P481 Repeated surgery for recurrent ileocolonic Crohn’s disease: does the outcome keep worsening operation after operation? A comparative study of 1224 consecutive procedures

F. Colombo1, A. Frontali1, C. Baldi1, G.M.B. Lamperti1, G. Maconi2, S. Ardizzone2, D. Foschi2, G. Sampietro1

1General Surgery, L.Sacco University Hospital, Milan, Italy, 2Gastroenterology Unit, L.Sacco University Hospital, Milan, Italy

Background

Despite relevant improvement in the medical treatment of ileocolic Crohn’s disease (CD), still surgery is needed in 80% of patients and clinical recurrence occurs in more than 50% of cases after surgery. Aim of the study is to assess the outcome for patients undergoing repeated surgery for recurrent CD.

Methods

All patients undergoing surgery for ileal or colonic CD between 1993 and 2018 in our tertiary care centre were retrospectively reviewed. We considered all small bowel resections, colonic resections, conventional (SP) and not conventional strictureplasties (NCSP).

Results

In the study period, 1224 CD patients underwent a surgical operation. We performed 713 (58.2%) primary operations (1R), 325 (26.5%) re-operations (2R group) and 186 (15.3%) three or more interventions (≥3R group). A CD diagnosis and the time of first surgery at early age were a negative prognostic factor, favouring repeated surgery in the time (R1 vs. R2 p < 0.004 and R1 vs. ≥3R p < 0.0001 comparing age at diagnosis; R1 vs. R2 p < 0.0001 and R1vs. ≥3R< 0.0001 regarding the time to firstt surgery). The indication for surgery (stenosis, fistula/abscess and refractoriness to medical therapy) does no’t change significantly with the number of surgeries regarding the analysis of variance of the three groups (p = 0.3). In the repeated surgery (R2 and ≥3R group) the incidence of recurrence appears to be more frequent at the anastomotic site for the 2R group (197/325, 61% vs. 103/186, 55% p < 0.05) and at SP site for the ≥3R group (19/325, 5% vs. 23/186, 12% p < 0.01). The ≥3R group underwent to a higher number of strictureplasties in comparison to the 1R group (100/186, 53% vs. 305/713, 43% p < 0.0001) and the 2R group (100/186, 53 vs. 127/325, 43% p < 0.0004). The duration of the surgical procedure tends to be higher for the ≥3R group but without reaching a statistical significance (p = 0.2). Postoperative morbidity (Clavien Dindo I-V) was increased in the 2R and in the ≥3R group but not in a significant way: (1R group n = 187 (26%); 2R group n = 102 (31%); ≥3R group n = 64 (34%); 1R vs. 2R p = 0.1, 2R vs. ≥3R group p = 0.5). In particular there was no difference between the groups in the incidence of severe postoperative morbidity (Clavien-Dindo ≥3): 1R group n = 69 (9.6%); 2R group n = 37 (11%); ≥3R group n = 24 (12%); 1R vs. 2R p = 0.4, 2R vs. ≥3R group p = 0.7.

Conclusion

In this series, we treated more than 40% of the small bowel segments with SP and NCSP at first surgery. This percentage gradually increases to 49% in 2R group and to 53% in ≥3R group. Even if the overall morbidity rate was higher, repeated surgery for recurrent CD doesn’t appear to be related to an increased risk of severe postoperative morbidity in our experience.