Impact of preoperative nutritional scores on 1-year postoperative mortality in patients undergoing valvular heart surgery.

OBJECTIVE

Malnutrition is a well-recognized risk factor for poor prognosis and mortality. We investigated whether preoperative malnutrition diagnosed with objective nutritional scores affects 1-year mortality in patients undergoing valvular heart surgery.

METHODS

In this retrospective cohort observational study, we evaluated the association among the Controlling Nutritional Status score, Prognostic Nutritional Index, and Geriatric Nutritional Risk Index with 1-year mortality in 1927 patients undergoing valvular heart surgery. We identified factors for mortality using multivariable Cox proportional hazard analysis and investigated the utility of nutritional scores for risk stratification.

RESULTS

Malnutrition, as identified by a high Controlling Nutritional Status score and low Prognostic Nutritional Index and Geriatric Nutritional Risk Index, was significantly associated with higher 1-year mortality. Kaplan-Meier survival curve showed that mortality significantly increased as the severity of malnutrition increased (log-rank test, P < .001). The predicted discrimination (C-index) was 0.79 with the Controlling Nutritional Status score, 0.77 with the Prognostic Nutritional Index, and 0.73 with the Geriatric Nutritional Risk Index. Each nutritional index (Controlling Nutritional Status; hazard ratio, 1.31, 95% confidence interval, 1.21-1.42, P < .001), the European System for Cardiac Operative Risk Evaluation II (hazard ratio, 1.07, 95% confidence interval, 1.04-1.09, P < .001), and chronic kidney disease (hazard ratio, 2.26, 95% confidence interval, 1.31-3.90, P = .003) were independent risk factors for mortality. The Controlling Nutritional Status score added to the European System for Cardiac Operative Risk Evaluation II significantly increased the predictive discrimination ability for mortality (C-index 0.82, 95% confidence interval, 0.78-0.87, P = .014) compared with the Controlling Nutritional Status or European System for Cardiac Operative Risk Evaluation II alone.

CONCLUSIONS

Preoperative malnutrition as assessed by objective nutritional scores was associated with 1-year mortality after valvular heart surgery. The Controlling Nutritional Status score had the highest predictive ability and, when added to the European System for Cardiac Operative Risk Evaluation II, provided more accurate risk stratification.