Cardiac troponin I is a sensitive and specific indicator of myocardial injury.1 In domestic animals, plasma or serum cTnI concentration increases as a result of primary myocardial disease or CHF associated with various cardiac diseases.2–9 Additionally, plasma cTnI concentration can be increased because of noncardiac diseases or conditions that result in clinically observable myocardial injury (eg, gastric dilatation and volvulus or trauma) or clinically inapparent myocardial damage.10–15
To our knowledge, investigations of the use of circulating cTnI concentration measurement for veterinary medical purposes have been limited; studies have included evaluation of the sensitivity and specificity of plasma or serum cTnI concentration for diagnosis of hypertrophic cardiomyopathy in cats2,5 and differentiation of congenital from acquired heart disease and classification of heart failure severity in dogs.9 In these instances, investigators compared affected and clinically normal animals.2,5,9 However, with the exception of screening tests, clinicians generally attempt to distinguish between possible causes of clinical signs in ill animals, rather than distinguishing between clinically normal and affected animals. Thus, given the finding that both cats and dogs with noncardiac disease can have increased circulating cTnI concentrations,13–15 it would be more appropriate to compare data obtained from animals with similar clinical signs, but diseases of different etiologies, to determine the true usefulness of assessment of circulating cTnI concentration for the diagnosis of cardiac disease as the cause of respiratory distress or dyspnea.
The purpose of the multicenter study reported here was to determine whether plasma cTnI concentration can be used to discriminate cardiac from noncardiac causes of dyspnea in cats. To this end, plasma cTnI concentration was assessed in cats with various forms of cardiac disease and in cats with primary respiratory causes of dyspnea. We hypothesized that measurement of plasma cTnI concentration could help differentiate cats with D-CHF from those with D-NCC.
Materials and Methods
Cats—Cats with dyspnea were identified prospectively at the cardiology referral clinics of the Ryan Veterinary Hospital of the University of Pennsylvania, Oradell Animal Hospital, and Redbank Animal Hospital. Cats were included in the study if they met previously described criteria2 with 2 minor modifications, namely that cats with CHF secondary to all forms of cardiomyopathy and cats with CHF secondary to hyperthyroidism were eligible for inclusion. All dyspneic cats underwent echocardiography and thoracic radiography. Congestive heart failure was identified on the basis of radiographic and echocardiographic findings (eg, large left atrium and evidence of severe cardiac disease; pulmonary artery blood pressure was estimated when possible) and response to diuretics. Cats with D-NCC underwent additional diagnostic procedures as necessary (transtracheal wash, pleurocentesis, bronchoscopy, exploratory thoracotomy, and fine-needle aspiration) to identify the cause of the pulmonary disease.
Plasma cTnI analysis—Plasma concentrations of cTnI were analyzed as previously described.2,16 Briefly, a blood sample (1 mL) was collected from each cat into a heparinized tube and centrifuged; plasma was analyzed within 2 hours of collection or stored at −80°C until analysis. Analyses were performed in a fluorometric analyzera with a 2-site sandwich assay that was based on solid-phase radial partition immunoassay technology.
Statistical analysis—Dyspneic cats were classified as having either cardiac disease with CHF or a noncardiac cause of respiratory distress. The plasma cTnI concentration between groups was compared by use of the Mann-Whitney U test. A scatterplot of the data was created for visual inspection and comparison of results. Data from a previous study2 of plasma cTnI concentrations in cats with subclinical cardiac disease and additional data obtained subsequently were included in the scatterplot to further define the clinical use of plasma cTnI concentration in discriminating causes of dyspnea in cats.
For the sensitivity and specificity analyses performed on data from the 2 groups of study cats, 4 diagnostic threshold values were established: 0.19, 0.5, 1.0, and 1.43 ng/mL. These threshold values were chosen to represent the 100% specificity thresholds for identifying either cardiac or noncardiac causes of dyspnea (0.19 and 1.43 ng/mL) and 2 additional thresholds that equally spanned the threshold range (0.5 and 1.0 ng/mL). Cumulative true-positive and false-positive rates were calculated, and a receiver operating characteristic curve was determined via logarithmic fitting. For all analyses, a value of P < 0.05 was considered significant.
