Full title: The prevalence of and predictors for perioperative hypothermia in postanaesthesia care unit.

Abstract

Hypothermia is a common clinical issue during the perioperative period. The patients with perioperative inadvertent hypothermia are associated with higher risk of postoperative complications and higher hospitalisation costs. The aim of this study is to explore the prevalence of and predictors for hypothermia in patients entering postanaesthesia care unit. Retrospective cohort study. A rigorous retrospective cohort study was conducted according to the STROBE reporting checklist. A total of 7216 patients were enrolled in this study. The perioperative variables potentially related to hypothermia were collected. Univariate analysis and multivariate logistic regression analysis were performed to investigate the contributing factors. The overall prevalence of inadvertent hypothermia in postanaesthesia care unit was 21.3% (n=1505). The multivariate logistic regression analysis showed that age >65 (OR=1.561, 95% CI 1.371-1.778, p<.001), non-supine position [lateral decubitus position (OR=1.341, 95% CI 1.133-1.586, p=.001), lithotomy position (OR=1.639, 95% CI 1.295-2.075, p<.001)], non-superficial surgery (OR=2.195, 95% CI 1.566-3.077, p<.001), non-open surgery [laparoscopic surgery (OR=1.205, 95% CI 1.020-1.423, p=.029), endoscopic surgery (OR=1.430, 95%CI 1.084-1.887, p=.011)], the volume of intravenous infusion fluid >1000ml (OR=1.814, 95% CI 1.500-2.194, p<.001), blood transfusion (OR=1.552, 95% CI 1.159-2.078, p=.003), operation performed in the summer or fall (OR=1.874, 95%CI 1.656-2.122, p<.001) and use of dexmedetomidine (OR=1.147, 95%CI 1.015-1.296, p=.028) were associated with increased risk of hypothermia. In contrast, our finding showed that body mass index ≥25kg/m2 (OR=0.556, 95%CI 0.491-0.630, p<.001), higher baseline body temperature (OR=0.641, 95%CI 0.541-0.761, p<.001) and duration of fasting ≥18h (OR=0.487, 95%CI 0.345-0.689, p<.001) were associated with decreased risk of hypothermia. Compared with non-hypothermic patients, patients with hypothermia were associated with prolonged length of hospital days (5 vs. 4days, p<.001). For patients admitted to postanaesthesia care unit after elective procedures, old age, non-supine position, non-open surgery, non-superficial surgery, large volume of intravenous infusion fluid, dexmedetomidine, blood transfusion and Summer or Fall operative season were associated with increased risk of hypothermia, whereas high body mass index, high baseline body temperature and long fasting duration were associated with decreased risk of hypothermia. The outcomes of this study will raise the concerns of perioperative care team on hypothermia in surgical patients. Measures should be taken to improve perioperative hypothermia and clinical outcome.