Reproductive outcomes after fertility-sparing interventions for symptomatic adenomyosis: a systematic review and meta-analysis

The titles and abstracts of 513 records retrieved from databases after de-duplication were screened, and the full texts of 159 articles were further assessed for eligibility. A total of 127 articles that did not meet the inclusion criteria were excluded. We excluded articles that did not report any reproductive outcome ( n = 93), case reports ( n = 15), repeated articles ( n = 6), review articles ( n = 9), articles in other languages ( n = 2), and articles that reported the total number of participants rather than the number of participants desiring fertility ( n = 2). Therefore, 32 articles [ 19 – 50 ] that met the prespecified criteria were included in the meta-analysis. Of these, 19 articles [ 19 – 37 ] focused on adenomyomectomy, one article [ 38 ] focused on UAE, and 13 articles [ 33 , 39 – 50 ] focused on image-guided thermal ablation (HIFU: 10 articles [ 33 , 39 – 47 ], RFA: 2 articles [ 48 , 49 ], MWA: 1 article [ 50 ]). One article [ 33 ] compared the reproductive outcomes of patients treated with laparoscopic and HIFU. As only one article regarding UAE was included, the meta-analysis for UAE could not be performed. (Fig. 1 ) Fig. 1 PRISMA flow chart for study selection PRISMA flow chart for study selection Among the included articles comprising 2501 participants, 903 (36.1%) underwent adenomyomectomy, 6 (0.2%) underwent UAE, and 1592 (63.7%) received thermal ablation (1435 underwent HIFU, 118 underwent RFA, and 39 underwent MWA). The proportions of study design, geographic area, and language of articles are presented in Table  1 . Table 1 Summary of the included studies No. of articles according to interventions Adenomyomectomy 19 (55.9%) UAE 1 (8.8%) Image-guided thermal ablation 13 (38.2%) No. of participants according to interventions  Adenomyomectomy 903 (36.1%)  UAE 6 (0.2%)  Image-guided thermal ablation 1592 (53.7%) No. of articles according to study design  Single-arm clinical studies 17 (53.1%)  Non-randomized controlled trials 15 (46.9%)  Prospective studies 10 (31.3%)  Retrospective studies 22 (68.7%) No. of articles according to region  Asia 30 (93.8%)  Europe 2 (6.2%) No. of articles according to language  English 30 (93.8%)  Chinese 2 (6.2%) No. of articles according to evidence quality  High 8 (25.0%)  Medium 23 (71.9%)  Low 1 (3.1%) Summary of the included studies Table  2 provides a comprehensive summary of the first author, publication year, treatments, surgical approach, the number and age of participants desiring fertility, follow-up duration, type of adenomyosis, history of subfertility, and dimensions of the uterus and adenomyosis. Table  3 offers a detailed overview of reproductive outcomes from each article, including pregnancy rate, delivery rate, pregnancy loss rate, spontaneous miscarriage rate, rate of adverse pregnancy outcomes, preterm delivery rate, IVF-ET conception rate, cesarean section rate, birthweight, and gestational age at delivery following treatments. The included articles exhibited a range of follow-up durations (spanning from 6 months to 10 years). Nineteen articles (55.9%) specified the type of adenomyosis, and 17 articles (50.0%) documented the participants’ subfertility history before treatments. Due to data limitations, it was challenging to determine whether there were statistically significant differences between these treatment groups in terms of participant age, follow-up duration, and baseline uterus and adenomyosis sizes. Table 2 Characteristics of the included studies ( N  = 32) First author, year Intervention (Adjuvant treatment) No. of participants preparing for pregnancy Follow-up period Age (year) Type of adenomyosis ( n ) Rate of subfertility history (%) Volume of Uterus (cm 3 ) Volume of Adenomyosis (cm 3 ) Adenomyomectomy ( N  = 19):  Fujishita, 2004 [ 19 ] Adenomyomectomy (Laparotomy) with/without GnRH-a 4 45.6 ± 15.3 months 33.2 ± 4.7 Focal (3) Diffuse (1) 100.0 Diameter 109 mm (89–151 mm) —  Hadisaputra, 2006 [ 20 ] Adenomyomectomy (Laparoscopy) with GnRH-a 10 6–22 months 37.7 ± 5.7 — — — Diameter 76.85 mm (15–799 mm)  Wang, 2009 [ 21 ] Adenomyomectomy (Laparotomy/Laparoscopy) with GnRH-a 44 24 months 38.9 ± 3.8 Focal (44) — — Diameter 55.3 ± 12.1 mm Adenomyomectomy (Laparotomy/Laparoscopy) without