• Open Access

Cardiac Troponin I in Calves with Congenital Heart Disease

Authors


Corresponding author: Kazuyuki Suzuki, DVM, PhD, 582 Midorimachi, Bunnkyoudai, Ebetsu, Hokkaido, 069-8501, Japan; e-mail: kazuyuki@rakuno.ac.jp

Abstract

Background

The association between plasma cardiac troponin I (cTnI) and the magnitude of cardiac enlargement in calves with congenital heart disease (CHD) are not well defined.

Objective

To investigate the relationship between plasma cTnI concentrations and cardiac size in healthy calves and calves with CHD.

Animals

A total of 19 healthy calves (control) and 12 Holstein calves with CHD (patent ductus arteriosus, ventricular septal defect, tetralogy of Fallot or double outlet right ventricle).

Methods

Case control study. All animals underwent a comprehensive transthoracic echocardiographic study to document cardiac health or presence of CHD. The vertebral heart score (VHS) was determined in each animal using right lateral survey radiographic images. Blood samples were collected via jugular venipuncture and plasma cTnI concentration and creatine kinase (CK) activity were determined by a 3rd generation immunoassay and an automatic biochemical analyzer, respectively. Groups were compared using Mann-Whitney U-test and receiver-operating characteristics (ROC) curve analysis.

Results

Calves with CHD had significantly larger VHS values and higher plasma cTnI concentrations (< .001) compared to control. Creatine kinase activity was not different between the control and CHD groups of calves. Diagnostic cutoffs of VHS and plasma cTnI for discrimination of groups were 8.9 vertebrae and 0.035 ng/mL, respectively. The cTnI concentration in plasma was significantly correlated with VHS (r 2=0.512, < .001).

Conclusion and Clinical Relevance

Our results suggest that determination of plasma cTnI concentrations in calves with clinical signs compatible with CHD might prove useful as a guide to quantify cardiac remodeling associated with increased cardiac size.

Abbreviations
CHD

congenital heart disease

CK

creatine kinase

DORV

double outlet right ventricle

cTnI

cardiac troponin-I

PDA

patent ductus arteriosus

Se

sensitivity

Sp

specificity

TOF

tetralogy of Fallot

VHS

vertebral heart score

VSD

ventricular septal defect

Introduction

The prevalence of congenital heart disease (CHD) in cattle has been reported to be approximately 0.7%,[1] with ventricular septal defects (VSD) representing the most common type of CHD.[2] Sudden death, stunted growth, and poor breeding performance are frequently observed as complications of CHD in cattle.[3-5] Early identification of CHD and assessment of myocardial involvement might be important for prognostication, risk stratification, and treatment of calves with CHD.

Echocardiography is considered the “gold standard” in the diagnosis of heart disease in many species. In calves, echocardiography has proven to be diagnostically useful in the detection and quantification of VSDs and other types of CHD.[3] In addition, thoracic radiography is used in the prediction of disease severity using morphometric indices.[6] The vertebral heart score (VHS) is commonly used in the evaluation of cardiac size in dogs and cats with heart disease.[6, 7] VHS measurements such as assessment of cardiac enlargement tend to increase in dogs with cardiac disease, because cardiac disease leads to hypertrophy or cardiac dilation, both of which increase the external cardiac dimensions.[7] Determination of the VHS is a method of cardiac size measurement that compares the radiographic dimensions of the cardiac silhouette with the length of thoracic vertebral bodies.[7] However, the main limitation of echocardiography and thoracic radiography is that they require dedicated equipment that are relatively expensive.[4] Therefore, both echocardiographic and radiographic examinations are very useful diagnostic tests, but are not widely available in routine clinical practice. In most clinical cases, they are impractical in the absence of referral options.

Various circulating proteins are used as cardiac biomarkers in the diagnosis and potential prognosis of bovine heart disease.[3, 5] Although the myocardial isoenzyme MB of creatine kinase (CK),[5] lactic dehydrogenase type 1,[5] and atrial natriuretic peptide[4] are all increased in heart disease, these biomarkers lack specificity in the diagnosis of myocardial injury in cattle.[3, 8] Cardiac troponin-I (cTnI) is a myofibrillar protein required for myocardial contraction and relaxation, and serum concentrations increase in response to myocardial injury,[9] which is a valid biomarker of myocardial injury in many mammalian species. cTnI has not been fully evaluated in cardiac disease in cattle.[10, 11] Increased concentrations of circulating cTnI have been reported from cows with endocarditis,[3] traumatic reticuloperitonitis,[12] pericardial disease,[13] calves with foot and mouth disease,[10, 14] myocarditis,[8] and have been validated histologically in cattle with experimentally induced myocardial damage.[15] Severity of myocardial injury and blood concentrations of cTnI are correlated.[8] Therefore, the purpose of the study reported herein was to determine plasma cTnI concentrations in calves with CHD and to compare plasma cTnI with radiographic indices of cardiac enlargement.

