For all tests, values of P < 0.05 were considered significant.
Results
One hundred and thirteen horses with a complaint of colic were enrolled in the study. Two horses were excluded because they were determined to have primary cardiac disease at postmortem examination, leaving 111 horses in the study. Age ranged from 1 to 29 years (mean, 9.8 years). There were 53 Thoroughbreds, 13 Quarter Horses, 12 Standardbreds, 11 warmbloods, 7 draft horses, 6 Arabians, 3 Morgans, 3 crossbred horses, 2 ponies, and 1 American Saddlebred; 45 were females, 48 were geldings, and 18 were sexually intact males. On the basis of duration of clinical signs at admission, 17 horses had peracute colic, 52 had acute colic, and 37 had chronic colic. In 5 horses, duration of colic prior to admission was not recorded.
Fifty-one horses were treated surgically, and 60 horses were treated medically. Of these horses, 36 had strangulating lesions, 30 had obstructive lesions, 12 had enterocolitis (3 of which also had an obstructive lesion), and 36 had uncomplicated colic treated medically. Disease severity was mild in 48 horses, moderate in 17, and severe in 46. Surgical treatment was significantly (P = 0.004) associated with outcome; horses that underwent surgery were 7.86 times as likely to have a negative outcome.
The number of horses with abnormalities in the 24-hour Holter ECG, high cTnI concentrations, high CK-MB concentrations, high lactate concentrations, and low magnesium concentrations and the number of horses in which each test was completed were summarized (Table 1). Although 24-hour Holter ECG recordings were obtained in all horses, the recordings were nondiagnostic in 14 and were not completed in 10 that died or were euthanized before the recording was finished. Samples for CK-MB testing were obtained for all horses, but the kits malfunctioned in 9. In 5 horses, the values obtained for the plasma magnesium concentration were lost from the data set. Lactate analysis was inadvertently omitted from the clinical laboratory testing in 22 horses.
Distribution of 111 horses (No. affected [No. tested]) with colic grouped by treatment, gastrointestinal tract lesion type, and disease severity that had various abnormal findings.
| Treatment | Lesion type | Disease severity | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Variable | Surgical (n = 51) | Medical (n = 60) | Strangulating (n = 36) | Obstructive (n = 30 [3 inflammatory]) | Inflammatory (n = 12 [3 obstructive]) | Other (n = 36) | Severe (n = 46) | Moderate (n = 17) | Mild (n = 48) | Negative outcome (n = 14) |
| SVA | 5 (39) | 2 (48) | 1 (24) | 2 (28) | 2 (11) | 3 (27) | 3 (34) | 0 (13) | 4 (40) | 1 (4) |
| VA | 12 (39) | 6 (48) | 9 (24) | 6 (28) | 3 (11) | 1 (27) | 11 (34) | 3 (13) | 4 (40) | 2 (4) |
| SVA and VA | 1 (39) | 0 (48) | 1 (24) | 0 (28) | 0 (11) | 0 (27) | 1 (34) | 0 (13) | 0 (40) | 0 (4) |
| cTnI > 0.1 ng/mL at admission | 21 (51) | 17 (60) | 18 (36) | 6 (30) | 4 (12) | 10 (35) | 21 (46) | 5 (17) | 12 (48) | 10 (14) |
| cTnI > 0.1 ng/mL at any time | 34 (51) | 24 (60) | 27 (36) | 8 (30) | 7 (12) | 16 (35) | 33 (46) | 8 (17) | 17 (48) | 12 (14) |
| CK-MB > 10 ng/mL at admission | 26 (45) | 30 (57) | 23 (34) | 14 (26) | 7 (11) | 12 (32) | 29 (44) | 6 (13) | 21 (45) | 12 (13) |
| CK-MB > 10 ng/mL at any time | 44 (45) | 38 (58) | 33 (34) | 23 (26) | 9 (11) | 18 (33) | 42 (44) | 13 (13) | 27 (46) | 13 (13) |
| Mg < 0.46 mmol/L | 1 (47) | 7 (59) | 1 (36) | 2 (29) | 1 (12) | 4 (31) | 2 (44) | 0 (15) | 6 (47) | 0 (14) |
| Lactate > 2 mmol/L | 25 (46) | 10 (43) | 18 (32) | 10 (25) | 0 (6) | 7 (28) | 18 (39) | 7 (13) | 10 (37) | 7 (14) |
| Negative outcome | 10 | 4 | 11 | 1 | 1 | 1 | 13 | 1 | 0 | 14 |
SVA = Supraventricular arrhythmia. VA = Ventricular arrhythmia.
