PURPOSE: To compare flap-specific complications of rectus abdominis myocutaneous (RAM) and myoperitoneal (RAMP) flap neovagina reconstructions performed concurrently with radical pelvic procedures. MATERIALS AND METHODS: Retrospective single institution chart review of all patients with RAM or RAMP flap neovaginal reconstructions performed on a Gynecologic Oncology service, 1988-2003. Analysis for associations with flap-specific morbidity was performed. RESULTS: Neovaginal reconstructions comprised 32 RAM and 7 RAMP flaps. Twenty-two (69%) RAM patients underwent total pelvic exenteration compared to 1 (14%) RAMP patient (P < 0.013). Overall, 33 (85%) of the patient population had previously been treated with radiation. Flap-specific complications developed in 12 (32%) RAM versus 4 (57%) of the RAMP patients (P > 0.1). Donor site complications and incisional hernias were increased in RAMP patients (both P < 0.03), with trends for increasing risk of vaginal stricture/stenosis and superficial wound separations (both P < 0.1). Complete vaginal stenosis developed in only 1 (3%) RAM versus 3 (43%) RAMP patients. Furthermore, 3 RAMP patients developed complete stenosis when the vaginal defect was circumferential and involved >65% of the vagina while this did not occur in 22 similar RAM patients (P < 0.0005). Only patients with partial longitudinal defects maintained vaginal patency after RAMP flap. Fifteen (58%) of 26 patients surviving >12 months reported coitus, with no significant difference between the groups. CONCLUSIONS: When there is circumferential loss of the upper 2/3 of the vagina. RAMP flaps are not suitable for neovaginal reconstruction after radical pelvic surgery because of an increased risk of vaginal stenosis compared to RAM flaps. Patients with partial longitudinal vaginal defects, however, may have successful neovaginal reconstruction with RAMP flaps.