Preoperative risk factors for unexpected postoperative intensive care unit admission: A retrospective case analysis

Introduction: The purpose of this retrospective case-control study was to investigate preoperative risk factors for unexpected postoperative intensive care unit (ICU) admissions in patients undergoing non-emergent surgical procedures in a tertiary medical centre. Methods: A medical record review of adult patients undergoing elective non-cardiac and non-transplant major surgical procedures during the period of January 2011 through December 2015 in the operating rooms of a large university hospital was carried out. The primary outcome assessed was unexpected ICU admission, with mortality as a secondary outcome. Demographic data, length of hospital and ICU stay and preoperative comorbidities were also obtained as exposure variables. Propensity score matching was then employed to yield a study and control group. Results: The group of patients who met inclusion criteria in the study and the control group that did not require ICU admission were obtained, each containing 1191 patients after propensity matching. Patients with acute and/or chronic kidney injury (odds ratio (OR) 2.20 [1.75-2.76]), valvular heart disease (OR: 1.94 [1.33-2.85]), peripheral vascular disease (PVD) (OR: 1.41 [1.02-1.94]) and congestive heart failure (CHF) (OR: 1.80 [1.31-2.46]) were all associated with increased unexpected ICU admission. History of cerebrovascular accident (CVA) (OR: 3.03 [1.31-7.01]) and acute and/or chronic kidney injury (OR: 1.62 [1.12-2.35]) were associated with increased mortality in all patients; CVA was also associated with increased mortality (OR: 3.15 [1.21-8.20]) specifically in the ICU population. Conclusions: CHF, acute/chronic kidney injury, PVD and valve disease were significantly associated with increased unexpected ICU admission; patients with CVA suffered increased mortality when admitted to the ICU. (C) 2018 Societe francaise d'anesthesie et de reanimation (Sfar). Published by Elsevier Masson SAS. All rights reserved.