For the 22 clinically normal control horses, mean ± SD cTnI concentration was 0.048 ± 0.028 ng/mL (range, 0.02 to 0.15 ng/mL) and mean CK-MB concentration was 4.14 ± 2.04 ng/mL (range, 1.8 to 9.9 ng/mL). The cTnI concentrations in the 4 horses treated surgically for reasons other than colic were not significantly (P = 0.41) different in samples obtained at admission and 12 and 24 hours after surgery. Mean cTnI concentration in these horses was 0.022 ± 0.12 ng/mL (range, 0.01 to 0.04 ng/mL).
For all samples obtained from horses with colic, mean ± SD cTnI concentration was 0.43 ± 1.83 ng/mL (range, 0.0 to 15.42 ng/mL) and mean CK-MB concentration was 26.93 ± 31.96 ng/mL (range, 1 to 150 ng/mL). Fifty-eight (52.5%) horses had high cTnI concentration in at least 1 sample, and in 38 (34.2%) horses, the sample obtained at admission had high cTnI concentration.
Abnormal cTnI concentrations at admission were significantly correlated with surgical treatment (P = 0.045), ventricular arrhythmias (P = 0.018), and outcome (P = 0.017). Horses with an abnormal cTnI concentration at admission were 2.4 times as likely to have surgical treatment, 3.86 times as likely to have ventricular arrhythmias, and 4.17 times as likely to have a negative outcome. Abnormal CK-MB concentrations at admission were also significantly correlated with outcome (P = 0.005) and were associated with severe lesions (P = 0.03). Horses with high CK-MB concentration at admission were 12 times as likely to have a negative outcome.
Mean ± SD cTnI concentration at admission in the surgically treated horses was 0.74 ± 2.6 ng/mL versus 0.12 ± 0.3 ng/mL for the medically treated horses (P = 0.009). Absolute cTnI concentration at admission, but not at 12 or 24 hours, was significantly associated with surgical treatment (P = 0.009), a strangulating lesion (P = 0.026), severe disease (P = 0.02), and outcome (P = 0.013). Absolute CK-MB concentration at admission was significantly associated with severe disease (P = 0.016) and outcome (P = 0.005); greater CK-MB concentration at 24 hours was also significantly (P = 0.044) associated with outcome. Mean ± SD cTnI and CK-MB concentrations at admission for each of the treatment groups, type of lesions, and severity of lesions were summarized (Table 2). There were no significant associations with any of the other variables recorded.
Mean ± SD plasma cTnI and CK-MB concentrations (ng/mL) at initial evaluation of 111 horses with colic grouped by treatment, type of lesion, severity, outcome, ventricular arrhythmias detected, and plasma lactate concentration.
| cTnI | CK-MB | |||
|---|---|---|---|---|
| Variable | No. of horses | Mean ± SD | No. of horses | Mean ± SD |
| Treatment | ||||
| Surgical | 51 | 0.74 ± 2.6a | 45 | 31.6 ± 36.5a |
| Medical | 60 | 0.12 ± 0.3b | 57 | 22.7 ± 26.8a |
| Type of lesion | ||||
| Strangulated | 36 | 1.0 ± 3.2a | 34 | 33.2 ± 34.0a |
| Obstructive | 30 | 0.23 ± 0.4b | 26 | 19.7 ± 25.8a |
| Inflammatory | 12 | 0.15 ± 0.17b | 11 | 45.8 ± 49.7a |
| Other | 35 | 0.13 ± 0.32b | 32 | 20.7 ± 25.4a |
| Disease severity | ||||
| Severe | 46 | 0.82 ± 2.8a | 44 | 36.6 ± 37.8a |
| Moderate | 17 | 0.23 ± 0.4b | 13 | 19.7 ± 25.8b |
| Mild | 48 | 0.14 ± 0.32b | 45 | 20.0 ± 25.3b |
| Outcome | ||||
| Positive | 97 | 0.18 ± 0.38a | 89 | 24.4 ± 30.4a |
| Negative | 14 | 2.14 ± 4.8b | 13 | 44.2 ± 37.8b |
| Ventricular arrhythmias | ||||
| Absent | 69 | 0.14 ± 0.29a | 88 | 22.5 ± 29.3a |
| Present | 18 | 0.41 ± 0.65a | 14 | 32.3 ± 33.2a |
| Lactate | ||||
| < 2 mmol/L | 54 | 0.17 ± 0.4a | 48 | 17.6 ± 18.9a |
| ≥ 2 mmol/L | 35 | 0.97 ± 3.2a | 31 | 35.6 ± 36.3a |
Within a column, values with different superscript letters are significantly (P < 0.05) different for that variable. Results were not available for all horses for some variables.