Results
Forty-three dyspneic cats were included in the study; 31 had D-CHF and 12 had D-NCC. Among the cats with D-CHF, 13 had hypertrophic cardiomyopathy, 8 had unclassified cardiomyopathy, 7 had dilated cardiomyopathy, 2 had hyperthyroidism, and 1 had tricuspid valve dysplasia. Among the cats with D-NCC, 4 had asthma, 3 had neoplastic pleural effusion, 2 had idiopathic pleural effusion, 1 had a lung tumor, and 1 had a laryngeal tumor; the diagnosis for 1 cat was not determined.
Plasma cTnI concentrations for the cats in the 2 groups were plotted (Figure 1). Cats with D-CHF had significantly (P < 0.001) greater cTnI concentration (median, 1.59 ng/ mL; range, 0.2 to 30.24 ng/mL) than cats with D-NCC (median, 0.165 ng/mL; range, 0.01 to 1.42 ng/mL). A third subset of cats with subclinical cardiac disease, some of which were included in a previous study,2 had plasma cTnI concentrations that spanned both D-CHF and D-NCC ranges (median, 0.39 ng/mL; range, 0.05 to 6.36 ng/mL). Data from this third group were not statistically compared with data from the other 2 groups.
Scatterplot of plasma cTnI concentrations in 31 cats with D-CHF and 12 cats with D-NCC; data from 19 nondyspneic cats with heart disease but no CHF (HD-no CHF) are also provided (data for 9 of these cats were previously reported2). The shaded area represents the reference interval for plasma cTnI concentration in healthy cats.15 With regard to diagnosis of CHF by use of plasma cTnI concentration measurement, the dashed line identifies 100% sensitivity (0.2 ng/mL), and the dotted line identifies 100% specificity (1.42 ng/mL). Fourteen of 31 cats with D-CHF had plasma cTnI concentrations < 1.42 ng/mL, and 6 of 21 cats with D-NCC had plasma cTnI concentrations > 0.2 ng/mL. Thus, by use of the 100% sensitivity and specificity threshold values, measurement of plasma cTnI concentration would discriminate between cardiac and noncardiac causes of dyspnea in approximately 50% of cats. Notice that 5 of the 19 cats with heart disease but no CHF had plasma cTnI concentrations > 1.42 ng/mL. The asterisk represents 1 cat with D-CHF that had a value (30.24 ng/mL) that exceeded the vertical scale range.
Citation: Journal of the American Veterinary Medical Association 233, 8; 10.2460/javma.233.8.1261
Scatterplot of plasma cTnI concentrations in 31 cats with D-CHF and 12 cats with D-NCC; data from 19 nondyspneic cats with heart disease but no CHF (HD-no CHF) are also provided (data for 9 of these cats were previously reported2). The shaded area represents the reference interval for plasma cTnI concentration in healthy cats.15 With regard to diagnosis of CHF by use of plasma cTnI concentration measurement, the dashed line identifies 100% sensitivity (0.2 ng/mL), and the dotted line identifies 100% specificity (1.42 ng/mL). Fourteen of 31 cats with D-CHF had plasma cTnI concentrations < 1.42 ng/mL, and 6 of 21 cats with D-NCC had plasma cTnI concentrations > 0.2 ng/mL. Thus, by use of the 100% sensitivity and specificity threshold values, measurement of plasma cTnI concentration would discriminate between cardiac and noncardiac causes of dyspnea in approximately 50% of cats. Notice that 5 of the 19 cats with heart disease but no CHF had plasma cTnI concentrations > 1.42 ng/mL. The asterisk represents 1 cat with D-CHF that had a value (30.24 ng/mL) that exceeded the vertical scale range.