GnRH-a 27 24 months 37.0 ± 4.8 Focal (27) — — Diameter 48.8 ± 10.9 mm  Wang, 2009 [ 22 ] Adenomyomectomy with/without GnRH-a 28 36 months 34.3 ± 2.1 Diffuse (28) 100.0 Diameter 101.7 ± 9.2 mm —  Takeuchi, 2010 [ 23 ] Adenomyomectomy (Laparoscopy) with/without GnRH-a 3 35.9 ± 21.4 months 25.2 ± 4.4 Focal (3) — — Diameter 31.8 ± 5.1 mm  AI Jama, 2011 [ 24 ] Adenomyomectomy (Laparoscopy) with GnRH-a 18 36 months 38.1 ± 0.9 — 100.0 Diameter 104 ± 73 mm —  Osada, 2011 [ 25 ] Adenomyomectomy (Laparotomy) without GnRH-a 26 10 years 36.9 ± 4.7 Diffuse (26) 53.8 — —  Sun, 2011 [ 26 ] Adenomyomectomy (Laparotomy/Laparoscopy) with GnRH-a 24 27.6 (6–58) months 33.6 ± 4.0 Focal (24) 100.0 — Diameter 35 ± 15 mm  Huang, 2012 [ 27 ] Adenomyomectomy (Laparotomy) with GnRH-a 9 67.7 (62–83) months 34.2 (31–37) Diffuse (9) 100.0 — —  Chang, 2013 [ 28 ] Adenomyomectomy (Laparotomy) with GnRH-a 56 36 months 38.3 ± 4.6 Focal (56) — — —  Saremi, 2014 [ 29 ] Adenomyomectomy (Laparotomy) without GnRH-a 70 24 (20–50) months 37.46 ± 5.37 — 65.7 — —  Kishi, 2014 [ 30 ] Adenomyomectomy without GnRH-a 102 24 (9–60) months 38 (26–51) — 100.0 — —  Tamura, 2017 [ 31 ] Adenomyomectomy (Laparotomy/Laparoscopy) without GnRH-a 89 — 34.8 ± 4.2 Focal (25) Diffuse (64) 100.0 — —  Gao, 2019 [ 32 ] Adenomyomectomy (Laparotomy) with LNG-IUS 2 36 months 37.8 ± 4.4 — — 314.8 ± 106.3 —  Huang, 2020 [ 33 ] Adenomyomectomy (Laparoscopy) with GnRH-a 43 31 months 35.0 (31.0, 38.0) — 100.0 179.9 (114.3, 240.7) —  Shi, 2021 [ 34 ] Adenomyomectomy (Laparoscopy) with/without GnRH-a 176 47 (24–80) months 32.52 ± 3.20 Focal (81) Diffuse (95) 100.0 — Diameter 54.2 ± 12.0 mm  Hlinecka, 2022 [ 35 ] Adenomyomectomy (Laparotomy/Laparoscopy) without GnRH-a 28 76.81 months 35.0 ± 5.3 Focal (5) Diffuse (23) 57.1 — Diameter 48 ± 16 mm  Zhou, 2022 [ 36 ] Adenomyomectomy (Laparoscopy) with GnRH-a 137 36 months 35.5 (25–45) Diffuse (137) 83.2 — Diameter 91.6 ± 89.1 mm  Hijazi, 2022 [ 37 ] Adenomyomectomy (Laparoscopy) with/without GnRH-a 7 24 months 35.68 ± 5.30 (24–45) Focal (3) Diffuse (4) 14.3 — Diameter 52.5 ± 16.3 mm UAE ( N  = 1): Kim, 2005 [ 38 ] UAE 6 35 (22–60) months 30.7 (28–35) — — — — Image-guided thermal ablation ( N  = 13):  Lee, 2015 [ 39 ] HIFU 299* 12 months 40.43 (24–51) — — 264.14 —  Jayaram, 2016 [ 40 ] HIFU 13 24 months Median 34.0 Diffuse (1) Focal (12) 100.0 — —  Zhou, 2016 [ 41 ] HIFU 68 5 years 37.1 ± 5.2 — 60.3 187.0 ± 75.0 36.0 ± 24.0  Guo, 2018 [ 42 ] HIFU without GnRH-a 45* 12 months 42.42 ± 5.09 — — 150.65 ± 12.89 54.20 ± 7.95 HIFU with GnRH-a 18* 12 months 41.44 ± 4.74 — — 175.16 ± 19.51 61.67 ± 12.98  Sun, 2019 [ 43 ] HIFU with GnRH-a 29 5 years 39 (27–46) — — 220.32 ± 78.21 120.32 ± 18.27  Jiang, 2019 [ 44 ] HIFU without GnRH-a 50* 6 months 39.98 ± 4.22 — — 179.06 ± 15.33 49.97 ± 6.38 HIFU with GnRH-a 50* 6 months 40.96 ± 4.42 — — 169.09 ± 14.88 48.63 ± 7.11  Huang, 2020 [ 33 ] HIFU with GnRH-a 50 31 months 36.0 (32.0, 38.0) Diffuse (30) Focal (20) 100.0 179.9 (114.3,240.7) —  Jeng, 2020 [ 45 ] HIFU 202* 21 (3–38) months 41.7 ± 5.3 — — 473.7 ± 346.5 162.4 ± 191.7  Li, 2021 [ 46 ] HIFU with/without GnRH-a 597* 36 months 40.6 ± 5.5 — — 220.1 ± 118.4 72.7 ± 66.1  Huang, 2022 [ 47 ] HIFU with GnRH-a 14* 12 months 39.9 ± 4.3 Diffuse (6) Focal (8) — — 102.7 ± 48.2  Nam, 2020 [ 48 ] RFA 58 57.4 (3–129) months 35.6 (26–44) Diffuse (9) Focal (49) 91.36 — —  Stepniewska, 2022 [ 49 ] RFA 60* 56 ± 29 months 43.0 ± 3.0 Diffuse (13) Focal (47) — — —  Yang, 2015 [ 50 ] MWA 39* 12 (10–21) months 38.42 ± 4.07 — — — — *: These articles did not specify whether the number of participants referred to the number of total participants or solely the participants preparing for pregnancy, and these articles were not included in the meta-analysis for pregnancy rate and delivery rate “—” means data deficiency GnRH-a Gonadotrophin-releasing hormone agonist, UAE Uterine artery embolization, HIFU High-intensity focused ultrasound, RFA Radiofrequency ablation, MWA Microwave ablation Characteristics of the included studies ( N  = 32) Focal (3) Diffuse (1) Focal (25) Diffuse (64) Focal (81) Diffuse (95) Focal (5) Diffuse (23) Diameter 48 ± 16 mm Focal (3) Diffuse (4) Diffuse (1) Focal (12) Diffuse (30) Focal (20) Diffuse (6) Focal (8) Diffuse (9) Focal (49) Diffuse (13) Focal (47) *: These articles did not specify whether the number of participants referred to the