Continuous 24-hour Holter recordings were obtained from 87 horses. Twenty-four (27.5%) horses had pathological cardiac arrhythmias detected during the Holter recording. Seventeen horses had ventricular arrhythmias, 6 had supraventricular arrhythmias, and 1 had both.
Mean ± SD plasma lactate concentration for horses requiring surgical treatment was 3.2 ± 0.34 mmol/L versus 1.6 ± 0.35 mmol/L for horses treated medically (P = 0.002). Mean lactate concentration for horses with ventricular arrhythmias was 4.2 ± 0.53 mmol/L versus 1.7 ± 0.3 mmol/L for the horses with no ventricular arrhythmias (P = 0.003). Mean lactate concentration was also significantly (P = 0.004) associated with outcome (survived to discharge from the hospital, 2.2 ± 0.3 mmol/L; died or euthanized, 3.9 ± 0.7 mmol/L).
An abnormal plasma lactate concentration (> 2 mmol/L) at admission was significantly correlated with surgical treatment (P = 0.004) and ventricular arrhythmias (P = 0.001). Horses with high lactate concentration were 3.93 times as likely to have surgical treatment and 6.86 times as likely to have ventricular arrhythmias.
Mean ± SD plasma magnesium concentration in horses with ventricular arrhythmias (0.64 ± 0.19 mmol/L) was significantly (P = 0.001) lower than in horses without ventricular arrhythmias (0.75 ± 0.14 mmol/L). The numbers of horses with low plasma magnesium concentration or an abnormal lactate concentration that also had high cTnI concentration or cardiac arrhythmias were summarized (Table 3).
Distribution (No. affected [No. tested]) of horses with arrhythmias, low magnesium concentration, or high lactate concentration and high (≥ 0.1 ng/mL) cTnI concentration at hospital admission or 12 or 24 hours after admission or surgery in a study of 111 horses with colic.
| cTnI ≥ 0.1 ng/mL | ||
|---|---|---|
| Variable | Admission | 12 or 24 hours |
| SVA | 2 (7) | 4 (7) |
| VA | 9 (18) | 13 (18) |
| SVA and VA | 0 (1) | 1 (1) |
| Magnesium < 0.46 mmol/L | 2 (8) | 2 (8) |
| Lactate > 2 mmol/L | 16 (35) | 16 (35) |
Heart rate was only weakly correlated with absolute cTnI (ρ = 0.29; P = 0.002) and CK-MB (ρ = 0.3; P = 0.002) concentrations. There was a significant association between heart rate and outcome (survived to discharge from the hospital, 54 ± 16 beats/min; died or euthanized, 68 ± 25 beats/min; P = 0.02) and between heart rate and treatment (medical, 50 ± 11 beats/min; surgical, 62 ± 21 beats/min; P = 0.002).
Fourteen horses died or were euthanized because of the severity of disease. Of these horses, 12 had high cTnI concentration and 13 had high CK-MB concentration (CK-MB concentration was not measured in 1 horse because of test kit malfunction). Four of the 14 horses had a 24-hour ECG completed before death, and arrhythmias were present in 3. Two horses had ventricular arrhythmias, and 1 horse had supraventricular arrhythmias (Table 1). A necropsy was performed in all horses that died or were euthanized, and histologic analysis of the myocardial tissue was performed in 5 horses. Samples of the left and right ventricular free walls, left and right atrial free walls, and interventricular septum were analyzed. Acute subendocardial and epicardial hemorrhage was found on histologic evaluation in one of these horses. This horse had high cTnI and CK-MB concentrations at admission and was not evaluated for arrhythmias because it died shortly afterwards. The other 4 horses with histologically normal myocardial tissue had high cTnI concentration at admission (n = 3) or in the second sample (1).
