Abstract

The aim of the present study was to retrospectively investigate the prognostic value of N-terminal pro-brain natriuretic peptide (NT-proBNP) in tricuspid valve replacement (TVR). A total of 73 TVR patients who had NT-proBNP measured on the first postoperative morning during a period of 10 years from February 2008 to December 2018 were included in the study. The endpoint was postsurgery all-cause in-hospital mortality. The outcome-based cut-point optimization was performed using X-tile software. NT-proBNP with the maximum χ2 score and the minimum P value will be used as the optimal cut-point. Kaplan-Meier analysis and log-rank test were adopted to calculate and compare survival rates stratified by tertiles and the cut-point. Predictive capabilities of NT-proBNP were tested using univariable and multivariable Cox regression. Overall, 20 (27.3%) in-hospital deaths occurred. Postsurgery hospital stay was 21 days (interquartile range, 16-32 day). NT-proBNP were divided into low (<1262 pg/mL), medium (1262-4003 pg/mL), and high (≥4003 pg/mL) tertiles. The optimal cut-off point determined using X-tile was 3639 pg/mL. Kaplan-Meier analysis revealed a strong association between worse survival and elevated NT-proBNP expressed as tertiles (log-rank P = 0.002) and stratified by optimal cut point (log-rank P < 0.001). Multivariable Cox survival analysis demonstrated that NT-proBNP was a strong predictor of mortality (logNT-proBNP hazard ratio [HR], 2.11; 95% confidence interval [CI], 1.33-3.37; P = 0.002). In NT-proBNP tertiles model, multivariable Cox survival analysis showed that patients in the medium and high NT-proBNP tertiles had 6.32-fold (adjusted HR, 7.32; 95% CI, 0.76-70.69; P = 0.085) and 16.11-fold (adjusted HR, 17.11; 95% CI, 1.92-152.68, P = 0.011) increased risk for mortality, respectively, compared with patients in the low tertile. Elevated postoperative NT-proBNP level is a potential independent and strong in-hospital postsurgery mortality risk factor in TVR, thus may serve as a useful surrogate marker for risk-stratification.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call