ObjectiveTo evaluate the outcome of artificial anal sphincter implantation for severe fecal incontinence in 37 consecutive patients operated on in a single institution from 1993 through 2001.Summary Background DataImplantation of an artificial anal sphincter is proposed in severe fecal incontinence when local treatment is unsuitable or has failed. The results of this technique have not been determined yet, and its place among the various operative procedures is still debated.MethodsArtificial anal sphincters were implanted in 37 patients from 1993 through 2001. All patients had complete fecal incontinence and had failed to respond to medical treatment. Median duration of incontinence was 16 years. The causes of incontinence were sphincter disruption (19 patients), hereditary malformations (2 patients), and neurologic disease (16 patients). Six patients had had previous surgery for fecal incontinence. Assessment was made by physical examination (anal continence, rectal emptying) and anorectal manometry.ResultsIn the first 12 patients, six devices had to be removed (50%); the cause of failure was found in all cases, and this allowed contraindications to be defined. Among the next 25 patients, 23 had an uncomplicated postoperative follow-up, and 5 developed seven complications: control pump change (n = 3), balloon migration (n = 1), and major rectal emptying difficulties in patients with obstructive internal rectal procidentia (n = 2). The artificial anal sphincter had to be removed definitively in three cases, representing the failure rate of this technique in the authors’ experience (12%); two other devices had to be removed temporarily and the patients are awaiting reimplantation. In this latter group of 25 patients, 80% have an activated sphincter: continence for liquid stool is normal in 78.9%, continence for gas in 63.1%. Seven patients have rectal emptying difficulties, minor in five and major in two. Manometric studies showed mean pressures of 110 and 37 cm H2O with closed and open sphincter, respectively, with a mean duration of artificial sphincter opening of 128 seconds.ConclusionsThe long-term functional outcome of artificial anal sphincter implantation for severe fecal incontinence is satisfactory; adequate sphincter function is recovered and the definitive removal rate is low. Good results are directly related to careful patient selection and appropriate surgical and perioperative management after a learning curve of the surgical team.