Brain natriuretic peptide as a biomarker of systemic right ventricular function in patients with transposition of great arteries after atrial switch operation

Abstract

Background

Assessment of systemic right ventricular (RV) function is inherently difficult. In adults, plasma brain natriuretic peptide (BNP) level has been shown to reflect systemic ventricular dysfunction. We sought to test the hypothesis that plasma BNP is a biomarker of systemic RV function in patients after atrial switch operation.

Methods

We determined the RV function in 44 patients, 35 after Senning and 9 after Mustard operation, aged 19.7 ± 4.0 years, by tissue Doppler echocardiography and determination of myocardial performance index (MPI). The results were compared to the left ventricular function of 14 age-matched controls. Their plasma BNP levels were correlated with indices of systemic ventricular function.

Results

Compared with controls, the patients had greater MPI (p < 0.001), lower systemic ventricular free wall-annular early diastolic (p < 0.001), late diastolic (p < 0.001), and systolic velocities (p = 0.001), lower septal-annular early diastolic (p < 0.001), late diastolic (p < 0.001), and systolic velocities (p < 0.001), and higher BNP levels (p = 0.03). Plasma BNP levels correlated positively with MPI (r = 0.43, p = 0.001) and negatively with the free wall- and septal-annular myocardial velocities (r = − 0.32 to − 0.47, p < 0.05). The area under the receiver operating characteristic curve for BNP to detect ventricular dysfunction (MPI > 0.45) in patients was 0.67 (p = 0.04). A BNP level of 36 pg/ml had a sensitivity of 55%, specificity of 86%, positive predictive value 80%, negative predictive value of 64%, and an accuracy of 70% for detecting systemic ventricular dysfunction.

Conclusions

Plasma BNP has modest accuracy in the detection of systemic RV dysfunction in patients after atrial switch operation.

Introduction

Systemic right ventricular (RV) dysfunction after atrial switch operation in patients with complete transposition of the great arteries (TGA) has been reported to occur at a prevalence ranging from 9–45%, being severe in 1–14% of patients [1], [2], [3], [4], [5], [6]. Management options for systemic RV failure include heart transplantation, arterial switch operation with takedown of atrial switch following retraining of the left ventricle by pulmonary artery banding [7], [8], and the recently reported cardiac resynchronization therapy [9]. The importance of serial assessment of systemic RV function in the long-term follow-up of patients after the Senning or Mustard procedure cannot therefore be overemphasized.

Quantitative two- or three-dimensional echocardiographic assessment of systemic RV function is nonetheless inherently difficult due to the complex geometry of the morphologic right ventricle [10], [11], [12]. On the other hand, the Doppler-derived myocardial performance index (MPI) [13], a global measure of combined systolic and diastolic function, has been shown to correlate with systemic RV ejection fraction as derived from cardiac magnetic resonance imaging [14], and peak oxygen consumption [15] in patients after atrial switch operation.

Assay of plasma biomarkers that reflect ventricular function and cardiac functional status is perhaps an alternative [16], [17], [18]. Plasma brain natriuretic peptide (BNP), in particular, may have a role in the evaluation of systemic RV function as this neurohormone is released from the ventricle in response to volume and pressure load [19]. It has been shown to be useful in the diagnosis, prognostication and refinement of management in adults with heart failure [16], [17], [18], [20]. While previous studies have demonstrated elevated BNP level in ventricular dysfunction complicating different types of congenital heart disease [21], [22], [23], [24], [25], [26], the role of plasma BNP in the assessment of systemic RV function after atrial switch operation has not been examined. The present study aimed to test the hypothesis that plasma BNP is a biomarker of systemic RV function in patients with complete TGA after atrial switch operation.