Hysteropreservation versus hysterectomy in the surgical treatment of uterine prolapse: systematic review and meta-analysis.

Abstract

The efficacy and safety of removing or preserving the uterus during reconstructive pelvic surgery is a matter of debate. We performed a systematic review and meta-analysis of studies that compared hysteropreservation and hysterectomy in the management of uterine prolapse. PubMed, Medline, SciELO and LILACS databases were searched from inception until January 2017. We selected only randomized controlled trials and observational cohort prospective comparative studies. Primary outcomes were recurrence and reoperation rates. Secondary outcomes were: operative time, blood loss, visceral injury, voiding dysfunction, duration of catheterization, length of hospital stay, mesh exposure, dyspareunia, malignant neoplasia and quality of life. Eleven studies (six randomized and five non-randomized) were included involving 910 patients (462 in the hysteropreservation group and 448 in the hysterectomy group). Pooled data including all surgical techniques showed no difference between the groups regarding recurrence of uterine prolapse (RR 1.65, 95% CI 0.88-3.10; p=0.12), but the risk of recurrence following hysterectomy was lower when the vaginal route was used with native tissue repair (RR 10.61; 95% CI 1.26-88.94; p=0.03). Hysterectomy was associated with a lower reoperation rate for any prolapse compartment than hysteropreservation (RR 2.05; 95% CI 1.13-3.74; p=0.02). Hysteropreservation was associated with a shorter operative time (mean difference -12.43min; 95% CI -14.11 to -10.74; p<0.00001) and less blood loss (mean difference -60.42ml; 95% CI -71.31 to -49.53ml; p<0.00001). Other variables were similar between the groups. Overall, the rate of recurrence of uterine prolapse was not lower but the rate of reoperation for prolapse was lower following hysterectomy, while operative time was shorter and blood loss was less with hysteropreservation. The limitations of this analysis were the inclusion of nonrandomized studies and the variety of surgical techniques. The results should be interpreted with caution due to potential biases.