What Role Should B-Type Natriuretic Peptide Play in Predicting Late Graft Dysfunction Following Cardiac Transplantation?

Abstract

Introduction: B-type natriuretic peptide (BNP) levels are a marker of left ventricular dysfunction and a powerful predictor of morbidity and mortality in patients with chronic heart failure. There have been limited reports indicating that BNP levels tend to be higher around the time of late graft dysfunction (LGD) following cardiac transplantation.1Bader F.M. Rogers R.K. Kfoury A.G. Gilbert E.M. Horne B.D. Stehlik J. et al.Time-dependent changes in B-type natriuretic peptide after heart transplantation: correlation with allograft rejection and function.Congest Heart Fail. 2009; 15: 63-67Crossref PubMed Scopus (11) Google Scholar, 2Martínez-Dolz L. Almenar L. Hervás I. Moro J. Agüero J. Sánchez-Lázaro I. et al.Prognostic relationship between two serial determinations of B-type natriuretic peptide and medium-long-term events in heart transplantation.JHLT. 2008; 27: 735-740Abstract Full Text Full Text PDF PubMed Scopus (8) Google Scholar, 3Damodaran A. Dardas T. Wu A.H. Dyke D.B. Hummel S.L. Cowger J.A. et al.Changes in serial B-type natriuretic peptide level independently predict cardiac allograft rejection.JHLT. 2012; 31: 708-714Abstract Full Text Full Text PDF PubMed Scopus (15) Google Scholar Objective: To assess the utility of serial BNP testing as a predictive indicator for LGD. Methods: We retrospectively reviewed heart transplantations performed at our center between 2012 and 2015, and compared the time course of BNP levels in the group of patients with LGD (defined as ejection fraction ≤50% and/or cardiac index <2) to that in the group without LGD. The predictive ability of BNP testing at different time points was assessed by receiver-operator characteristic (ROC) analysis. Multivariate logistic regression was used to assess the association between BNP and LGD. Results: Of 203 patients included, 73 (36%) patients (median age 60 years [50, 65]; 67% male) developed LGD. In both groups the average BNP was elevated at time of transplant and then decreased over time (P < .001). Throughout follow-up, BNP was greater in the LGD group than in the non-LGD group (Fig. 1A; with significant differences in the first 3 months and after 1 year; P < .05). ROC analysis showed that BNP at the time of LGD was more strongly associated with LGD than BNP at transplantation (Fig. 1B). The LGD group exhibited greater rates of biopsy-proven rejection (P < .0001) and poorer 1-year survival (P < .001). Multivariate analysis revealed that BNP at time of LGD (P < .0001) and ≥2R rejection (P = .043) were both associated with LGD status. Conclusions: These data showed elevated BNP levels surrounding the time of documented LGD after cardiac transplantation. Thus, our results add to the current literature supporting serial BNP testing as a prognostic tool for the occurrence of LGD in this patient population. Prospective studies are needed to validate these findings.