Discussion
A high prevalence of myocardial injury in horses with colic was detected; 58 (52.2%) horses had high cTnI concentration in 1 or more samples. In 38 (34.2%), the cTnI concentration was high at admission, similar to findings from a study20 that measured cTnI concentration at admission in horses with acute abdominal disease. These results suggested the possibility of concurrent myocardial damage. Myocardial injury in horses with acute abdominal disease could be caused by endotoxemia, sepsis, or myocardial hypoxia. Myocardial hypoxia secondary to hypotension during surgery was unlikely to be the sole cause of high postoperative cTnI concentrations in the horses in the present study, considering that cTnI concentrations in samples at initial evaluation were already high.
Endotoxemia was thought to be the most likely cause of myocardial injury in the horses in this study. Endotoxin may leak into the systemic circulation from damaged intestinal mucosa in horses with gastrointestinal tract disease, causing endotoxemia.5 Both endotoxemia and sepsis have been observed in horses with acute abdominal disease, causing release of inflammatory mediators and inducing shock and multiple-organ failure.21 In clinical studies,5,22 endotoxin was detected in the circulation of approximately 25% of horses with colic; however, these findings may underestimate the prevalence of endotoxemia because of the rapid clearance of endotoxin from the circulation.21 The administration of sublethal doses of endotoxin in horses and ponies causes arterial hypoxemia, systemic hypotension, lactic acidosis, and pulmonary hypertension.23 Experimentally induced endotoxemia reveals that the inflammatory response induced by endotoxin causes marked changes in the cardiovascular system, leading to hypotension, myocardial depression, multiple organ dysfunction, and death.24,25 A recent study26 detected increases in plasma cTnI concentration and cardiac arrhythmias in horses with experimentally induced endotoxemia, indicating that myocardial injury occurs in these horses.
Sepsis could have been a cause for the myocardial injury, as indicated by high cTnI concentrations and cardiac arrhythmias in some of the horses in this study. Horses with direct intestinal penetration by surgical, traumatic, or spontaneous rupture can develop sepsis.21 Cardiac troponin I concentrations are significantly increased in septicemic foals, compared with healthy foals22; this is presumably caused by acute myocardial injury. Septicemic foals with cardiac lesions at necropsy have higher cTnI concentrations than the general septicemic foal population.27 Myocardial dysfunction and high cTnI concentrations during sepsis are common in humans and have been postulated to occur because of cardiac oxygen demand exceeding supply, septic microemboli, toxic effects from endotoxin, and cardiac depressant effects from cytokines.28 High cTnI concentrations during sepsis in humans are associated with left ventricular dysfunction.28 These same mechanisms may be occurring in horses with colic and endotoxemia and sepsis.
Cardiac troponin I is a highly sensitive and specific clinical biomarker in a wide range of disease processes that cause cardiac injury in humans and other animals.29,30 Conservation of troponin structure across species allowed the cTnI test used in this study, which was designed for use in humans, to be used in horses.31–33 Differences in reported cTnI concentrations in clinically normal horses depend on the assay used because each cTnI test kit measures concentrations of different epitopes and fragments of cTnI. All cTnI concentrations in the present study were determined on the same analyzer. Our reference limit was similar to that reported for a similar analyzer.34
Cardiac troponin I is nearly undetectable in plasma or serum of clinically normal horses and increases with cardiac injury.11,12,15,35 High cTnI concentrations have been reported in horses with a variety of conditions such as intense exercise,36 infectious anemia and piroplasmosis,35 multiform ventricular tachycardia,12 experimental exposure to monensin, natural exposure to lasalosid, atypical myopathy, and rattlesnake envenomation.15–17,37,38 Troponin concentration in humans is presently used not only for identification of myocardial injury but also as a prognostic indicator because troponin concentration correlates with clinical severity of disease and life expectancy.29 The cTnI concentration also correlates with the extent of myocardial injury in dogs.39 In the present study, an abnormal cTnI concentration at admission was significantly correlated with the occurrence of ventricular arrhythmias, the need for surgical treatment, and outcome, which may indicate more severe cardiac injury in these horses. Also, higher cTnI concentrations at admission were significantly associated with surgical treatment, a strangulating lesion, severe gastrointestinal tract disease, and a negative outcome. Our data suggest that cTnI concentration measured at admission may be useful as an indicator of severity of disease and a predictor of outcome and the likelihood of concurrent myocardial injury. These results indicated that horses with more severe gastrointestinal tract disease had abnormal and higher cTnI concentrations and possibly more serious myocardial injury. Studies of cardiac tissue at necropsy in such cases would be needed to confirm this correlation. We cannot conclude from our results that the magnitude of the cTnI increase correlates with the degree of myocardial injury in horses; however, further studies may lead to that conclusion.
