CR32P PREDICTORS OF DAY STAY AFTER COLONIC SURGERY IN A STRUCTURED MULTI‐MODAL CARE PROGRAM

Background:   Enhanced Recovery After Surgery (ERAS) programs have gained popularity in colonic surgery. Hospital stay has been reduced dramatically as a result, but remains varied in the published literature. We were interested in investigating the factors that influence hospital stay in an ERAS setting. Methods:   Between October 2005 and November 2008 prospective data were collected on consecutive patients undergoing elective colonic resection, without a stoma, at a single site. Patients were managed within an ERAS program. Patients unable to communicate in English, with dementia or ASA >4 were excluded. Variables were tested using the Mann-Whitney U and Spearman's two tailed correlation table for continuous data. Cox regression analysis was used for modelling. Results:   102 patients were included. There were 57 right sided, 43 left sided and 2 total colectomies. Mean age was 66 years (95% CI, 40–88). Median day stay was 4 days (range 3–46). Indications for surgery were: malignancy (80%), diverticular disease (13%) and other (7%). Age, operation site, ASA, indication for surgery and BMI were not significant predictors of hospital stay. Factors with the strongest correlation for reduced day stay were Cr-Possum score (p = 0.006), female sex (p = 0.013), laparoscopic (n = 6) and transverse (n = 32) incisions (p < 0.001), and the operating surgeon (p < 0.001). Using Cox regression modeling, surgeon and patient gender could be eliminated as independent predictors of day stay. The other two factors added significant strength to the model. Conclusion:   Choice of incision and Cr-POSSUM score are independent predictors of day stay in the ERAS setting.