Study selection and characteristics.
The literature search yielded 4998 results. After the removal of duplicates, reviews and case reports, 1124 results remained for title and abstract screening, resulting in the selection of 55 studies for full-text assessment. Of these, 17 published papers including thirteen cohort studies13-29 met the defined inclusion criteria and compared the UAE and MYO groups. Fig.1 shows the flow diagram of study selection.
The Table 1 presented the characteristics of the included studies, which comprised of four RCTs14-21, two prospective cohorts22,28 and seven retrospective studies.23-27,29,30 The sample size of the studies ranged from 80 to 18433 participants. According to the NOS assessment, five studies20,22,23,25,26 were rated eight-stars and four studies21,22,26,29 were rated seven-stars. The four RCTs were considered high-quality studies, although the “other bias” category was a high risk of bias. The quality of evidence for all outcomes was above medium.
Patients’ characteristics.
The baseline characteristics of patients between the UAE and MYO groups were compared in Table 2. The mean age of the patients in the UAE group was 43.3 years, while it was 39.5 years in the MYO group. The racial composition of the UAE group was 40.9% white and 24.3% black, while the MYO group had 32.9% white and 35.4% black. A small proportion of patients in both groups had subfertility (UAE, 13.1%; MYO, 15.1%) and pelvic pain (UAE, 24.5%; MYO, 12.1%). Bleeding symptoms were present in 38.7% of the UAE group and 38.8% of the MYO group. The location of the largest fibroid was submucosa in 9.4% of the UAE group and 11.2% of the MYO group, subserosa in 29.3% of the UAE group and 27.3% of the MYO group, and intramural in 21.7% of the UAE group and 32.1% of the MYO group. The number of fibroids was 1–3 in 56.2% of the UAE group and 58.3% of the MYO group, 4–10 in 25.3% of the UAE group and 31.5% of the MYO group, and >10 in 29.3% of the UAE group and 12.1% of the MYO group. The gravidity rate was 86.9% in the UAE group and 78.8% in the MYO group. The mean uterine volume was 796.6 cm3 in the UAE group and 731.7 cm3 in the MYO group.
Early outcomes
Outcome measured in hospitalization or predischarge are summarized in detail in Table 3. The meta-analysis found that some outcomes in the UAE group were significantly lower than the MYO group including early major complications (OR, 0.44; 95% CI, 0.20 to 0.95; P =0.04; heterogeneity: I2 = 25%)15,18,20, readmission (OR,1.16; 95%CI, 1.01 to 1.33; P =0.04; heterogeneity: I2 = 0%)15,18,20,24, hospital stay (MD,-47.07; 95%CI, -58.94 to -35.2; P < 0.01; heterogeneity: I2 = 87%)20,22,23,30 and primary cost (MD, -47.56; 95%CI, -53.27 to -41.85; P < 0.01; heterogeneity: I2 = 65%)15,18,20 However, there was no significant difference between the UAE group and the MYO group in delayed discharge (OR, 2.05; 95% CI, 0.74 to 5.68; P =0.17; heterogeneity: I2 = 0%)15,18,20.
Long-term outcomes
Pregnancy condition
Table 4 indicated that there was no significant difference in the gravidity rates (OR, 0.89; 95%CI, 0.18 to 4.49; P =0.88; heterogeneity: I2 = 55%)17,18,20, live birth (OR, 0.62; 95%CI, 0.12 to 3.07; P =0.56; heterogeneity: I2 = 75%)17,18,20, miscarriage (OR, 2.43; 95%CI, 0.90 to 6.54; P =0.08; heterogeneity: I2 = 13%)17,18,20 and termination (OR, 1.04; 95%CI, 0.15 to 7.49; P =0.97; heterogeneity: I2 = 0%)17,18,20 between the UAE group and the MYO group.
Abnormal uterine bleeding
The analysis of the original data revealed that the incidence of abnormal uterine bleeding was improved from pre-operation to post-operation16,21, such as amenorrhea (0.4% to 17.2%), light (1.4% to 42.5%), normal (30.9% to 29.3%) and heavy (67.4% to 12.1%). However, the synthesized outcomes indicated that there was no significant difference between the UAE group and the MYO group in terms of amenorrhea (OR, 0.80; 95% CI, 0.35 to 1.84, P = 0.60; Heterogeneity: I² = 0%), light (OR, 1.83; 95% CI, 0.58 to 5.80, P = 0.31; Heterogeneity: I² = 0%), normal (OR, 0.50; 95% CI, 0.05 to 4.72, P = 0.55; Heterogeneity: I² = 59%) and heavy (OR, 0.96; 95% CI, 0.38 to 2.42; P = 0.93; Heterogeneity: I² = 0%) at the end of follow-up (Table 5).
Reintervention
Fig.2 demonstrates that the rate of reintervention was significantly higher in the UAE group than in the MYO group during one-year, two-year and four-year follow-up (OR, 1.77; 95% CI, 1.54 to 2.04, P < 0.01; Heterogeneity: I² = 0%; OR, 3.44; 1.52 to 7.77, P < 0.01; Heterogeneity: I² = 67%; OR, 1.84; 95% CI, 1.62 to 2.10, P < 0.01; Heterogeneity: I² = 39%, respectively). Furthermore, a meta-analysis of the re-intervention events revealed that women in the UAE group were at a significantly greater risk of hysterectomy than those in the MYO group during one-year, two-year and four-year follow-up (OR, 2.67; 95% CI, 2.18 to 3.25, P < 0.01; Heterogeneity: I² = 37%; OR, 4.06; 95% CI, 3.31 to 4.9, P < 0.01; Heterogeneity: I² = 16%; OR, 4.04; 95% CI, 3.45 to 4.72, P < 0.01; Heterogeneity: I² = 59%, respectively) (Fig. 3).
UFs scores for quality of life (QOL) and symptom severity
The results of the UFs questionnaire score have been a crucial factor in assessing the quality of life and symptom severity in women with UFs. 30,31 According to the data, there is no significant difference between the UFs treated with UAE or MYO in terms of improvement of QOL at two-year and four-year follow-up (OR, -0.56; 95% CI, -7.33 to 6.22, P = 0.87; Heterogeneity: I² = 78%; OR, -2.42; 95% CI, -5.31 to 0.48, P = 010; Heterogeneity: I² = 0%)18,22 (Fig.4). However, the fixed-effects model indicated, in Fig 5, that the UAE group had significantly higher scores of symptom severity compared to the MYO group at two-year and four-year follow-up (OR, 2.13; 95% CI, 0.41 to 3.85, P = 0.02; Heterogeneity: I² = 0 %; OR, -4.02; 95% CI, 0.82, 7.22, P = 0.01; Heterogeneity: I² = 0%).18,22