Citation: Journal of the American Veterinary Medical Association 233, 8; 10.2460/javma.233.8.1261
Scatterplot of plasma cTnI concentrations in 31 cats with D-CHF and 12 cats with D-NCC; data from 19 nondyspneic cats with heart disease but no CHF (HD-no CHF) are also provided (data for 9 of these cats were previously reported2). The shaded area represents the reference interval for plasma cTnI concentration in healthy cats.15 With regard to diagnosis of CHF by use of plasma cTnI concentration measurement, the dashed line identifies 100% sensitivity (0.2 ng/mL), and the dotted line identifies 100% specificity (1.42 ng/mL). Fourteen of 31 cats with D-CHF had plasma cTnI concentrations < 1.42 ng/mL, and 6 of 21 cats with D-NCC had plasma cTnI concentrations > 0.2 ng/mL. Thus, by use of the 100% sensitivity and specificity threshold values, measurement of plasma cTnI concentration would discriminate between cardiac and noncardiac causes of dyspnea in approximately 50% of cats. Notice that 5 of the 19 cats with heart disease but no CHF had plasma cTnI concentrations > 1.42 ng/mL. The asterisk represents 1 cat with D-CHF that had a value (30.24 ng/mL) that exceeded the vertical scale range.
Citation: Journal of the American Veterinary Medical Association 233, 8; 10.2460/javma.233.8.1261
A receiver operating characteristic curve was plotted to examine whether plasma cTnI concentrations could be used to differentiate cats with D-CHF from cats with D-NCC (Figure 2). The area under the curve was 0.844, which was consistent with a good fit. A plasma cTnI concentration ≥ 0.2 ng/mL had 100% sensitivity, 58% specificity, and 88% diagnostic accuracy for determination of D-CHF in cats. Plasma cTnI concentration > 1.42 ng/mL had 58% sensitivity, 100% specificity, and 66% diagnostic accuracy for determination of D-CHF in cats.
Receiver operating characteristic curve analysis to determine the accuracy of plasma cTnI concentration measurement for the diagnosis of CHF as the cause of dyspnea in 43 cats. The area under the curve is 0.844.
Citation: Journal of the American Veterinary Medical Association 233, 8; 10.2460/javma.233.8.1261
Receiver operating characteristic curve analysis to determine the accuracy of plasma cTnI concentration measurement for the diagnosis of CHF as the cause of dyspnea in 43 cats. The area under the curve is 0.844.
Citation: Journal of the American Veterinary Medical Association 233, 8; 10.2460/javma.233.8.1261
Receiver operating characteristic curve analysis to determine the accuracy of plasma cTnI concentration measurement for the diagnosis of CHF as the cause of dyspnea in 43 cats. The area under the curve is 0.844.
Citation: Journal of the American Veterinary Medical Association 233, 8; 10.2460/javma.233.8.1261
Discussion
In our study, all cats with cardiogenic dyspnea (ie, D-CHF) had plasma cTnI concentrations > 0.2 ng/mL. Conversely, of the 12 cats with D-NCC, 6 had plasma cTnI concentrations < 0.2 ng/mL, but 6 had values that exceeded the reference limit for healthy cats (< 0.17 ng/mL) obtained by use of the same analyzer system.15 The increases in plasma cTnI concentration in cats with D-NCC were typically smaller than those detected in cats with D-CHF, and no cats with D-NCC had plasma cTnI concentrations > 1.42 ng/mL. Thus, measurement of plasma cTnI concentration may help rule out CHF in a substantial portion of dyspneic cats; if the plasma cTnI concentration is within or near the reference range values for healthy cats, dyspnea in an affected cat can be attributed to noncardiac causes. Similarly, dyspneic cats with plasma cTnI concentrations that exceed 1.42 ng/mL most likely have CHF as the cause of respiratory distress. Thus, detection of a high plasma cTnI concentration (> 1.42 ng/ mL) in a dyspneic cat may provide a clinician with sufficient diagnostic evidence to provide initial treatment for CHF (eg, administration of diuretic agents), thereby sparing the patient additional, potentially compromising diagnostic testing or inappropriate fluid administration. Conversely, detection of a plasma cTnI concentration < 0.2 ng/mL in a dyspneic cat would indicate that a clinician should pursue diagnosis and treatment of noncardiac causes of dyspnea, thereby avoiding inappropriate administration of diuretics.
In dyspneic cats, plasma cTnI concentrations of 0.2 to 1.42 ng/mL were not indicative of the cause of dyspnea. Additional discriminatory diagnostic testing, such as thoracic radiography or echocardiography, would be required to determine the cause of the dyspnea.