The present study revealed that horses with colic and higher cTnI concentrations within 24 hours of admission or surgery were more likely to need surgical treatment and to have severe gastrointestinal tract disease, a strangulating lesion, and a negative outcome. These findings are similar to a study in which horses undergoing emergency abdominal surgery with high postoperative cTnI concentration were more likely to have a negative outcome.40 In a recent study20 in horses with acute abdominal disease, the proportion of horses with high cTnI concentration was significantly greater among horses with strangulating or inflammatory lesions, compared with healthy horses, and in nonsurvivors, compared with survivors. Our findings agree with the results of that study,20 in that the strangulating and inflammatory lesions were included in the severe group of gastrointestinal tract disease. Dogs with gastric dilatation-volvulus commonly develop cardiovascular complications (including shock, ischemia, and reperfusion injury) and arrhythmias. Cardiac troponin I concentration is high in a large percentage of dogs with gastric dilatation-volvulus.41,42 All dogs that died in those studies had high cTnI concentration after surgery.41,42
In the present study, cTnI concentration was determined at admission and 12 and 24 hours after admission or surgery. In humans, cTnI concentration increases at 4 to 6 hours and peaks at 18 to 24 hours after the onset of clinical signs.29 We chose our sampling times assuming that cTnI release and clearance in horses are similar to that in humans. A previous study43 revealed higher cTnI concentrations 3 to 6 hours after maximal exercise in healthy adult horses, compared with other time points. In horses with experimentally induced endotoxemia, significant increases in cTnI concentration were detected as early as 4 hours after the start of the endotoxin infusion but were not significantly different from preinfusion values at 6 hours after infusion.26 In addition, the reported half-life of cTnI in ponies is short.44 Therefore, additional sampling in the present study between the admission and 12-hour sampling time may have yielded more horses with high cTnI concentration. In the present study, horses were evaluated at different times following the appearance of clinical signs, so samples were obtained at various times depending on the chronicity of colic. High cTnI concentration was detected in 34.2% of the horses at admission, and an additional 20 horses that had a cTnI concentration within the reference limit at admission had high cTnI concentration in the 12- or 24-hour sample. In a previous study,20 cTnI concentration was high at admission in 32% of horses with acute abdominal disease. In another study40 in horses treated with emergency abdominal surgery, high cTnI concentrations detected in the postoperative period were more predictive of death than were high cTnI concentrations detected in the preoperative period. We conclude from our results that several samples obtained at different times are more likely to permit detection of high cTnI concentrations than 1 sample obtained at admission from horses with colic.
The effects of surgery or the stress of hospital admission was evaluated by measuring cTnI concentration in 4 horses undergoing surgical procedures for reasons other than colic. In each horse, cTnI concentration was evaluated at admission and 12 and 24 hours after surgery. Cardiac troponin I concentrations in the 12 samples obtained from these horses were within the reference range and were not significantly different from each other, making surgery and anesthesia by themselves an unlikely cause of increased cTnI concentration in the horses in this study. In a study45 of horses undergoing general anesthesia for elective surgery or imaging, all horses had cTnI concentrations (determined with the same analyzer as in the present study) in the reference range for up to 24 hours after anesthesia. Similar results were found in a study46 evaluating the effect of transvenous electrical cardioversion on plasma cTnI concentrations in which 5 horses undergoing general anesthesia for laryngeal or orthopedic procedures were used as controls. A small increase in cTnI concentration has been reported in some horses treated by use of transvenous electrical cardioversion, most likely associated with the delivery of the electrical shocks to the myocardium.46
Serum CK-MB concentration has been widely used for diagnosis of myocardial cell injury, but its sensitivity and specificity for myocardial cell damage in horses and other species have been questioned.31 The isoenzyme CK-MB can also be found in skeletal muscle as well as the heart, and the stress and restraint that occur in animals during handling can cause its release.31 In the present study, a high CK-MB concentration at admission was significantly associated with severe disease and outcome. We have to consider that CK-MB concentration could have also been high in many of these cases because of skeletal muscle damage or stress. The horses with negative outcome had a greater chance of having more serious muscle damage from rolling or lying down, which could have caused an increase in CK-MB concentration. A larger number of horses in this study had high CK-MB concentration than had high cTnI concentration, but this was most likely attributable to the lack of organ specificity of CK-MB. The fact that high CK-MB concentration was not associated with cardiac arrhythmias, but high cTnI concentration was associated with ventricular arrhythmias, may further indicate that the high CK-MB concentration could also have been caused by skeletal muscle release.
