OBJECTIVES: Perioperative hypothermia is linked to adverse effects that increase morbidity and mortality. The objectives of this study were to identify the risk factors for intraoperative hypothermia and construct an instrument for identifying patients at high risk. MATERIALS AND METHODS: We studied patients of all ages who had undergone surgery. Patients were assigned to a design group or a validation group by means of a list of randomly generated numbers. Intraoperative hypothermia was defined by an tympanic temperature of 35.9 degrees C or less. A bivariate analysis of the design group identified the predictive factors and a multivariate analysis (logistic regression with backward elimination of nonsignificant variables) provided a predictive model. Risk scores were obtained for each variable by converting them to a 4-degree risk scale (abbreviated model). Predictive power was determined by calculating the area under the receiver-operator characteristic curve (AUC). RESULTS: We enrolled 264 consecutive patients; 200 were assigned to the design group and 64 to the validation group. In the design group, the AUC was 0.85 for the complete model and 0.83 for the abbreviated model. In the validation group, the AUC was 0.85 for the complete model and 0.82 for the abbreviated model. The P value was <.01 for all curves. CONCLUSION: Age, weight, approximate duration of surgery, and body and ambient temperature during induction were the included factors that predicted intraoperative hypothermia in a heterogeneous sample of surgical patients.
RCT Entities:
OBJECTIVES: Perioperative hypothermia is linked to adverse effects that increase morbidity and mortality. The objectives of this study were to identify the risk factors for intraoperative hypothermia and construct an instrument for identifying patients at high risk. MATERIALS AND METHODS: We studied patients of all ages who had undergone surgery. Patients were assigned to a design group or a validation group by means of a list of randomly generated numbers. Intraoperative hypothermia was defined by an tympanic temperature of 35.9 degrees C or less. A bivariate analysis of the design group identified the predictive factors and a multivariate analysis (logistic regression with backward elimination of nonsignificant variables) provided a predictive model. Risk scores were obtained for each variable by converting them to a 4-degree risk scale (abbreviated model). Predictive power was determined by calculating the area under the receiver-operator characteristic curve (AUC). RESULTS: We enrolled 264 consecutive patients; 200 were assigned to the design group and 64 to the validation group. In the design group, the AUC was 0.85 for the complete model and 0.83 for the abbreviated model. In the validation group, the AUC was 0.85 for the complete model and 0.82 for the abbreviated model. The P value was <.01 for all curves. CONCLUSION: Age, weight, approximate duration of surgery, and body and ambient temperature during induction were the included factors that predicted intraoperative hypothermia in a heterogeneous sample of surgical patients.