Reinterventions after complicated or failed STARR procedure

Abstract

The stapled transanal rectal resection (STARR) procedure has been suggested as a simple surgical option for patients presenting with evacuatory difficulty in the clinical presence of a rectocele. Most of these patients have a multiplicity of pelvic floor pathology unaddressed by the performance of one procedure. The aim of the study was to assess an unselected group of patients referred to a tertiary coloproctological unit following performance of the STARR procedure for obstructed defecation (OD) where the procedure was complicated or had failed. Anorectal, urogynecological, and psychological examination with objective constipation/incontinence scoring, anal-vaginal-perineal ultrasound, manometry, and defecography were selectively performed utilizing the Iceberg Diagram to detect occult pelvic floor pathology. Twenty patients were referred with 13 cases (female, 10; median age, 65 years; range, 40-72) operated upon. Post-STARR surgery was performed for three complications and ten failures including recurrent OD, severe proctalgia, and fecal incontinence. Overall, 11 patients underwent biofeedback therapy and psychotherapy. Of the operated group, 11 patients had a median of four associated disorders. Seven patients had a significant psychological overlay with severe depression or anxiety and four heterogeneous anal sphincter defects. Operative procedures were tailored to the clinical findings using enterocele repair, staple removal, fistulectomy, rectosigmoid resection, and levatorplasty where appropriate. Twelve patients were evaluated after a median follow-up of 18 months. Of these, six (all with psychoneurosis) remained unchanged. Three patients with no psychological overlay were asymptomatic with a further two improved. The STARR procedure, when complicated or failed, has a poor outcome following surgical reintervention. It requires careful patient selection to determine the associated pelvic floor pathology and pre-existent psychopathology.