number of total participants or solely the participants preparing for pregnancy, and these articles were not included in the meta-analysis for pregnancy rate and delivery rate “—” means data deficiency GnRH-a Gonadotrophin-releasing hormone agonist, UAE Uterine artery embolization, HIFU High-intensity focused ultrasound, RFA Radiofrequency ablation, MWA Microwave ablation Table 3 Reproductive outcomes of the included studies ( N  = 32) First author, year Intervention (No. of patients preparing for pregnancy) Pregnancy, n (rate, %) Mode of Conception, n (rate, %) Delivery, n (rate, %) Term delivery, n (rate, %) Preterm delivery, n (rate, %) Pregnancy loss, n (rate, %) (Including spontaneous miscarriage, ectopic pregnancy, termination of pregnancy, and stillbirth) Adverse pregnancy outcomes, n (rate, %) (Including neonatal death and major congenital anomaly, n ) Birth weight (g) Time to pregnancy leading to live birth ( n ) Delivery method, n (rate, %) Adenomyomectomy ( N  = 19):  Fujishita, 2004 [ 19 ] Adenomyomectomy with/without GnRH-a (4) 2 (50.0) Spontaneous conception 2 (100.0) IVF/ET 0 1 1 0 (Ongoing pregnancy 1) 0 — — — Cesarean Sect. 1 (100.0) Vaginal delivery 0  Hadisaputra, 2006 [ 20 ] Adenomyomectomy with GnRH-a (10) 3 (30.0) — 1 1 0 (Ongoing pregnancies 2) 0 — Malpresentation (1) 3500 — Cesarean Sect. 1 (100.0) Vaginal delivery 0  Wang, 2009 [ 21 ] Adenomyomectomy with GnRH-a (44) 35 (79.5) — 32 (72.7) 27 (84.4) 5 (15.6) 3 (8.6) Spontaneous miscarriage 3 (8.6) — — — — Adenomyomectomy without GnRH-a (27) 20 (74.1) — 17 (63.0) 15 (88.2) 2 (11.8) 3 (15.0) Spontaneous miscarriage 3 (15.0) — — — —  Wang, 2009 [ 22 ] Adenomyomectomy with/without GnRH-a (28) 13 (46.4) Spontaneous conception 13 (100.0) IVF/ET 0 9 (32.1) — — 4 (30.8) Spontaneous miscarriage 4 (30.8) — — — —  Takeuchi, 2010 [ 23 ] Adenomyomectomy (Laparoscopy) with/without GnRH-a (3) 2 (66.7) — 2 (66.7) 2 (100.0) 0 0 0 — 2 years (1), 7 years (1) Cesarean Sect. 1 (50.0) Vaginal delivery 1 (50.0)  AI Jama, 2011 [ 24 ] Adenomyomectomy with GnRH-a (18) 8 (44.4) Spontaneous conception 8 (100.0) IVF/ET 0 6 (33.3) 6 (100.0) 0 2 (25.0) Spontaneous miscarriage 2 (25.0) 1 (16.7) Retained placenta (1) — 4–30 months (6) Cesarean Sect. 6 (100.0) Vaginal delivery 0  Osada, 2011 [ 25 ] Adenomyomectomy without GnRH-a (26) 16 (61.5) Spontaneous conception 4 (25.0) IVF/ET 12 (75.0) 14 (53.8) 14 (100.0) 0 2 (12.5) Spontaneous miscarriage 2 (12.5) 0 — — Cesarean Sect. 14 (100.0) Vaginal delivery 0  Sun, 2011 [ 26 ] Adenomyomectomy with GnRH-a (24) 8 (33.3) Spontaneous conception 3 (37.5) IVF/ET 5 (62.5) 3 (12.5) — — 5 (62.5) Spontaneous miscarriage 5 (62.5) — — — —  Huang, 2012 [ 27 ] Adenomyomectomy with GnRH-a (9) 3 (33.3) Spontaneous conception 3 (100.0) IVF/ET 0 2 (22.2) — — 1 (33.3) Spontaneous miscarriage 1 (33.3) — — — Cesarean Sect. 2 (100.0) Vaginal delivery 0  Chang, 2013 [ 28 ] Adenomyomectomy with GnRH-a (56) 23 (41.1) Spontaneous conception 23 (100.0) IVF/ET 0 15 (26.8) 13 (86.7) 2 (13.3) 12 (52.1) Spontaneous miscarriage 4 (17.4) Ectopic pregnancy 1 (4.3) Termination of pregnancy 7 (30.4) 0 — 1 year (8), 2 years (5), 3 years (2) —  Saremi, 2014 [ 29 ] Adenomyomectomy without GnRH-a (70) 21 (30.0) Spontaneous conception 7 (33.3) IVF/ET 14 (66.7) 16 (22.9) 15 (93.8) 1 (6.2) 5 (23.8) Spontaneous miscarriage 4 (19.0) Still birth 1 (4.8) (37 weeks) 6 (37.5) Asherman’s syndrome (4) Spontaneous uterine rupture (2) — — Cesarean Sect. 16 (100.0) Vaginal delivery 0  Kishi, 2014 [ 30 ] Adenomyomectom without GnRH-a (102) 42 (41.2) — 32 (31.4) 28 (87.5) 4 (12.5) 10 (23.8) Spontaneous miscarriage 10 (23.8) 4 (12.5) Placenta accreta (2) Threatened preterm birth (2) — — Cesarean Sect. 32 (100.0) Vaginal delivery 0  Tamura, 2017 [ 31 ] Adenomyomectomy without GnRH-a (89) 37 (41.6) Spontaneous conception 13 (35.1) IVF/ET 24 (64.9) — 12 (32.4) Spontaneous miscarriage 12 (32.4) — — — —  Gao, 2019 [ 32 ] Adenomyomectomy with LNG-IUS (2) 1 (50.0) Spontaneous conception 0 IVF/ET 1 (100.0) 1 (50.0) 0 1 (100.0) 0 1 (100.0) Placenta previa and gestational hypertension (1) — 36 months (1) —  Huang, 2020 [ 33 ] Adenomyomectomy with GnRH-a (43) 13 (30.2) Spontaneous conception 8 (61.5) IVF/ET 5 (38.5) 12 (27.9) — — 1 (7.7) Spontaneous miscarriage 1 (7.7) 3 (25.0) Fetal distress (2) Placenta accreta (1) — — Cesarean Sect. 5 (41.7) Vaginal delivery 7 (58.3)  Shi, 