Cardiac arrhythmias have been reported in horses with gastrointestinal tract disorders.1–4 Ventricular arrhythmias are an indication of myocardial disease.3 The majority (70.5%) of horses with ventricular arrhythmias in the present study also had high cTnI concentrations, consistent with concurrent myocardial injury. Of these horses, 27.5% of those for which a continuous 24-hour ECG (Holter monitor) was completed had serious cardiac arrhythmias, and there was an association between abnormal cTnI concentrations and the occurrence of ventricular arrhythmias, further suggesting myocardial injury. A recent study20 failed to detect cardiac arrhythmias in horses with acute abdominal disease and high cTnI concentrations, in contrast to the present study. However, the monitoring time during that study20 was shorter than in the present study. Results of the present study indicated considerable prevalence of cardiac arrhythmias and evidence of myocardial damage as detected by use of high cTnI concentration in horses with colic. In addition, the treatments received during hospitalization were not noted, and some horses received drugs commonly used to treat colic, such as lidocaine, that could have decreased the frequency of postoperative ventricular arrhythmias.
Electrolyte imbalances, in particular hypokalemia, hypocalcemia, and hypomagnesemia, have been associated with cardiac arrhythmias in humans and horses.12,47 In the present study, horses that had ventricular arrhythmias had lower magnesium concentration than horses with no arrhythmias. No other electrolyte concentrations were significantly different between groups in these horses. Therefore, the contribution of other electrolyte abnormalities to development of cardiac arrhythmias in the horses in this study seemed to be unimportant. Only 4 of 18 horses with ventricular arrhythmias and none with supraventricular arrhythmias had a plasma magnesium concentration < 0.46 mmol/L. Two of the horses with ventricular arrhythmias also had an abnormal cTnI concentration, indicating that hypomagnesemia as the possible sole cause for these arrhythmias was infrequent but was possible in the other 2 horses.
In this study, high blood lactate concentration at admission was significantly correlated with surgical treatment, ventricular arrhythmias, and outcome. In a study40 evaluating l-lactate and cTnI concentrations in horses undergoing emergency abdominal surgery, high blood lactate concentration at hospital admission was present in the majority (88%) of horses, most likely secondary to perfusion deficits. Hyperlactemia occurs secondary to hypovolemia and tissue hypoxia or secondary to an inflammatory process.48 Persistent hyperlactemia in horses with colic has been associated with poor prognosis.40
The small number of horses in each category was a limitation in this study. Also, most horses survived to discharge from the hospital, making outcome comparisons difficult. Most of the horses that died or were euthanized did not have a 24-hour ECG or all of the samples drawn before death, making it difficult to detect associations between cTnI concentration and arrhythmias or other variables. A thorough cardiovascular postmortem examination was not performed on all horses, which limited the ability to detect small focal areas of myocardial injury.
This study revealed a high prevalence of high cTnI concentrations in horses with colic, indicative of myocardial injury; there were also a high percentage of horses with colic that had cardiac arrhythmias. Horses with colic and high cTnI concentrations had a less favorable prognosis. Cardiac troponin I concentration may be useful as a predictor of the severity of gastrointestinal tract disease and outcome in horses. The isoenzyme CK-MB was a better indicator of outcome than cTnI but was not indicative of cardiac injury in this group of horses.
ABBREVIATIONS
| CK | Creatine kinase |
| cTnI | Cardiac troponin I |
Stratus CS stat fluorometric analyzer, Stratus CS, Dade-Behring, Bear, Del.
JMP, version 4.0.4, SAS Institute Inc, Cary, NC.
References
1. Cornick JL, Seahorn TL. Cardiac arrhythmias identified in horses with duodenitis/proximal jejunitis: six cases (1985–1988). J Am Vet Med Assoc 1990; 197: 1054–1059.
