RESUMO
BACKGROUND: NT-proBNP is being used as a biomarker for prognosticating and delineating cardiac dysfunction. The cut-off value for deciding normal versus abnormal levels has always been a point of contention since it depends on the degree of dysfunction as well as other associated conditions often termed non-cardiac factors and parameters. Such association had not been formally presented. OBJECTIVE: To determine the direction and magnitude of effect of cardiac and non-cardiac parameters on NT-proBNP variability. MATERIAL AND METHOD: The present study included 78 cardiac ambulatory patients with a history of heart failure and/or low left ventricular ejection fraction. Their cardiac and non-cardiac parameters were recorded at the time of blood sampling for NT-proBNP. Multivariate linear regression analysis was used to correlate cardiac and non-cardiac parameters with NT-proBNP level and, from this, a predictive equation was derived. RESULTS: Log [NT-proBNP (pmol/l)] was 1.424 + 0.348 (for EF of 18-27) + 0.636 (for EF < 18) + 0.021 CTR - 0.002 SMW- 0.326 for female + 0.430 Cr - 0.010 BW [EF = LV ejectionfraction in %; CTR = cardio-thoracic ratio in %; SMW = 6-minute walking distance in meters; Cr = serum creatinine in mg/dl; BW = body weight in kg]. The adjusted R-square for this regression was 0.659. Omitting the non-cardiac variables (sex, Cr, BW) would decrease the adjusted R-square to 0.493. CONCLUSION: Cut-off value for NT-proBNP concentration in subjects without severe systolic heart failure has to account for these non-cardiac factors.