2021 [ 34 ] Adenomyomectomy with/without GnRH-a (176) 142 (80.7) Spontaneous conception 36 (25.4) IVF/ET 106 (74.6) 97 (55.1) 81 (83.5) 16 (16.5) 47 (32.6) Spontaneous miscarriage 40 (27.8) Ectopic pregnancies 2 (1.4) Stillbirth 5 (3.4) 12 (12.4) Placenta previa (7) Placenta accreta (4) Placenta increta (1) — 12.89 ± 8.66 months (97) —  Hlinecka, 2022 [ 35 ] Adenomyomectomy without GnRH-a (28) 21 (75.0) Spontaneous conception 6 (28.6) IVF/ET 15 (71.4) 13 (46.4) — — 8 (38.1) Spontaneous miscarriage 6 (28.6) 1 (7.7) Placenta previa (1) — — Cesarean Sect. 12 (92.3) Vaginal delivery 1 (7.7)  Zhou, 2022 [ 36 ] Adenomyomectomy with GnRH-a (137) 62 (45.3) Spontaneous conception 35 (56.5) IVF/ET 27 (43.5) 45 39 6 (Ongoing pregnancies 3) 14 Spontaneous miscarriage 14 — — — Cesarean Sect. 44 (97.8) Vaginal delivery 1 (2.2)  Hijazi, 2022 [ 37 ] Adenomyomectomy with/without GnRH-a (7) 3 (42.9) — 3 (42.9) 2 (66.7) 1 (33.3) 0 1 (33.3) Cervical incompetence (1) — 12 months (3) Cesarean Sect. 3 (100.0) Vaginal delivery 0 UAE ( N  = 1): Kim, 2005 [ 38 ] UAE (6) 5 (83.3) — 5 (83.3) 4 (80.0) 1 (20.0) 0 1 (20.0) Premature rupture of membrane (1) 2820 (1850–3400) 18 (3–34) months (5) Cesarean Sect. 2 (40.0) Vaginal delivery 3 (60.0) Image-guided thermal ablation ( N  = 13):  Lee, 2015 [ 39 ] HIFU (299*) 6 (2.0) — 1 1 0 (Ongoing pregnancies 3) 2 Spontaneous miscarriage 2 — — — Cesarean Sect. 0 Vaginal delivery 1 (100.0)  Jayaram, 2016 [ 40 ] HIFU (13) 2 (15.4) Spontaneous conception 0 IVF/ET 2 (100.0) 2 (15.4) 2 (100.0) 0 0 — — 12 months (1), 18 months (1) Cesarean Sect. 1 (50.0) Vaginal delivery 1 (50.0)  Zhou, 2016 [ 41 ] HIFU (68) 54 (79.4) — 23 (33.8) 23 (100.0) 0 31 (57.4) Spontaneous miscarriage 20 (37.0) Termination of pregnancy 11 (20.4) 0 — 10 (1–31) months (23) Cesarean Sect. 18 (78.3) Vaginal delivery 5 (21.7)  Guo, 2018 [ 42 ] HIFU (45*) 3 (6.6) — 1 (2.2) 1 (100.0) 0 2 (66.7) Spontaneous miscarriage 2 (66.7) 0 — 6 months (1) — HIFU with GnRH-a (18*) 1 (5.5) — 0 0 0 1 (100.0) Spontaneous miscarriage 1 (100.0) — — — —  Sun, 2019 [ 43 ] HIFU with GnRH-a (29) 17 (58.6) — 8 (27.6) — — 9 (52.9) Spontaneous miscarriage 0 Termination of pregnancy 9 (52.9) 0 — — —  Jiang, 2019 [ 44 ] HIFU (50*) 4 (8.0) — 1 (2.0) — — 3 (75.0) Spontaneous miscarriage 1 (25.0) Termination of pregnancy 2 (50.0) — — — Cesarean Sect. 0 Vaginal delivery 1 (100.0) HIFU with GnRH-a (50*) 5 (10.0) — 3 (6.0) — — 2 (40.0) Spontaneous miscarriage 2 (40.0) — — — Cesarean Sect. 0 Vaginal delivery 3 (100.0)  Huang, 2020 [ 33 ] HIFU with GnRH-a (50) 26 (52.0) Spontaneous conception 20 (76.9) IVF/ET 6 (23.1) 18 (36.0) 17 (94.4) 1 (5.6) 8 (30.8) Spontaneous miscarriage 6 (23.1) Termination of pregnancy 2 (7.7) 5 (27.8) Placenta accreta (2) Postpartum Hemorrhage (2) Premature rupture of membranes (1) — 10 (7–31) months (18) Cesarean Sect. 12 (66.7) Vaginal delivery 6 (33.3)  Jeng, 2020 [ 45 ] HIFU (202*) 4 (2.0) — 1 — — (Ongoing pregnancies 2) 1 Spontaneous miscarriage 1 — — — Cesarean Sect. 1 (100.0) Vaginal delivery 0  Li, 2021 [ 46 ] HIFU with/without GnRH-a (597*) 50 (8.4) — — — — — — —  Huang, 2022 [ 47 ] HIFU with GnRH-a (14*) 1 (7.1) — 1 (7.1) — — 0 — — 8 months (1) —  Nam, 2020 [ 48 ] RFA (58) 29 (50.0) Spontaneous conception 25 (86.2) IVF/ET 4 (13.8) 22 (37.9) 20 (90.9) 2 (9.1) 8 (27.6) Spontaneous miscarriage 8 (27.6) 5 (22.7) Placenta previa (1) Postpartum hemorrhage (2) Malpresentation (1) Gestational hypertension (1) 3030 (1300–3700) 13.4 (1–59) months (22) Cesarean Sect. 14 (63.6) Vaginal delivery 8 (36.4)  Stepniewska, 2022 [ 49 ] RFA (60*) 2 (3.3) Spontaneous conception 2 (100.0) IVF/ET 0 2 (3.3) 1 (50.0) 1 (50.0) 0 0 — 36 months (1), 53 months (1) Cesarean Sect. 1 (50.0) Vaginal delivery 1 (50.0)  Yang, 2015 [ 50 ] MWA (39*) 6 (15.4) Spontaneous conception 6 (100.0) IVF/ET 0 0 0 0 (Ongoing pregnancy 1) 6 Spontaneous miscarriage (1) Termination of pregnancy (4) Cervical pregnancy (1) — — — — *: These articles did not specify whether the number of participants referred to the number of total participants or solely the participants preparing for pregnancy, and these articles were not included in the meta-analysis for pregnancy rate and delivery rate “—” means data deficiency. If any ongoing pregnancy was reported in the study, the delivery rate, pregnancy loss rate, and preterm delivery rate were not calculated GnRH-a Gonadotrophin-releasing