Materials and Methods

All procedures were performed in accordance with the Guide for the Care and Use of Laboratory Animals of the School of Veterinary Medicine at Rakuno Gakuen University and the National Research Council.[16] Twelve calves with CHD aged (mean ± SD) 20 ± 12 days old and 43 ± 13 kg of body weight were included. All animals had heart murmurs upon auscultation. The definitive diagnosis of CHD was made in each animal by transthoracic echocardiography with two-dimensional and B-mode images acquired from right parasternal short-axis and long-axis views1 using a 5.0 MHz sector transducer.2 A diagnosis of patent ductus arteriosus (PDA) and VSD was made by direct visualization of the defects in short and long axis views. Diagnostic criteria for tetralogy of Fallot (TOF) included presence of a large VSD, severe pulmonic stenosis, concentric right ventricular hypertrophy, and an overriding, anteriorly displaced aortic root by echocardiography. A double outlet right ventricle (DORV) was diagnosed using echocardiograph if more than 90% of the two great vessels arose from the morphologic right ventricle.

The control group comprised 19 normal calves, 25 ± 10 days old and 45 ± 13 kg of body weight. The health status of the control animals was determined on the basis of a physical examination, serum biochemical analyses, echocardiography, and thoracic radiography.

Two sets of thoracic radiographic images with the animals in right lateral recumbency were obtained in each calf using a portable X-ray machine3 and digital computed radiography system,4 in nonsedated calves. Duplicate lateral projections from each calf were used for VHS measurements, measured separately, and values averaged for final statistical analysis.[6, 7] The long axis of the heart was measured from the heart base to the apex, using the ventral margin of the carina and the left mainstem bronchus as dorsal landmarks. The short axis of the heart was measured perpendicular to the long axis, from cranial to caudal border at the widest portion of the heart, approximately at the level of the ventral border of the caudal vena cava. Individual measurements were then transposed to the long axial length of thoracic spine, beginning at the cranial margin of T4 and extended caudally. Measurements were translated into a vertebral number to the nearest 0.1 vertebral units (v), with a single vertebral unit consisting of a vertebral body and the accompanying caudal intervertebral disk space. Vertebral numbers were then summed for the total VHS. All VHS measurements were made by the same examiner (KS) blinded to results of the echocardiographic examination at the time of analysis, in accordance with methods described by Buchanan and Bucheler.[17]

A volume of 4 mL of whole blood was collected via jugular venipuncture into heparinized tubes for CK and cTnI analysis and centrifuged at room temperature within 15 minutes of collection. The plasma obtained was separated and stored at −80°C until assay. Plasma CK activity was measured using a kinetic method5 and an automatic biochemical analyzer,6 in accordance with the manufacturer's instructions. Plasma cTnI concentration was measured using a human cTnI 1-step sandwich EIA Kit (3rd generation)7 and an automated immunoassay instrument,8 and conducted in accordance with the manufacturer's instructions. The lower limit of detection for this assay is 0.01 ng/mL.

Statistical Analysis

Statistical analyses were performed using a commercial software package.9 Normally distributed data are reported as mean ± standard deviation (SD) and non-normally distributed data are expressed as median and range. For non-normally distributed data such as VHS, CK, and cTnI, the Mann-Whitney U-test was employed for comparison between groups. Receiver-operating characteristic (ROC) curves were constructed and used to determine diagnostic cutoffs, sensitivity (Se), and specificity (Sp) of each parameter in the diagnosis of CHD. The optimal diagnostic cutoff point was calculated via the Youden index.[18] The Youden index (J) is defined as the maximum vertical distance between the ROC curve and the diagonal or chance line and is calculated as = maximum [Se + Sp − 1]. The point on the ROC curve that corresponds to J is then used as the optimal cutoff point.[18] The relationship between the VHS and cTnI concentrations was evaluated by the Pearson's product-moment correlation coefficient. Plasma cTnI concentrations below 0.01 ng/mL were statistically analyzed as 0.01 ng/mL.[19] Significance was assumed when < .05.

