Endometrial resection and ablation versus hysterectomy for heavy menstrual bleeding.

backgroundHeavy menstrual bleeding (HMB) is common in otherwise healthy women of reproductive age, and can affect physical health and quality of life. Surgery is usually a second-line treatment of HMB. Endometrial resection/ablation (EA/ER) to remove or ablate the endometrium is less invasive than hysterectomy. Hysterectomy is the definitive treatment and can be via open (laparotomy) approach, or via minimally invasive approaches (vaginally or laparoscopically). Each approach has its own advantages and risk profile.

objectivesTo compare the effectiveness, acceptability and safety of endometrial resection or ablation versus different routes of hysterectomy (open, minimally invasive hysterectomy, or unspecified route) for the treatment of HMB. SEARCH

methodsWe searched the Cochrane Gynaecology and Fertility specialised register, CENTRAL, MEDLINE, Embase and PsycINFO (July 2020), and reference lists, grey literature and trial registers. SELECTION CRITERIA: Randomised controlled trials (RCTs) that compared techniques of endometrial resection/ablation with hysterectomy (by any technique) for the treatment of HMB in premenopausal women. DATA COLLECTION AND ANALYSIS: We used standard methodological procedures expected by Cochrane. MAIN

resultsWe included 10 RCTs (1966 participants) comparing EA/ER to hysterectomy (open (abdominal), minimally invasive (laparoscopic or vaginal), or unspecified (or at surgeon's discretion) route of hysterectomy). The results were rated as moderate-, low- and very low-certainty evidence. Endometrial resection/ablation versus open hysterectomy We found two trials. Women having EA/ER are probably less likely to perceive an improvement in HMB compared to women having open hysterectomy (risk ratio (RR) 0.90, 95% confidence interval (CI) 0.84 to 0.95; 2 studies, 247 women; moderate-certainty evidence) and probably have a 13% risk of requiring further surgery for treatment failure (compared to 0 on the open hysterectomy group; 2 studies, 247 women; moderate-certainty evidence). Both treatments probably lead to similar quality of life at two years (mean difference (MD) -5.30, 95% CI -11.90 to 1.30; 1 study, 155 women; moderate-certainty evidence) and satisfaction rate at one year (RR 0.91, 95% CI 0.82 to 1.00; 1 study, 194 women; moderate-certainty evidence). There may be no difference in serious adverse events (RR 1.29, 95% CI 0.32 to 5.20; 2 studies, 247 women; low-certainty evidence). EA/ER probably reduces time to return to normal activity compared to open hysterectomy (MD -21.00 days, 95% CI -24.78 to -17.22; 1 study, 197 women; moderate-certainty evidence). Endometrial resection/ablation versus minimally invasive hysterectomy We found five trials. The proportion of women with perception of improvement in HMB at two years may be similar between groups (RR 0.97, 95% CI 0.90 to 1.04; 1 study, 79 women; low-certainty evidence). Blood loss may be higher in the EA/ER group when assessed using the Pictorial Blood Assessment Chart (MD 44.00, 95% CI 36.09 to 51.91; 1 study, 68 women; low-certainty evidence). Quality of life is probably lower in the EA/ER group compared to the minimally invasive hysterectomy group at two years according to the 36-item Short Form (SF-36) (MD -10.71, 95% CI -15.11 to -6.30; 2 studies, 145 women; moderate-certainty evidence) and Menorrhagia Multi-Attribute Scale (RR 0.82, 95% CI 0.70 to 0.95; 1 study, 616 women; moderate-certainty evidence). EA/ER probably increases the risk of further surgery for HMB compared to minimally invasive hysterectomy (RR 7.70, 95% CI 2.54 to 23.32; 4 studies, 922 women; moderate-certainty evidence) and treatments probably have similar rates of any serious adverse events (RR 0.75, 95% CI 0.35 to 1.59; 4 studies, 809 women; moderate-certainty evidence). Women with EA/ER are probably less likely to be satisfied with treatment at one year (RR 0.90, 95% CI 0.85 to 0.94; 1 study, 558 women; moderate-certainty evidence). We were unable to pool data for time to return to work or normal life because of extreme heterogeneity (99%); however, the three studies reporting this all had the same direction of effect favouring EA/ER. Endometrial resection/ablation versus unspecified route of hysterectomy We found three trials. EA/ER may lead to a lower perception of improvement in HMB compared to unspecified route of hysterectomy (RR 0.89, 95% CI 0.83 to 0.95; 2 studies, 403 women; low-certainty evidence). Although EA/ER may lead to similar quality of life using the SF-36 General Health Perception at two years' follow-up (MD -1.90, 95% CI -8.67 to 4.87; 1 study, 209 women; low-certainty evidence), the proportion of women with improvement in general health at one year may be lower (RR 0.85, 95% CI 0.77 to 0.95; 1 study, 185 women; low-certainty evidence). EA/ER probably has a risk of 5.4% of requiring further surgery for treatment failure (compared to 0 with total hysterectomy; 2 studies, 374 women; moderate-certainty evidence) and reduces the proportion of women with any serious adverse event (RR 0.21, 95% CI 0.06 to 0.80; 2 studies, 374 women; moderate-certainty evidence). Both treatments probably lead to a similar satisfaction rate at one year' follow-up (RR 0.96, 95% CI 0.88 to 1.04; 3 studies, 545 women; moderate-certainty evidence). EA/ER may lead to shorter time to return to normal activity (MD -18.90 days, 95% CI -24.63 to -13.17; 1 study, 172 women; low-certainty evidence). AUTHORS'

conclusionsEndometrial resection/ablation (EA/ER) offers an alternative to hysterectomy as a surgical treatment for HMB. Effectiveness varies with EA/ER compared to different hysterectomy approaches. The perception of improvement in HMB with EA/ER is probably lower compared to open and unspecified route of hysterectomy, but may be similar compared to minimally invasive. Quality of life with EA/ER is probably similar to open and unspecified route of hysterectomy, but lower compared to minimally invasive hysterectomy. Further surgery for treatment failure is probably more likely with EA/ER compared to all routes of hysterectomy. Satisfaction rates also vary. EA/ER probably has a similar rate of satisfaction compared to open and unspecified route of hysterectomy, but a lower rate of satisfaction compared to minimally invasive hysterectomy. The proportion having any serious adverse event appears similar in all groups, but specific adverse events did reported difference between EA/ER and different routes. We were unable to draw conclusions about the time to return to normal activity, but the direction of effect suggests it is likely to be shorter with EA/ER.