hormone agonist, UAE Uterine artery embolization, HIFU High-intensity focused ultrasound, RFA Radiofrequency ablation, MWA Microwave ablation, IVF-ET In vitro fertilization and embryo transfer Reproductive outcomes of the included studies ( N  = 32) 2 (50.0) Spontaneous conception 2 (100.0) IVF/ET 0 1 1 0 (Ongoing pregnancy 1) Cesarean Sect. 1 (100.0) Vaginal delivery 0 3 (30.0) — 1 1 0 (Ongoing pregnancies 2) — Malpresentation (1) Cesarean Sect. 1 (100.0) Vaginal delivery 0 35 (79.5) — 32 (72.7) 27 (84.4) 5 (15.6) 3 (8.6) Spontaneous miscarriage 3 (8.6) 20 (74.1) — 17 (63.0) 15 (88.2) 2 (11.8) 3 (15.0) Spontaneous miscarriage 3 (15.0) 13 (46.4) Spontaneous conception 13 (100.0) IVF/ET 0 9 (32.1) — — 4 (30.8) Spontaneous miscarriage 4 (30.8) 2 (66.7) — 2 (66.7) 2 (100.0) 0 Cesarean Sect. 1 (50.0) Vaginal delivery 1 (50.0) 8 (44.4) Spontaneous conception 8 (100.0) IVF/ET 0 6 (33.3) 6 (100.0) 0 2 (25.0) Spontaneous miscarriage 2 (25.0) 1 (16.7) Retained placenta (1) Cesarean Sect. 6 (100.0) Vaginal delivery 0 16 (61.5) Spontaneous conception 4 (25.0) IVF/ET 12 (75.0) 14 (53.8) 14 (100.0) 0 2 (12.5) Spontaneous miscarriage 2 (12.5) Cesarean Sect. 14 (100.0) Vaginal delivery 0 8 (33.3) Spontaneous conception 3 (37.5) IVF/ET 5 (62.5) 3 (12.5) — — 5 (62.5) Spontaneous miscarriage 5 (62.5) 3 (33.3) Spontaneous conception 3 (100.0) IVF/ET 0 2 (22.2) — — 1 (33.3) Spontaneous miscarriage 1 (33.3) Cesarean Sect. 2 (100.0) Vaginal delivery 0 23 (41.1) Spontaneous conception 23 (100.0) IVF/ET 0 15 (26.8) 13 (86.7) 2 (13.3) 12 (52.1) Spontaneous miscarriage 4 (17.4) Ectopic pregnancy 1 (4.3) Termination of pregnancy 7 (30.4) 21 (30.0) Spontaneous conception 7 (33.3) IVF/ET 14 (66.7) 16 (22.9) 15 (93.8) 1 (6.2) 5 (23.8) Spontaneous miscarriage 4 (19.0) Still birth 1 (4.8) (37 weeks) 6 (37.5) Asherman’s syndrome (4) Spontaneous uterine rupture (2) Cesarean Sect. 16 (100.0) Vaginal delivery 0 42 (41.2) — 32 (31.4) 28 (87.5) 4 (12.5) 10 (23.8) Spontaneous miscarriage 10 (23.8) 4 (12.5) Placenta accreta (2) Threatened preterm birth (2) Cesarean Sect. 32 (100.0) Vaginal delivery 0 37 (41.6) Spontaneous conception 13 (35.1) IVF/ET 24 (64.9) 12 (32.4) Spontaneous miscarriage 12 (32.4) 1 (50.0) Spontaneous conception 0 IVF/ET 1 (100.0) 1 (50.0) 0 1 (100.0) 1 (100.0) Placenta previa and gestational hypertension (1) 13 (30.2) Spontaneous conception 8 (61.5) IVF/ET 5 (38.5) 12 (27.9) — — 1 (7.7) Spontaneous miscarriage 1 (7.7) 3 (25.0) Fetal distress (2) Placenta accreta (1) Cesarean Sect. 5 (41.7) Vaginal delivery 7 (58.3) 142 (80.7) Spontaneous conception 36 (25.4) IVF/ET 106 (74.6) 97 (55.1) 81 (83.5) 16 (16.5) 47 (32.6) Spontaneous miscarriage 40 (27.8) Ectopic pregnancies 2 (1.4) Stillbirth 5 (3.4) 12 (12.4) Placenta previa (7) Placenta accreta (4) Placenta increta (1) 21 (75.0) Spontaneous conception 6 (28.6) IVF/ET 15 (71.4) 13 (46.4) — — 8 (38.1) Spontaneous miscarriage 6 (28.6) 1 (7.7) Placenta previa (1) Cesarean Sect. 12 (92.3) Vaginal delivery 1 (7.7) 62 (45.3) Spontaneous conception 35 (56.5) IVF/ET 27 (43.5) 45 39 6 (Ongoing pregnancies 3) 14 Spontaneous miscarriage 14 Cesarean Sect. 44 (97.8) Vaginal delivery 1 (2.2) 3 (42.9) — 3 (42.9) 2 (66.7) 1 (33.3) 1 (33.3) Cervical incompetence (1) Cesarean Sect. 3 (100.0) Vaginal delivery 0 5 (83.3) — 5 (83.3) 4 (80.0) 1 (20.0) 1 (20.0) Premature rupture of membrane (1) Cesarean Sect. 2 (40.0) Vaginal delivery 3 (60.0) 6 (2.0) — 1 1 0 (Ongoing pregnancies 3) 2 Spontaneous miscarriage 2 Cesarean Sect. 0 Vaginal delivery 1 (100.0) 2 (15.4) Spontaneous conception 0 IVF/ET 2 (100.0) 2 (15.4) 2 (100.0) 0 Cesarean Sect. 1 (50.0) Vaginal delivery 1 (50.0) 54 (79.4) — 23 (33.8) 23 (100.0) 0 31 (57.4) Spontaneous miscarriage 20 (37.0) Termination of pregnancy 11 (20.4) Cesarean Sect. 18 (78.3) Vaginal delivery 5 (21.7) 3 (6.6) — 1 (2.2) 1 (100.0) 0 2 (66.7) Spontaneous miscarriage 2 (66.7) 1 (5.5) — 0 0 0 1 (100.0) Spontaneous miscarriage 1 (100.0) 17 (58.6) — 8 (27.6) — — 9 (52.9) Spontaneous miscarriage 0 Termination of pregnancy 9 (52.9) 4 (8.0) — 1 (2.0) — — 3 (75.0) Spontaneous miscarriage 1 (25.0) Termination of pregnancy 2 (50.0) Cesarean Sect. 0 Vaginal delivery 1 (100.0) 5 (10.0) — 3 (6.0) — — 2 (40.0) Spontaneous miscarriage 2 (40.0) Cesarean Sect. 0 Vaginal delivery 3 (100.0) 26 (52.0) Spontaneous conception 20 (76.9) IVF/ET 6 (23.1) 18 (36.0) 17 (94.4) 1 (5.6) 