Twenty-four studies were included in the review: 4 RCTs reported in 8 publications (n=755) and 11 observational studies reported in 16 publications (number of participants unclear). The sample sizes ranged from 22 to 319 in the RCTs and from 100 to 678 in the observational studies.
Two RCTs reported blinding and sample size calculations and three reported the method of randomisation.
Outcomes favouring total hysterectomy.
Urinary incontinence (4 RCTs, 9 observational studies): overall, no difference in urinary incontinence was found between total and subtotal hysterectomy (OR 1.15, 95% CI: 0.93, 1.43); there was some evidence of statistical heterogeneity (p=0.09; I-squared 37.6%). The subgroup analysis limited to the 4 RCTs showed a reduced risk of incontinence with total hysterectomy (OR 1.73, 95% CI: 1.10, 2.73).
Prolapse (2 RCTs, 3 observational studies): subtotal hysterectomy carried a higher risk of prolapse (OR 4.54, 95% CI: 1.53, 13.47).
Cervical stump problems (1 RCT, 6 observational studies): these occurred at frequencies ranging from 5 to 22% in the group of subtotal hysterectomy. Three studies suggested a lower rate of such complications with total hysterectomy.
Outcomes favouring subtotal hysterectomy.
Operation time (3 RCTs, 2 observational studies): this was on average 12.59 minutes shorter (95% CI: 9.40, 15.78) with subtotal hysterectomy than with total hysterectomy.
Peri-operative bleeding (3 RCTs, 2 observational studies): subtotal hysterectomy was associated with less peri-operative bleeding (WMD -83.27, 95% CI: -116.44, -50.09).
Peri- and post-operative complications (3 RCTs, 6 observational studies): subtotal hysterectomy was associated with lower peri- and post-operative complications (OR 0.69, 95% CI: 0.56, 0.86). However, there was strong evidence of heterogeneity (p=0.0003; I-squared 72.9%).
All other evaluated outcomes (lower urinary tract symptoms, quality of life, psychiatric symptoms, constipation, pelvic pain and sexual function) were comparable between the treatment groups.