In the present study, more than half of the cats could be confidently stratified as having D-CHF or D-NCC solely on the basis of the results of the plasma cTnI analysis. Whether these findings are reflective of the dyspneic feline population at large remains to be determined by further studies. Our results differ from recently published results for dogs, wherein assessment of plasma cTnI concentrations failed to differentiate causes of dyspnea.14
There are several limitations to our study. First, the group of cats with D-NCC was small (n = 12) and did not include cats with all types of respiratory disease that may be associated with dyspnea. For example, no confirmed cases of pulmonary hypertension resulting from primary lung disease or pulmonary thromboembolism were included in the study. Pulmonary thromboembolism is known to cause increases in circulating cTnI concentration as a result of damage to the right ventricular myocardium.17 Although a definitive diagnosis was not obtained in 1 cat in the present study, CHF was excluded as a cause of dyspnea, so the cat was included in the D-NCC group; the plasma cTnI concentration in this cat was 0.87 ng/mL. Excluding this cat from analysis did not alter the study findings—plasma cTnI concentrations in the 2 groups remained significantly different. The small group of cats with D-NCC likely reflects the proportion of such patients that are examined at referral centers; most cats with asthma or other noncardiac causes of dyspnea are likely managed by primary care providers. Therefore, the results of our study should be interpreted cautiously in light of the small number of cats with D-NCC, and additional studies may be warranted to confirm our findings.
Results of a previous study2 involving one of the authors indicated that cats with subclinical cardiac disease often have plasma cTnI concentrations > 0.17 ng/mL; however, in the 9 cats with subclinical hypertrophic cardiomyopathy (ie, no clinical signs) evaluated in that study, plasma cTnI concentration did not exceed 2 ng/mL. We subsequently examined an additional 10 cats with subclinical cardiac disease and found that plasma cTnI concentration in 4 of these cats was 2.2 to 6.4 ng/mL. This finding is noteworthy because if these cats had concurrent D-NCC, they would have nevertheless been classified as having CHF on the basis of plasma cTnI values alone. Additionally, results of another study15 have suggested that renal disease may result in increases in circulating cTnI concentrations in cats, further confounding attempts to discriminate causes of dyspnea in cats with comorbidities.
Although the receiver operating characteristic curve analysis revealed that the accuracy of plasma cTnI concentrations for identification of CHF in dyspneic cats was fairly high (area under the curve, 0.844), we did not provide optimal diagnostic threshold values because dyspnea associated with cardiac or noncardiac causes requires correct diagnosis—false-positive or false-negative diagnoses of D-CHF or D-NCC can be fatally compromising to the patient. Instead, we elected to report threshold values that minimized the improper diagnosis of either condition.
The causes of increased plasma cTnI concentrations in cats with D-NCC remain undetermined. Other investigators have speculated that circulating cTnI concentration could increase as a result of pulmonary endothelial damage, with release of angiotensin-converting enzyme and subsequent myocardial damage.14 However, this seems unlikely in the cats of our study because feline asthma does not induce pulmonary endothelial damage to any great extent. Nevertheless, in the cats with D-NCC, potential causes of increased plasma cTnI concentrations might include myocardial hypoxemia secondary to severe pulmonary disease, increased myocardial wall stress and injury secondary to pulmonary hypertension associated with primary pulmonary disease, septic myocardial injury as a result of infectious pneumonia, or nonspecific inflammatory myocardial injury by inflammatory mediators involved in the pulmonary disease process.17 However, these are all conjectural and remain to be more clearly elucidated.
From a practical clinical perspective, the lack of readily available, rapid, and inexpensive blood cTnI analyzers for cage-side assessments may currently limit the diagnostic use of circulating cTnI concentration measurement in cats with severe respiratory distress because the turnaround time would likely be unacceptably long. Acute severe dyspnea in cats requires rapid intervention and therefore requires rapid diagnosis. Nonetheless, once such analyzers become available, our data support their use in discriminating causes of dyspnea in cats.
