Heart valve surgery in hemodialysis-dependent patients: nutrition status impact on surgical outcome.

Abstract

Valve surgery in hemodialysis-dependent patients is associated with postoperative complications and a high mortality rate, and such patients frequently suffer cachexia. This study aimed to determine pre- and intraoperative risk factors associated with in-hospital mortality and long-term survival in hemodialysis-dependent patients undergoing heart valve surgery from the viewpoint of nutrition status. Eighty-seven hemodialysis-dependent patients who underwent valve surgery between January 1998 and October 2015 were retrospectively reviewed. Thirty-seven potential perioperative risk factors were evaluated. The in-hospital mortality rate was 12.6% (11 patients). Univariate analysis identified New York Heart Association Functional Classification III or IV, emaciation (body mass index<17.6kg/m(2)), total cholesterol<120mg/dl, serum albumin<3.0mg/dl, emergent/urgent surgery, and intraoperative blood transfusion>3000ml as predictors of in-hospital death. Multivariate logistic regression analysis confirmed low serum albumin<3.0mg/dl (hazard ratio 7.22; p=0.032) and emergent/urgent operation (hazard ratio 43.57; p=0.035) as independent predictors of in-hospital death. The 1- and 3-year actuarial survival rates were 64.9±5.4 and 51.8±5.8%, respectively. Long-term survival estimated by log-rank test was negatively impacted by anemia (hemoglobin<10mg/dl), low serum albumin, emergent/urgent operation, and infective endocarditis. Multivariate analysis using Cox proportional hazards modeling indicated low serum albumin (hazard ratio 2.12; p=0.047) and emergent/urgent operation (hazard ratio 8.97; p=0.0002) as independent predictors of remote death. Hypoalbuminemia and emergent/urgent operation are strong predictors of in-hospital and remote death. Malnutrition before surgery should be considered for operative risk estimation, and adequate preoperative nutrition management may improve surgical outcomes for hemodialysis-dependent patients.