Predicting length of stay out of hospital following lung resection using preoperative health status measures.

Abstract

Studies evaluating predictors of operative outcome for lung resection have focused on physiological measures of cardiorespiratory impairment, but these have proved inadequate. This study evaluated the predictive abilities of six preoperative variables: the global quality of life (QL), social function (SF), and emotional function (EF) scales of the European Organization for the Research and Treatment of Cancer's (EORTC) QLQ-C30 questionnaire, 6-min walk distance (6MWD), forced expiratory volume (FEV1), and diffusion capacity (DLCO). Operative outcome was represented by the surrogate measure length of stay, out of hospital within the first 30 days (LOSOH). A single-centre prospective cohort study evaluating 70 subjects was conducted using multiple regression. LOSOH was bimodally distributed, therefore analysis was undertaken for the entire sample and for two separate groups (A and B). Group B (n = 4) experienced severe complications (LOSOH = 0-5 days) and was too small for statistical analysis. Group A (n = 66) suffered fewer and less severe complications (LOSOH = 14-26 days). For the entire sample, age was the sole predictor of LOSOH (R2 = 0.123, p = 0.003). In Group A, the strongest predictors of LOSOH were global QL score and 6MWD (R2 = 0.224, p < 0.001). LOSOH was inversely correlated with complications. While it remains difficult to predict severe complications in this population, within Group A, health status measures demonstrated a limited ability to predict LOSOH.