Variables associated with loss of ileoanal pouches constructed in childhood.

AIMS: To quantify the incidence of loss of an ileoanal pouch in children and to identify variables associated this event.

METHODS

Logistic regression of possible explanatory variables: age, sex, indication for procto-colectomy, pre-operative continence status, use of immunosuppressive drugs, open or laparoscopic surgery, number of stages, anastomotic leak, operative complication excluding anastomotic leak, performance of revisional surgery, albumin concentration and platelet count at time of surgery, rank order in series. Univariate logistic regression was used to identify significant variables which were then assessed in a multivariate model and construction of Kaplan-Meier graphs.

RESULTS

103 children, 56 girls, underwent ileoanal pouch at median age 14years (SD 3.7). Indications and mean age were: ulcerative colitis (n=71, 14years), polyposis syndromes (n=13, 15years), chronic idiopathic constipation (n=9, 11years), Hirschsprung's disease (n=4, 1year), Crohn's disease (n=2, 16years), and fibrosing colonopathy (n=2, 11years). 13 patients had their pouch excised or permanently diverted. Three patients had successful revisional pouch surgery. Only pre-operative fecal incontinence and anastomotic leak were significantly associated with pouch excision/diversion, however on multivariate analysis, only fecal incontinence remained significant, odds ratio 21 (95%CI 1.8-235, p=0.01). Pouch survival was significantly worse where there was fecal incontinence pre-operatively, p<0.0001 or an anastomotic leak, p<0.001.

CONCLUSIONS

13% of children subjected to restorative procto-colectomy ultimately receive a permanent ileostomy. Fecal incontinence prior to surgery is a relative contra-indication. Anastomotic leak increases the probability of later pouch excision.

LEVEL OF EVIDENCE: IV.