8 (30.8) Spontaneous miscarriage 6 (23.1) Termination of pregnancy 2 (7.7) 5 (27.8) Placenta accreta (2) Postpartum Hemorrhage (2) Premature rupture of membranes (1) Cesarean Sect. 12 (66.7) Vaginal delivery 6 (33.3) 4 (2.0) — 1 — — (Ongoing pregnancies 2) 1 Spontaneous miscarriage 1 Cesarean Sect. 1 (100.0) Vaginal delivery 0 50 (8.4) — 1 (7.1) — 1 (7.1) — — 29 (50.0) Spontaneous conception 25 (86.2) IVF/ET 4 (13.8) 22 (37.9) 20 (90.9) 2 (9.1) 8 (27.6) Spontaneous miscarriage 8 (27.6) 5 (22.7) Placenta previa (1) Postpartum hemorrhage (2) Malpresentation (1) Gestational hypertension (1) Cesarean Sect. 14 (63.6) Vaginal delivery 8 (36.4) 2 (3.3) Spontaneous conception 2 (100.0) IVF/ET 0 2 (3.3) 1 (50.0) 1 (50.0) Cesarean Sect. 1 (50.0) Vaginal delivery 1 (50.0) 6 (15.4) Spontaneous conception 6 (100.0) IVF/ET 0 0 0 0 (Ongoing pregnancy 1) 6 Spontaneous miscarriage (1) Termination of pregnancy (4) Cervical pregnancy (1) *: These articles did not specify whether the number of participants referred to the number of total participants or solely the participants preparing for pregnancy, and these articles were not included in the meta-analysis for pregnancy rate and delivery rate “—” means data deficiency. If any ongoing pregnancy was reported in the study, the delivery rate, pregnancy loss rate, and preterm delivery rate were not calculated GnRH-a Gonadotrophin-releasing hormone agonist, UAE Uterine artery embolization, HIFU High-intensity focused ultrasound, RFA Radiofrequency ablation, MWA Microwave ablation, IVF-ET In vitro fertilization and embryo transfer An assessment of the risk of bias and evidence quality for each article was conducted using the Newcastle-Ottawa Scale quality assessment tool, evaluating selection, comparability, and outcome (Table  4 ). Of these articles, eight (25.0%) were classified as high quality, 23 (71.9%) were classified as medium quality, and only one (3.1%) was classified as poor quality. The primary factors diminishing the quality of studies were: (1) comparability of cases and controls: most articles (53.1%) were clinical single-arm studies lacking a control group; (2) length of follow-up duration: a short follow-up period of less than 24 months in 21.9% of the articles was insufficient for reproductive outcomes to manifest; (3) lost to follow-up: a rate of loss to follow-up exceeding 20% was considered to impact the quality of the studies; (4) risk of selection bias. Table 4 Quality assessment with Newcastle-Ottawa scale quality assessment tool First author, year Intervention Selection Comparability Outcome Quality assessment Fujishita, 2004 [ 19 ] Adenomyomectomy ↔↔↔ ↔ ↔↔ Medium Hadisaputra, 2006 [ 20 ] Adenomyomectomy ↔↔↔↔ ↔ ↔ Medium Wang, 2009 [ 21 ] Adenomyomectomy ↔↔↔↔ ↔↔ ↔↔ High Wang, 2009 [ 22 ] Adenomyomectomy ↔↔↔ ↔↔ ↔↔ High Takeuchi, 2010 [ 23 ] Adenomyomectomy ↔↔ — ↔↔ Medium AI Jama, 2011 [ 24 ] Adenomyomectomy ↔↔↔↔ ↔↔ ↔↔ High Osada, 2011 [ 25 ] Adenomyomectomy ↔↔↔ — ↔↔ Medium Sun, 2011 [ 26 ] Adenomyomectomy ↔↔↔↔ ↔ ↔↔ High Huang, 2012 [ 27 ] Adenomyomectomy ↔↔↔ — ↔↔ Medium Chang, 2013 [ 28 ] Adenomyomectomy ↔↔↔ — ↔↔ Medium Saremi, 2014 [ 29 ] Adenomyomectomy ↔↔↔ — ↔↔ Medium Kishi, 2014 [ 30 ] Adenomyomectomy ↔↔↔ — ↔↔ Medium Tamura, 2017 [ 31 ] Adenomyomectomy ↔ ↔ ↔ Low Gao, 2019 [ 32 ] Adenomyomectomy ↔↔↔ — ↔ Medium Huang, 2020 [ 33 ] Adenomyomectomy ↔↔↔↔ ↔↔ ↔↔ High Shi, 2021 [ 34 ] Adenomyomectomy ↔↔↔ — ↔ Medium Hlinecka, 2022 [ 35 ] Adenomyomectomy ↔↔↔ ↔ ↔↔ Medium Zhou, 2022 [ 36 ] Adenomyomectomy ↔↔↔ — ↔↔ Medium Hijazi, 2022[ 37 ] Adenomyomectomy ↔↔↔ — ↔↔ Medium Kim, 2005 [ 38 ] UAE ↔↔↔ — ↔↔ Medium Lee, 2015 [ 39 ] Image-guided thermal ablation (HIFU) ↔↔↔ — ↔ Medium Jayaram, 2016 [ 40 ] Image-guided thermal ablation (HIFU) ↔↔↔ — ↔ Medium Zhou, 2016 [ 41 ] Image-guided thermal ablation (HIFU) ↔↔↔ — ↔↔ Medium Guo, 2018 [ 42 ] Image-guided thermal ablation (HIFU) ↔↔↔↔ ↔↔ ↔ High Sun, 2019 [ 43 ] Image-guided thermal ablation (HIFU) ↔↔↔ — ↔↔ Medium Jiang, 2019 [ 44 ] Image-guided thermal ablation (HIFU) ↔↔↔↔ ↔↔ ↔ High Huang, 2020 [ 33 ] Image-guided thermal ablation (HIFU) ↔↔↔↔ ↔↔ ↔↔ High Jeng, 2020 [ 45 ] Image-guided thermal ablation (HIFU) ↔↔↔ ↔ — Medium Li, 2021 [ 46 ] Image-guided thermal ablation (HIFU) ↔↔↔↔ ↔↔ ↔ High Huang, 2022 [ 47 ] Image-guided thermal ablation (HIFU) ↔↔↔ ↔ ↔ Medium Nam, 2020 [ 48 ] Image-guided thermal ablation (RFA) ↔↔↔ — ↔↔ Medium Stepniewska, 2022 [ 49 ] Image-guided thermal ablation (RFA) ↔↔↔ — ↔↔ Medium Yang, 2015 [ 50 ] Image-guided thermal ablation (MWA) ↔↔↔ — ↔ Medium All included studies were classified as high (7–9 stars), medium (4–6 stars), or low (0–3 stars) evidence quality according to selection, comparability, and outcome. “—” means 0 star Quality assessment with Newcastle-Ottawa scale quality assessment tool All included studies were classified as high (7–9 stars), medium (4–6 stars), or low (0–3 stars) evidence quality according to selection, comparability, and outcome. “—” means 0 star Table  5 presents the synthesized reproductive outcomes following these fertility-sparing interventions for symptomatic adenomyosis. Tables  6 and 7 provide the results of subgroup analysis and sensitivity analysis, respectively. Figure  2 illustrates the histogram of the meta-analysis for reproductive outcomes. Figures  3 , 4 , 5 , 6 , 7 , 8 , 9 and 10 display the forest plots of the meta-analysis for reproductive outcomes. Table 5 Results of Meta-analysis for reproductive outcomes after different fertility-sparing interventions Outcome Measures No. of studies Pooled proportion (%) (95% CI) P value I², % No. of patients Pregnancy rate  Adenomyomectomy 19 50.1 (40.0—60.2) < 0.001 89.0 903  Image-guided thermal ablation 5 52.0 (32.4—71.6) < 0.001 89.8 218  Overall 23 50.5 (41.9—59.2) < 0.001 88.7 1121 Delivery rate  Adenomyomectomy 15 39.5 (29.9—49.2) < 0.001 83.7 663  Image-guided thermal ablation (HIFU/RFA) 5 32.5 (26.0—38.9) 0.367 7.0 218  Overall 19 37.1 (29.7—44.5) < 0.001 79.9 881 Pregnancy loss rate  Adenomyomectomy 16 19.8 (12.2—27.5) < 0.001 88.3 408  Image-guided thermal ablation (HIFU/RFA) 9 39.5 (13.8—65.1) < 0.001 98.9 144  Overall 24 28.3 (17.2—39.4) < 0.001 97.5 552 Spontaneous miscarriage rate  Adenomyomectomy 16 16.3 (9.7—22.9) < 0.001 84.4 408  Image-guided thermal ablation (HIFU/RFA) 9 27.1 (8.1—46.1) < 0.001 99.0 144  Overall 24 21.4 (12.5—30.3) < 0.001 97.6 552 Rate of adverse pregnancy outcomes  Adenomyomectomy 11 21.4 (7.5—35.3) < 0.001 99.0 211  Image-guided thermal ablation (HIFU/RFA) 6 1.0 (−1.6—3.7) 0.022 62.1 74  Overall 16 15.3 (7.9—22.7) < 0.001 98.4 285 Preterm delivery rate  Adenomyomectomy 10 18.4 (2.9—33.9) < 0.001 99.0 235  Image-guided thermal ablation (HIFU/RFA) 6 0.3 (−1.0—1.7) 0.404 1.9 68  Overall 16 13.5 (5.0—22.1) < 0.001 98.4 303 IVF-ET conception rate  Adenomyomectomy 14 40.5 (28.8—52.1) < 0.001 99.2 370  Image-guided thermal ablation 5 27.5 (−17.3—72.2) < 0.001 99.7 65  Overall 18 37.5 (25.2—49.8) < 0.001 99.5 435 Cesarean section rate  Adenomyomectomy 12 99.6 (98.3—100.8) 0.038 46.6 147  Image-guided thermal ablation (HIFU/RFA) 9 44.6 (13.4—75.9) < 0.001 99.1 73  Overall 20 73.5 (61.2—85.7) < 0.001 99.6 220 UAE Uterine artery embolization, HIFU High-intensity focused ultrasound, RFA Radiofrequency ablation, MWA Microwave ablation, IVF-ET In vitro fertilization and embryo transfer Table 6 Results of subgroup analysis for adenomyomectomy group Outcome Measures No. of studies Pooled proportion (%) (95% CI) P value I², % No. of patients Subgroup analysis of pregnancy rate  Adenomyomectomy (Laparotomy) 6 41.5 (30.0—53.0) 0.118 43.1 167  Adenomyomectomy (Laparoscopy) 7 48.4 (27.6—69.1) < 0.001 92.8 394  Adenomyomectomy with adjuvant hormone therapy 9 43.7 (31.0—56.5) < 0.001 79.7 343  Adenomyomectomy without adjuvant hormone therapy 6 52.7 (38.6—66.8) < 0.001 86.3 342  Adenomyomectomy (average/median age < 35 years) 7 50.0 (29.0—71.1) < 0.001 91.0 333  Adenomyomectomy (average/median age ≥ 35 years) 12 50.1 (39.7—60.4) < 0.001 83.3 570  Adenomyomectomy (Focal) 5 58.3 (37.2—79.3) < 0.001 86.2 154  Adenomyomectomy (Diffuse) 4 47.3 (39.7—54.8) 0.357 7.2 200  Adenomyomectomy (average diameter of lesion < 50 mm) 4 61.9 (39.4—84.4) 0.004 77.7 82  Adenomyomectomy (average diameter of lesion ≥ 50 mm) 5 58.6 (38.0—79.1) < 0.001 93.2 374  Adenomyomectomy (year of publication before 2012) 9 53.0 (39.7—66.3) < 0.001 71.9 193  Adenomyomectomy (year of publication after 2012) 10 47.9 (33.2—62.5) < 0.001 93.5 710 Subgroup analysis of delivery rate  Adenomyomectomy (Laparotomy) 5 31.0 (19.2—42.8) 0.072 53.4 163  Adenomyomectomy (Laparoscopy) 5 42.3 (26.1—58.6) 0.006 72.4 247  Adenomyomectomy with adjuvant hormone therapy 7 33.8 (15.6—52.0) < 0.001 87.1 196  Adenomyomectomy without adjuvant