Intrinsic predictors of prolonged length of stay in a colorectal enhanced recovery pathway: a prospective cohort study and multivariate analysis.

Abstract

This was a prospective cohort study to determine the intrinsic non-modifiable factors influencing length of stay (LOS) in unselected consecutive patients undergoing elective colorectal surgery within an enhanced recovery pathway. This study interrogated a prospective database of consecutive elective procedures from October 2006 to April 2011 at a tertiary referral academic hospital in the UK to identify independent predictors of prolonged length of stay (pLOS). pLOS was defined as longer than median length of stay (mLOS). Differences in determinants were identified in three groups of increasing operative complexity. In all, 872 procedures were identified and ranged from a simple ileostomy reversal to complex total pelvic exenteration. Preoperative anaemia and American Society of Anesthesiologists (ASA) Grade III+ predicted pLOS in stoma reversal surgery patients (n=191, mLOS4days). In colonic and small bowel surgery (n=444, mLOS 8days), an open procedure, new stoma formation, planned critical care admission and ASA III+ predicted pLOS. New stoma formation and planned critical care admission predicted pLOS in patients undergoing pelvic rectal surgery (n=237, mLOS 11days). pLOS was associated with significantly higher morbidity across Dindo-Clavien grades and a longer time to postoperative functional recovery and discharge. Operative complexity is associated with longer LOS even with an established enhanced recovery pathway in place. Intrinsic non-modifiable predictors of pLOS differ with operative complexity, and this should be taken into account when planning benchmarking and research across units.