Introduction: Blood biomarkers have the potential to monitor the severity of chronic heart failure (CHF). Studies correlating repeated measurements of blood biomarkers with repeatedly assessed New York Heart Association (NYHA) class over a prolonged follow-up period, and concomitantly investigating their associations with clinical endpoints, have not yet been performed.
Methods: Between 2011-2013, 263 CHF patients were included. At inclusion and subsequently every 3 months, we measured N‑terminal pro-B-type natriuretic (NT-proBNP), high-sensitivity troponin T (Hs-TnT) and C‑reactive protein (CRP), and assessed NYHA class. The primary endpoint comprised heart failure hospitalisation, cardiovascular mortality, cardiac transplantation or left ventricular assist device implantation. Time-dependent Cox models were used.
Results: Mean age was 67 ± 13 years, 72% were men and 27% were in NYHA class III-IV. We obtained 886 repeated measures (median 3 [IQR 2-5] per patient). The primary endpoint was reached in 41 patients during a median follow-up of 1.0 [0.6-1.4] year. Repeatedly measured NT-proBNP and Hs-TnT were significantly associated with repeatedly assessed NYHA class, whereas CRP was not (NT-proBNP: β [95% CI]: 1.56 [1.17-2.06]ln(ng/l) increase per point increase in NYHA class, p = 0.002; HsTNT: β [95% CI]: 1.58 [1.21-2.07]). Serially measured NT-proBNP (HR [95% CI]:2.86 [1.73-4.73]), CRP (1.69 [1.21-2.34]) and NYHA class (2.33 [1.51-3.62]) were positively and independently associated with the primary endpoint, whereas Hs-TnT lost statistical significance after multivariable adjustment. A model containing serially measured NYHA class and NT-proBNP displayed a C-index of 0.84, while serially measured NYHA class and CRP showed a C-index of 0.82.
Conclusion: Temporal NT-proBNP, CRP and NYHA class patterns are independently associated with adverse clinical outcome. Serially measured NT-proBNP and NYHA class are best suited for monitoring CHF outpatients.
Keywords: Biomarkers; Heart failure; NYHA class; Repeated measurements.