2. Reimer JM, Sweeney RW, Reef VB. ECG of the month. J Am Vet Med Assoc 1997; 191: 42–43.
3. Reimer JM, Reef VB, Sweeney RW. Ventricular arrhythmias in horses: 21 cases. J Am Vet Med Assoc 1992; 201: 1237–1243.
4. Wilkins PA, Bain FT. ECG of the month. J Am Vet Med Assoc 1993; 203: 972–973.
5. King JN, Gerring EL. Detection of endotoxemia in cases of equine colic. Vet Rec 1988; 123: 269–271.
6. McClure JJ, McClure JR, Johnson DW. Detection of endotoxin in the blood and peritoneal fluid in ponies with experimentally created strangulation obstruction of the small intestine, in Proceedings. 28th Annu Meet Am Asssoc Equine Pract 1982; 1102–1108.
7. Bonagura JD, Reef VB, Schwarzwald CC. Cardiovascular diseases. In: Reed S, Bayly W, Sellon D, eds. Equine internal medicine. 3rd ed. Philadelphia: WB Saunders Co, 2010;372–487.
8. Reef VB. Cardiovascular ultrasonography. In: Reef VB, ed. Equine diagnostic ultrasound. Philadelphia: WB Saunders Co, 1998;215–272.
9. Argiroudis SA, Joyce EK, Blackmore DJ. Observations on the isoenzymes of creatine kinase in equine serum and tissues. Equine Vet J 1982; 14: 317–321.
10. Fujii Y, Ikeda S, Watanabe H. Analysis of creatine kinase isoenzyme in racehorse serum and tissues. Bull Equine Res Inst 1983; 17: 21–31.
11. Cornelisse CJ, Schott HC II, Olivier NB, et al. Concentration of cardiac troponin I in a horse with a ruptured aortic regurgitation jet lesion and ventricular tachycardia. J Am Vet Med Assoc 2000; 217: 231–235.
12. Schwarzwald CC, Hardy J, Buccellato M. High cardiac troponin I serum concentration in a horse with multiform ventricular tachycardia and myocardial necrosis. J Vet Intern Med 2003; 17: 364–368.
13. Schoeber K, Kirbach B, Oechtering G. Myocardial cell injury from traumatic, degenerative, and metabolic heart disease in dogs, in Proceedings. 17th Annu Meet Am Coll Vet Intern Med 1999;66–67.
14. Smith SC, Landeson JH, Mason JW, et al. Elevations of cardiac troponin I associated with myocarditis. Experimental and clinical correlates. Circulation 1997; 95: 163–168.
15. Divers TJ, Kraus MS, Jesty SA, et al. Clinical findings and serum cardiac troponin I concentrations in horses after intragastric administration of sodium monensin. J Vet Diagn Invest 2009; 21: 338–343.
16. Decloedt A, Verheyen T, De Clercq D, et al. Acute and long-term cardiomyopathy and delayed neurotoxicity after accidental lasalocid poisoning in horses. J Vet Intern Med 2012; 26: 1005–1011.
17. Verheyen T, Decloedt D, De Clercq D, et al. Cardiac changes in horses with atypical myopathy. J Vet Intern Med 2012; 26: 1019–1026.
18. Nath LC, Anderson GA, Hinchcliff KW, et al. Serum cardiac troponin I concentrations in horses with cardiac disease. Aust Vet J 2012; 90: 351–357.
19. Slack J, Boston RC, Soma L, et al. Cardiac troponin I in racing Standardbreds. J Vet Intern Med 2012; 26: 1202–1208.
20. Nath LC, Anderson GA, Hinchcliff KW, et al. Clinicopathologic evidence of myocardial injury in horses with acute abdominal disease. J Am Vet Med Assoc 2012; 241: 1202–1208.
21. Moore JN, Morris DD. Endotoxemia and septicemia in horses: experimental and clinical correlates. J Am Vet Med Assoc 1992; 200: 1903–1914.
22. Meyers K, Reed S, Keck M, et al. Circulating endotoxin like substance(s) and altered hemostasis in horses with gastrointestinal disorders: an interim report. Am J Vet Res 1982; 43: 2233–2238.
23. Moore JN, Garner HE, Shapland JE, et al. Lactic acidosis and arterial hypoxemia during sublethal endotoxemia in conscious ponies. Am J Vet Res 1980; 41: 1696–1698.
24. Natanson C, Eichenholz P, Danner R, et al. Endotoxin and tumor necrosis factor challenges in dogs simulate the cardiovascular profile of human septic shock. J Exp Med 1989; 169: 823–832.
25. Suffredini AF, Fromm RE, Parker MM, et al. The cardiovascular response of normal humans to the administration of endotoxin. N Engl J Med 1989; 321: 280–287.