Results

Calves with CHD were diagnosed as having patent ductus arteriosus (n = 2), VSD (n = 4), tetralogy of Fallot (n = 3) or a double outlet right ventricle (n = 3) based on echocardiographic image evaluation in each calf. In healthy calves, plasma cTnI was below the limit of detection in 13/19 calves (68.4%) with a median plasma cTnI concentration of 0.010 ng/mL (min to max, 0.010–0.050 ng/mL). The median VHS and plasma CK values in control calves were 8.50 v (min to max, 8.0–9.0 v) and 118 ± 54 IU/L (min to max, 54–409 IU/L), respectively. The median VHS (11.00 v; min to max, 8.0 to 12.5 v, < .01) and plasma cTnI concentration (0.055 ng/mL; min to max 0.030 to 0.170 ng/mL, < .001) were significantly higher in calves with CHD compared to control (Fig 1) whereas there was no significant difference (> .05) in plasma CK concentrations between groups. ROC curves for VHS and plasma cTnI concentrations are presented in Fig 2. Table 1 summarizes various cutoff points for VHS and plasma cTnI in the discrimination of calves with CHD from healthy calves considering 100% Se, 100% Sp or the optimal combined cutoff (J). The proposed optimal cutoff point for VHS with regard to Se and Sp (J) was determined to be 8.9 v (AUC = 0.831, < .01, Se 75.0%, and Sp 94.7%). In the same manner, the proposed optimal cutoff point for plasma cTnI was determined to be 0.035 ng/mL (AUC = 0.921, < .001, Se 83.3%, and Sp 78.9%). Figure 3 graphically depicts the relationship between VHS and plasma cTnI concentration observed in control and disease animals. Plasma cTnI concentrations were significantly and positively correlated to VHS (r 2 = 0.512, < .001).

Figure 1.

Medians of vertebral heart score (VHS) and plasma cardiac troponin-I (cTnI) concentrations in the calves with congenital heart disease. The horizontal line in each box represents the median value. The boxes represent the interquartile range (25–75 percentiles). Outliers are plotted separately as dots. **Statistical difference (< .001) between groups (Mann-Whitney U-test).

Figure 2.

Receiver-operating characteristics (ROC) curves for vertebral heart score (VHS; left panel) and cardiac troponin-I concentration (cTnI; right panel) in discriminating calves with congenital heart disease (CHD) from healthy calves. The area under the ROC curve (AUC) is shown for each ROC curve. The optimal cutoff point (open circle) for the test was calculated using the Youden index.

Figure 3.

Relationship between vertebral heart score (VHS) to plasma cardiac troponin-I (cTnI) concentration in calves with congenital heart disease (CHD) and healthy calves. Plasma cTnI concentrations were significantly and positively correlated to VHS (y = 0.0229− 0.1755, r2 = 0.512, < .001) by Pearson's product-moment correlation coefficient. Open circles; healthy calves; closed circles; CHD; the central line represents the regression line; the dashed lines represent the 95% confidence interval; vertical (8.9 v) and horizontal lines (0.035 ng/mL) indicate the best cutoff values for VHS and cTnI, respectively.

Table 1. The cutoff points for the vertebral heart score (VHS) and plasma cardiac troponin-I (cTnI) concentrations in calves with congenital heart disease
 100% SensitivityYouden Index[17]100% Specificity
VHS (v)7.08.99.4
cTnI (ng/mL)0.0150.0350.055

Discussion

The purpose of this study was to determine plasma cTnI concentrations in calves with or without CHD and to compare those with radiographic indices of cardiac size. Calves with CHD had significantly increased VHS as well as plasma cTnI concentrations compared to healthy animals, whereas plasma CK did not differ between groups. The diagnostic cutoffs identified for VHS and plasma cTnI were 8.9 v and 0.035 ng/mL, respectively. Plasma cTnI concentrations were significantly and positively correlated to VHS. Our results suggest that determining plasma cTnI concentrations in calves with clinical signs compatible with CHD may prove useful as a guide to quantify cardiac remodeling associated with increased cardiac size.

