The effect of purse-string approximation versus linear approximation of ileostomy reversal wounds on morbidity rates and patient satisfaction: the 'STOMA' trial.

Abstract

Ileostomy reversal is associated with surgical site infection (SSI) rates as high as 37%. Recent literature suggests that employing a purse-string approximation (PSA) of the reversal wound reduces this rate of SSI. Thus we wished to perform a randomised controlled trial to compare SSI rates in purse-string versus linear closure (PLC) wounds following ileostomy reversal. A randomised, controlled trial was conducted at University Hospital Limerick. Sixty-one patients undergoing ileostomy reversal were included. Thirty-four patients were randomised to PSA and 27 patients to linear closure. The primary endpoint was incidence of SSI and secondary endpoints measured were quality of life and satisfaction with cosmesis. Statistical analysis was performed on a per protocol basis using SPSS version 22.0. Three patients in the PSA group developed an SSI compared to 8 in the PLC group at 30days (8 vs 30%, p=0.03). The mean time to SSI diagnosis was faster in the PSA group (3 vs 12.3days, p=0.08). Patients who developed SSI experienced a longer mean length of stay (6.8 vs 11.4days, p=0.012). On multivariate analysis, PLC was the only predictive factor of SSI formation (p<0.001). There was no difference in patient satisfaction between the two study groups (p=0.14). PSA of wounds following ileostomy reversal significantly reduces SSI formation compared to linear approximation without any effect on patient satisfaction.