26. Nostell K, Brojer J, Hoglund K, et al. Cardiac troponin I and the occurrence of cardiac arrhythmias in horses with experimentally induced endotoxemia. Vet J 2012; 192: 171–175.
27. Slack JA, Mc Guirk SM, Erb HN, et al. Biochemical markers of cardiac injury in normal, surviving septic, or nonsurviving septic neonatal foals. J Vet Intern Med 2005; 19: 577–580.
28. 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.
29. O'Brien PJ. Cardiac troponin is the most effective translational safety biomarker for myocardial injury in cardiotoxicity. Toxicology 2008; 245: 206–218.
30. Babuin L, Jaffe AS. Troponin: the biomarker of choice for the detection of cardiac injury. CMAJ 2005; 173: 1191–1202.
31. O'Brien PJ, Landt Y, Lsdenson JH. Differential reactivity of cardiac and skeletal muscle from various species in a cardiac troponin I immunoassay. Clin Chem 1997; 43: 2333–2338.
32. O'Brien PJ, Dameron GW, Beck ML, et al. 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.
33. Rishniw M, Simpson KW. Cloning and sequencing of equine cardiac troponin I and confirmation of its usefulness as a target analyte for commercial troponin I analyzers. J Vet Diagn Invest 2005; 17: 582–584.
34. Smith GW, Constable PD, Foreman JH, et al. Cardiovascular changes associated with intravenous administration of fumonisin B1 in horses. Am J Vet Res 2002; 63: 538–545.
35. Diana A, Guglielmini C, Candini D, et al. Cardiac arrhythmias associated with piroplasmosis in the horse: a case report. Vet J 2007; 174: 193–195.
36. Hoolbrook TC, Birks EK, Sleeper MM, et al. Endurance exercise is associated with increased plasma cardiac troponin I in horses. Equine Exerc Physiol 2006; 36: 27–31.
37. Decloedt A, Verheyen T, Sys S, et al. Tissue doppler imaging and 2-dimensional speckle tracking of left ventricular function in horses exposed to lasalocid. J Vet Intern Med 2012; 26:1009–1016.
38. Gilliam LL, Holbrook TC, Ownby CL, et al. Cardiotoxicity, inflammation, and immune response after rattlesnake envenomation in the horse. J Vet Intern Med 2012; 26: 1457–1463.
39. 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.
40. Radcliffe RM, Divers TJ, Fletcher DJ, et al. Evaluation of l-lactate and cardiac troponin I in horses undergoing emergency abdominal surgery. J Vet Emerg Crit Care (San Antonio) 2012; 22: 313–319.
41. Schober KE, Cornand C, Kirbach B, et al. Serum cardiac troponin I and cardiac troponin T concentrations in dogs with gastric dilatation-volvulus. J Am Vet Med Assoc 2002; 221: 381–388.
42. Burgener IA, Kovacevic A, Mauldin N, et al. Cardiac troponins as indicators of acute myocardial damage in dogs. J Vet Intern Med 2006; 20: 277–283.
43. Durando MM, Reef VB, Kline K, et al. Acute effects of short duration maximal exercise on cardiac troponin I in healthy horses. Equine Comp Exerc Physiol 2006; 4: 217–223.
44. Kraus MS, Kaufer BB, Damiani A, et al. Elimination half-life of intravenously administered equine cardiac troponin I in healthy ponies. Equine Vet J 2013; 45: 56–59.
45. Slack J, Boston R, Driessen B, et al. Effect of general anesthesia on plasma cardiac troponin I concentrations in healthy horses. J Vet Cardiol 2011; 13: 163–169.
46. Jesty SA, Kraus MS, Gelzer AR, et al. Effect of transvenous electrical cardioversion on plasma cardiac troponin I concentrations in horses with atrial fibrillation [Erratum published in J Vet Intern Med 2009; 23:1357]. J Vet Intern Med 2009; 23: 1103–1107.
47. Opie LH. Electricity out of control: ventricular arrhythmias. In: Opie LH, ed. The heart. 3rd ed. Philadelphia: Lippincott Williams & Wilkins, 1998;589–608.
48. Tennent-Brown BS, Wilkins PA, Lindborg S, et al. Sequential plasma lactate concentrations as prognostic indicators in adult equine emergencies. J Vet Intern Med 2010; 24: 198–205.