Cardiac size can be determined from lateral thoracic radiographs using the VHS system, as proposed by Buchanan and Bucheler.[17] In this study, the median VHS in calves without cardiac abnormality was 8.50 v with a range between 8.0 and 9.0 v. Calves had a mean VHS lower than that of normal dogs (9.7 ± 0.5 v),[17] but higher than VHS of cats (7.5 ± 0.03 v).[20] Cardiomegaly can be determined from thoracic radiographs and objectively measured using VHS determination. Cardiomegaly or eccentric cardiac remodeling is correlated to heart failure class and may, therefore, be used as an index of disease severity. Our results demonstrate that calves with CHD are characterized by significantly larger hearts; and the proposed diagnostic cutoff for VHS in detecting CHD, to be interpreted in concert with historical and clinical findings, was found to be 8.9 v. These findings are in agreement with studies in dogs[7, 17] and cats,[20] where VHS was found to be increased with cardiac disease and correlated to heart failure class. In this study, we demonstrated that cTnI was significantly higher in calves with CHD compared to control calves. However, there was no significant difference in plasma CK concentrations between groups. Although CK, particularly its more cardiac-specific isoenzyme MB-CK, has been used for the detection of heart disease in animals and people, the diagnostic accuracy is hampered by the lack of tissue specificity.[3] Currently, the preferred cardiac biomarker for detection of myocardial damage is cTnI with nearly absolute myocardial specificity, ease of determination, and clinically useful release-excretion kinetics as compared to CK, myoglobin, and lactate dehydrogenase.[8] The molecular structure of cTnI is highly conserved across mammalian species and assays developed for its measurement in human patients have been validated in a number of studies in dogs,[9, 19, 21, 22] cats,[23] horses,[24, 25] and cattle.[3, 8, 10, 12-14, 26] The cTnI assays used in people can also be used in cattle, as cTnI in both species is structurally highly homologous.[10] Bovine cTnI has >96% amino acid sequence homology to human cTnI,[10, 27] suggesting that the bovine protein can be reliably detected and quantified in samples obtained from cattle.[26]

In this study, we used a 3rd generation 1-step sandwich EIA Kit to measure plasma cTnI concentrations. Plasma cTnI concentrations for the 19 clinically healthy calves described herein were consistent with values previously reported for healthy calves,[8, 11] mature cattle,[13, 26] horses,[24, 25] and dogs.[22] Recent veterinary publications report increased cTnI concentrations in dogs[21] and cattle[26] with CHD compared to normal animals. Our findings are in agreement with these previous studies.

Although calves with CHD may be completely asymptomatic for some time with only a heart murmur present, they typically develop clinical signs early in life and have a guarded to poor long-term prognosis. Once cardiac dysfunction from chronic pressure and volume overload has developed, various clinical manifestations, (stunted growth, venous pulsations, liver enlargement, ventral edema) are observed.[28] However, causes other than CHD for the presence of cardiac murmurs in calves exist. Innocent murmurs caused by turbulent blood flow are also common in neonate and young calves without anatomic obstructions. Occasionally conditions such as anemia, excitement, and fever occur that may cause turbulent flow sufficient to cause audible vibrations of the adjacent cardio-hemic structures.[29] Therefore, it is important to differentiate true anatomic cardiac lesions from physiologic functional or innocent murmurs. However, without additional diagnostic procedures such as chest radiography or echocardiography, assessment of disease relevance is challenging in clinical practice. Therefore, an easily performed, noninvasive and predictive test in combination with thorough physical examination detecting cardiovascular disease would be useful in prognostication and clinical decision making. Under such circumstances, assessment of plasma cTnI in calves with suspected CHD based on findings from physical examination (eg, a heart murmur) would be useful. Our results suggest that using a cutoff value of ≥0.035 ng/mL, cTnI in plasma may distinguish calves with and without significant CHD with reasonable Se and Sp.

Several previous reports demonstrated a positive relationship between circulating cTnI concentrations and severity of heart disease.[15, 23, 30] A number of studies have been performed in dogs showing very good correlation between histopathologically detectable myocyte damage and blood cTnI concentrations.[30, 31] In addition, cTnI was significantly associated with left ventricular shortening fraction and myocardial histopathologic lesion score in cattle with monensin-induced cardiomyocyte damage.[15] The major finding of this study is that plasma cTnI concentrations correlate with the severity of cardiac enlargements in calves with CHD.

In conclusion, plasma cTnI concentrations provide a sensitive and specific test for distinguishing calves with CHD from healthy calves and in assessing of the severity of cardiac remodeling as assessed by cardiac size.

Acknowledgment

This study was supported by a grant-in-aid for Science Research from the Ministry of Education, Culture and Sciences of Japan (no.21580393) awarded to K. Suzuki.

Footnotes

  1. 1

    Prosound SSD-5000SV, Aloka Co, Tokyo, Japan

  2. 2

    UST-5297, Aloka Co

  3. 3

    PX-10HF, Medison Acoma Co, Ltd, Tokyo, Japan

  4. 4

    FCR XG-1V, Fujifilm Co., Tokyo, Japan

  5. 5

    Creatine Kinase (CK) Kinetic Method for CX and LX system, Teco Diagnostics, Anaheim, CA

  6. 6

    Synchron CX4 delta, Beckman Coulter Inc, Anaheim, CA

  7. 7

    ST AIA-PACK cTnI 3rd-Gen, Tosoh Medics, Kanagawa, Japan

  8. 8

    Tosoh AIA-360 Tosho Medics

  9. 9

    IBM SPSS Statistics, Ver.19, IBM Co, Somers, NY

Ancillary