Objectives Inferior outcomes of minimally invasive surgery (MIS) in cervical cancer may be attributable to exposure of peritoneum to tumor at colpotomy. Vaginal surgery may minimize surgical morbidity while avoiding dissemination. We sought to compare cervical cancer outcomes by surgical approach.

Methods A retrospective cohort study of cervical cancer patients in ten Canadian centers between 2007–2017. Patients with FIGO 2018 stage IA1, LVI+, and stages IA2-IIIC tumors <4cm were included. Patients undergoing MIS, abdominal (AH) and vaginal or laparoscopy-assisted vaginal hysterectomy (CLVH) were compared. PFS and OS were assessed using the product-limit method, and Cox regression was performed to evaluate association of surgery with outcomes.

Results 1066 patients met inclusion criteria (518 MIS, 436 AH and 110 CLVH). Radical hysterectomy was performed in 80% (CLVH), 96.9% (MIS) and 89.8% (AH) of cases. CLVH cases included more adeno/adenosquamous cancers (70.9% vs. 38.3%(MIS) and 50%(AH), p<0.001), more microinvasive disease (30.9% vs 21.4%(MIS) and 15.8%(AH), p=0.005), smaller tumors (8mm vs. 13mm(MIS), 15mm(AH), p=0.006), fewer LVI+ (20.9% vs. 39%(MIS), 35.9%(AH), p=0.001) and similar rates of lymphatic spread (11.1% vs 11.1%(MIS), 9.7%(AH)). CLVH was associated with fewer intraoperative (5.6% vs 5.6%(MIS), 10.1%(AH), p=0.023) and postoperative (11.8% vs 18.9%(MIS), 24.5%(AH), p=0.006) complications and readmissions (4.6% vs. 11.9%(MIS), 13.9%(AH), p=0.028). CLVH was further associated with a lower risk of recurrence, even when adjusted for age and stage (HR=2.6, 95% CI 1.04–6.51 (AH) and HR=3.07, 95% CI 1.23–7.64 (MIS)).

Conclusions CLVH for cervical cancer is associated with excellent perioperative outcomes. Oncological outcomes appear promising and warrant prospective exploration.