Enfermedades cardiovasculares y diagnóstico

Troponin and Nt-pro-BNP Fluctuation During Hemodialysis and Impact on Cardiovascular Prognosis

Abstract

Patrick Prud�??homme*, Hajar El-Kamouni, Gabriel Fortin, Anne-Marie Côté3, Mélanie Godin, Paul Farand, Serge Lepage, Thao Huynh and Michel Nguyen

Introduction: Chronic elevation of high sensitivity troponin (hs-cTnT) and N-terminal brain-natriuretic peptide (Nt-pro-BNP) in end-stage renal disease is associated with worse cardiovascular outcomes. Little is known about how hemodialysis (HD) parameters acutely influence those biomarkers and prognosis.

Objective: To determine if the variation of those biomarkers on the short term is influenced by factors that are already associated with hemodialysisinduced cardiac injury and if it has a prognostic value.

Methods: Forty participants undergoing HD at our institution were enrolled. Factors known to influence hs-cTnT and Nt-pro-BNP levels were the principal exclusion criteria. Eight patients were excluded mainly for EF<40%. Six consecutive serum samples were analysed for hs-cTnT and Nt-pro-BNP before and after three HD sessions. Hs-cTnT and Nt-pro-BNP percentage variation after HD were analysed in respect with HD weight reduction percentage, ultrafiltration rate, Kt/V (HD adequacy parameter), presence of diastolic dysfunction, indexed left ventricular mass, peridialysis hypotension, blood filtration rate and with the other biomarker. Linear regression analysis was used in a fixed-effect model for multivariate assessment with variables already mentionned. Major adverse clinical events (hospitalization for heart failure, acute coronary syndrome and cardiovascular death) were recorded for a period of 21 months to analyze the sensitivity and specificity of biomarker fluctuation for predicting clinical events.

Results: Mean decrease after dialysis for hs-cTnT was 38.3% +- 3.9%, while it was 56% +- 3.5% for Nt-pro-BNP. There was a fair and significant association between variation in hs-cTnT and Nt-pro-BNP and the model’s variables (Pearson coefficient of 0.646 (p<0.001) and 0. 0.53 (p=0.001), respectively). Variables having the most important influences on the biomarkers fluctuation were ultrafiltration rate (β = -0.558, p=0.001) and interdialytic weight decrease percentage (β = 0.399, p=0.020). Smaller troponin decrease (25th percentile) showed fair sensitivity (80%) for adverse clinical events. Hs-cTnT and Nt-pro-BNP levels are diminished after HD in a manner that is reproducible for the same patient and they change in a parallel manner.

Conclusion: Hs-cTnT and Nt-pro-BNP changes are highly reproducible for the same patient and vary parallelly. While seeming contradictory, but consistent with past literature, higher total fluid removal and slower ultrafiltration rates are associated with more important biomarker decreases. Since static levels of those biomarkers correlate with mortality, adopting a slower fluid removal during hemodialysis and prolonging sessions could improve mortality over the long run. This could be assessed in future dedicated studies.