STUDY Objective: To determine the safety of manual vaginal morcellation by evaluating the rates of incidental uterine malignancy and manual vaginal morcellation after vaginal or laparoscopic-assisted vaginal hysterectomy.

Design: Retrospective analysis (Canadian Task Force classification II-2). Setting: University of Texas Southwestern Medical Center, Dallas, TX. Patients: Women (n = 1,629) undergoing vaginal or laparoscopic-assisted vaginal hysterectomy. Interventions: Vaginal hysterectomy (n = 1,091) or laparoscopic-assisted vaginal hysterectomy (n = 538) with and without scalpel morcellation. Measurements and main

Results: The number of uterine malignancies, rate of vaginal morcellation, surgical indications, pathology diagnoses, and uterine weights were evaluated. Chi-square analysis was used to compare categoric data, and analysis of variance was used to compare uterine weights. There were no cases of leiomyosarcomas. There were 2 other sarcomas, 4 smooth muscle tumors of uncertain malignant potential, and 8 endometrial adenocarcinomas. The vaginal morcellation rate was 19.4%, but no malignancy was morcellated. Myomas were more common preoperatively and histologically in morcellated specimens. Mean (± standard deviation) uterine weights for morcellated versus nonmorcellated laparoscopic-assisted vaginal hysterectomy specimens were 285.5 ± 159.3 versus 140.1 ± 83.6 g (p < .001), respectively, and 199.9 ± 92.8 versus 111.9 ± 61.4 (p < .001), respectively, for vaginal hysterectomy.

Conclusion: Vaginal manual morcellation is safe with a low risk of incidental malignancy. Variables that influence the decision for the vaginal approach may also affect malignancy risk and morcellation decisions. Thus, all patients undergoing vaginal or laparoscopic-assisted vaginal hysterectomy should be counseled regarding incidental malignancy, risk of morcellation, and alternatives for intact specimen removal.