Background There is an ongoing debate whether the type of anastomosis following intestinal resection for Crohn’s disease (CD) can have an impact on complications and postoperative recurrence. The aim of the present study is to describe the outcome of side-to-side (S-S) versus end-to-end (E-E) anastomosis after ileocolic resection for CD. Methods A retrospective single center comparative study was conducted in consecutive CD patients who underwent primary ileocecal resection between 2005 and 2013. All patients underwent colonoscopy 6-months postoperatively to assess endoscopic recurrence, defined as modified Rutgeerts’ score >i2b. Surgical recurrence implied reoperation due to CD activity at the anastomotic site. Modified surgical recurrences was defined as the need for re-operation or balloon-dilation. Perioperative factors related to recurrence were evaluated. Results Of the128 patients included, 52 (40.6%) received an E-E anastomosis. Median follow-up was longer in the E-E group (8.62 vs 13.68 years, p<0.001). Apart from the primary anastomosis, patient, disease and surgical characteristics were similar between both groups. Postoperative complications and anastomotic leak were comparable (S-S 6.6% vs E-E 7.7%, p=1.00). During the entire postoperative follow-up, biologicals were used in 55.3% and 63.5% (p=0.37) in S-S and E-E patients, respectively. Endoscopic recurrence did not differ between S-S and E-E patients (40.8% vs 49.0%, p=0.37). However, at ten years follow-up, surgical and modified surgical recurrence rate were significantly higher in the E-E group (19.7% vs 7.4%, p=0.02 and 39.2% vs 14.1%, p=0.001). Type of anastomosis was the only independent risk factor for surgical and modified surgical recurrence. Conclusion The type of anastomosis did not influence endoscopic recurrence and immediate postoperative disease complications. However, the wide diameter and the morphologic characteristic of the stapled S-S anastomosis result in a lower risk for balloon dilatation and surgical reintervention in the long-term.