Finally, the diagnostic limits established in the present study apply only to the specific analyzer that we used; results obtained from different analyzers do not agree sufficiently to be interchangeable.18 Studies with new analyzers will need to be conducted to establish appropriate diagnostic cutoff values for each analyzer, but similar findings would be anticipated.
On the basis of the results of the present study, we suggest that measurement of plasma cTnI concentrations may be a useful adjunct diagnostic test in cats with dyspnea of which the cause is unknown, especially in institutions that are able to rapidly assess this variable. These preliminary data indicate that additional investigations of dyspneic cats involving more rapid cTnI analyzers (specifically in-house and cage-side models) as they become available are warranted to better define the diagnostic usefulness of this test.
ABBREVIATIONS
| CHF | Congestive heart failure |
| cTnI | Cardiac troponin I |
| D-CHF | Dyspnea attributable to congestive heart failure |
| D-NCC | Dyspnea attributable to noncardiac causes |
Stratus CS stat fluorometric analyzer, Dade Behring, Newark, Del.
References
- 1.↑
O'Brien PJ, Smith DE, Knechtel TJ, et al. Cardiac troponin I is a sensitive, specific biomarker of cardiac injury in laboratory animals. Lab Anim 2006;40:153–171.
- 2.↑
Herndon WE, Kittleson MD, Sanderson K, et al. Cardiac troponin I in feline hypertrophic cardiomyopathy. J Vet Intern Med 2002;16:558–564.
- 3.
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.
- 4.
Connolly DJ, Guitian J, Boswood A, et al. Serum troponin I levels in hyperthyroid cats before and after treatment with radioactive iodine. J Feline Med Surg 2005;7:289–300.
- 5.
Connolly DJ, Cannata J, Boswood A, et al. Cardiac troponin I in cats with hypertrophic cardiomyopathy. J Feline Med Surg 2003;5:209–216.
- 6.
Lobetti R, Dvir E, Pearson J. Cardiac troponins in canine babesiosis. J Vet Intern Med 2002;16:63–68.
- 7.
Oyama MA, Sisson DD. Cardiac troponin-I concentration in dogs with cardiac disease. J Vet Intern Med 2004;18:831–839.
- 8.
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.
- 9.↑
Spratt DP, Mellanby RJ, Drury N, et al. Cardiac troponin I: evaluation I of a biomarker for the diagnosis of heart disease in the dog. J Small Anim Pract 2005;46:139–145.
- 10.
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.
- 11.
Schober KE, Kirbach B, Oechtering G. Noninvasive assessment of myocardial cell injury in dogs with suspected cardiac contusion. J Vet Cardiol 1999;1:17–26.
- 12.
Selting KA, Lana SE, Ogilvie GK, et al. Cardiac troponin I in canine patients with lymphoma and osteosarcoma receiving doxorubicin: comparison with clinical heart disease in a retrospective analysis. Vet Comp Oncol 2004;2:142–156.
- 13.
Kirbach B, Schober K, Oechtering G, et al. Diagnostic of myocardial cell injury in cats with blunt thoracic trauma by circulating biochemical markers. Tierarztl Prax Ausg K Klientiere Heimtiere 2000;28:25–33.
- 14.↑
Prosek R, Sisson DD, Oyama MA, et al. Distinguishing cardiac and non-cardiac dyspnea in 48 dogs using plasma atrial natriuretic factor, B-type natriuretic factor, endothelin and cardiac troponin-I. J Vet Intern Med 2007;21:238–242.
- 15.↑
Porciello F, Rishniw M, Herndon WE, et al. Cardiac troponin I is elevated in dogs and cats with azotemic renal failure and in dogs with non-cardiac systemic disease. Aust Vet J 2008;86:in press.
- 16.
Sleeper MM, Clifford CA, Laster LL. Cardiac troponin I in the normal dog and cat. J Vet Intern Med 2001;15:501–503.
- 17.↑
Korff S, Katus HA, Giannitsis E. Differential diagnosis of elevated troponins. Heart 2006;92:987–993.
- 18.↑
Adin DB, Oyama MA, Sleeper MM, et al. Comparison of canine cardiac troponin I concentrations as determined by 3 analyzers. J Vet Intern Med 2006;20:1136–1142.
