Methods. One thousand and ten patients undergoing non-emergent cardiac surgery were recruited prospectively. Baseline clinical details were obtained and the European System for Cardiac Operative Risk Evaluation (EuroSCORE) and Parsonnet score were calculated. Preoperative NT-proBNP levels were measured using the Roche Elecsys assay. The primary endpoint was 30 day mortality.
Results. Median NT-proBNP levels were 624 ng litre(-1) among patients who died within 30 days of surgery (n=29), compared with 279 ng litre(-1) in survivors [odds ratio (OR) 1.03 per 250 ng litre(-1), 95% confidence interval 1.01-1.05, P=0.001). NT-proBNP levels remained predictors of 30 day mortality in models including either the additive EuroSCORE (OR 1.03 per 250 ng litre(-1), P=0.01), the logistic EuroSCORE (OR 1.03 per 250 ng litre(-1), P=0.004), or the Parsonnet score (OR 1.02 per 250 ng litre(-1), P=0.04). Levels of NT-proBNP were also predictors of prolonged (>1 day) stay in the intensive care unit (OR 1.03 per 250 ng litre(-1), P<0.001) and of a hospital stay >1 week (OR 1.07 per 250 ng litre(-1), P<0.001). They remained predictive of these outcomes in regression models that included either the EuroSCORE or the Parsonnet score and in a model that included all study variables.
Conclusions. NT-proBNP levels predict early outcome after cardiac surgery. Their prognostic utility is modest-but is independent of traditional indicators and conventional risk prediction scores.
- complications, morbidity
- hormones, atrial natriuretic peptide
- surgery, cardiovascular
- major non-cardiac surgery