In patients with fistulizing perianal Crohn's disease (CD), the need for a secondary surgical step is not defined. The aim was to assess the efficacy of surgical closure compared to a single seton removal in patients with drained fistulizing perianal CD treated with adalimumab. This was a multicentre, randomized controlled trial, comparing seton removal+surgical closure (closure group) to seton removal alone (control group) with a stratification according to the American Gastroenterological Association classification. The primary end-point was fistula closure at month 12 defined by the association of the following criteria: no seton, absence of a visible external opening, absence of discharge from the tract after finger compression, absence of an internal opening, absence of perianal pain/abscess and absence of fistula-related abnormalities. Among the 64 included patients (262 expected) (48 complex fistula, 75%), 33 were randomized to the closure group and 31 to the control group. In the closure group, 26 patients (78.8%) had glue. At month 12, overall fistula closure was achieved in 35 of the evaluable 58 patients (60%): 18/32 (56%) in the surgery group and 17/26 (65%) in the control group (P=0.479). In the closure group, fistula closure was observed in 13/25 (52%) and 5/7 (71%) patients with complex and simple fistula respectively (P=0.426), compared with 12/18 (67%) and 5/8 (63%), respectively in the control group (P=1.000). Seton removal alone seems to be no more effective than a secondary surgical step (in particular glue injection) in patients having fistulizing perianal CD controlled by an initial drainage combined with adalimumab. The results should be interpreted with caution.