hormone therapy 5 41.7 (27.8—55.7) < 0.001 80.6 253  Adenomyomectomy (average/median age < 35 years) 5 35.0 (12.6—57.4) < 0.001 88.7 240  Adenomyomectomy (average/median age ≥ 35 years) 10 41.6 (30.5—52.8) < 0.001 81.6 423  Adenomyomectomy (Focal) 4 46.2 (19.9—72.6) < 0.001 92.4 154  Adenomyomectomy (Diffuse) 3 37.4 (19.7—55.1) 0.113 54.1 63  Adenomyomectomy (average diameter of lesion < 50 mm) 4 44.1 (15.7—72.5) < 0.001 87.0 82  Adenomyomectomy (average diameter of lesion ≥ 50 mm) 3 60.3 (45.5—75.2) 0.050 66.6 227  Adenomyomectomy (year of publication before 2012) 7 43.5 (25.1—61.8) < 0.001 86.3 179  Adenomyomectomy (year of publication after 2012) 8 35.8 (24.6—47.0) < 0.001 81.9 484 Subgroup analysis of pregnancy loss rate  Adenomyomectomy (Laparotomy) 5 21.9 (1.8—42.0) < 0.001 85.8 64  Adenomyomectomy (Laparoscopy) 5 11.4 (−1.1—23.9) < 0.001 93.6 168  Adenomyomectomy with adjuvant hormone therapy 7 22.0 (7.2—36.9) < 0.001 83.5 91  Adenomyomectomy without adjuvant hormone therapy 6 23.5 (16.3—30.8) 0.293 18.6 157  Adenomyomectomy (average/median age ≤ 35 years) 9 22.0 (10.3—33.7) 35 years) 7 18.1 (7.6—28.7) < 0.001 79.6 166  Adenomyomectomy (Focal) 4 22.6 (5.8—39.5) < 0.001 89.7 88  Adenomyomectomy (Diffuse) 3 18.8 (5.6—32.0) 0.418 0 32  Adenomyomectomy (average diameter of lesion < 50 mm) 4 25.1 (1.7—48.4) < 0.001 88.9 51  Adenomyomectomy (average diameter of lesion ≥ 50 mm) 3 13.6 (−6.8—34.1) < 0.001 96.5 180 Subgroup analysis of spontaneous miscarriage rate  Adenomyomectomy (Laparotomy) 5 11.3 (0.7—21.9) 0.051 57.6 64  Adenomyomectomy (Laparoscopy) 5 10.1(−1.0—21.3) < 0.001 92.0 168  Adenomyomectomy with adjuvant hormone therapy 7 14.0 (3.4—24.6) 0.004 69.0 91  Adenomyomectomy without adjuvant hormone therapy 6 22.0 (15.6—28.3) 0.472 0 157  Adenomyomectomy (average/median age ≤ 35 years) 9 19.8 (9.1—30.5) 35 years) 7 13.0 (5.7—20.3) 0.016 59.6 166  Adenomyomectomy (Focal) 4 13.9 (2.0—25.9) < 0.001 80.4 88  Adenomyomectomy (Diffuse) 3 18.8 (5.6—32.0) 0.418 0 32  Adenomyomectomy (average diameter of lesion < 50 mm) 4 22.1 (1.0—43.2) < 0.001 86.8 51  Adenomyomectomy (average diameter of lesion ≥ 50 mm) 3 12.0 (−5.7—29.8) < 0.001 95.5 180 Subgroup analysis of adverse pregnancy outcomes rate  Adenomyomectomy (Laparotomy) 4 34.0 (9.2—58.7) 35 years) 6 28.0 (7.6—48.3) < 0.001 99.5 84  Adenomyomectomy (average diameter of lesion < 50 mm) 2 0.7 (−3.5—4.9) 0.325 0 15  Adenomyomectomy (average diameter of lesion ≥ 50 mm) 2 12.7 (6.2—19.2) 0.446 0 100 Subgroup analysis of preterm delivery rate  Adenomyomectomy (Laparotomy) 4 30.0 (−28.7—88.6) < 0.001 99.7 46  Adenomyomectomy (Laparoscopy) 4 5.3 (−2.2—12.8) < 0.001 84.0 108  Adenomyomectomy with adjuvant hormone therapy 4 32.3 (−27.0—91.6) < 0.001 99.6 54  Adenomyomectomy without adjuvant hormone therapy 4 5.7 (−1.6—13.0) 0.060 59.5 79  Adenomyomectomy (average/median age ≤ 35 years) 3 9.8 (−5.6—25.2) 0.001 86.8 102  Adenomyomectomy (average/median age >35 years) 7 20.1 (−0.1—40.3) < 0.001 99.3 133  Adenomyomectomy (average diameter of lesion < 50 mm) 2 3.5 (−6.9—14.0) 0.150 51.6 19  Adenomyomectomy (average diameter of lesion ≥ 50 mm) 3 16.5 (10.2—22.8) 0.818 0 132 Subgroup analysis of IVF-ET conception rate  Adenomyomectomy (Laparotomy) 6 39.2 (13.6—64.7) < 0.001 99.5 66  Adenomyomectomy (Laparoscopy) 4 39.1 (−7.1—85.3) < 0.001 99.3 225  Adenomyomectomy with adjuvant hormone therapy 7 33.1 (14.7—51.5) < 0.001 99.4 118  Adenomyomectomy without adjuvant hormone therapy 4 68.7 (59.4—78.0) 0.876 0 95  Adenomyomectomy (average/median age ≤ 35 years) 9 37.6 (21.3—54.0) 35 years) 5 47.0 (22.2—71.9) < 0.001 99.6 69  Adenomyomectomy (Focal) 4 23.5 (9.8—37.3) < 0.001 96.8 94  Adenomyomectomy (Diffuse) 2 29.0 (−32.0—89.9) < 0.001 92.5 31  Adenomyomectomy (average diameter of lesion < 50 mm) 2 69.2 (52.4—85.9) 0.652 0 29  Adenomyomectomy (average diameter of lesion ≥ 50 mm) 2 59.5 (29.0—89.9) < 0.001 94.5 204 Subgroup analysis of cesarean section rate  Adenomyomectomy (Laparotomy) 4 100.0 (98.8—101.0) 1.000 0 33  Adenomyomectomy (Laparoscopy) 6 97.8 (93.2—102.3) 0.002 73.9 69 UAE Uterine artery embolization, HIFU High-intensity focused ultrasound, RFA Radiofrequency ablation, MWA Microwave ablation, IVF-ET In vitro fertilization and embryo transfer Table 7 Results of sensitivity analysis for studies with follow-up ≥ 36 months after adenomyomectomy Outcome Measures No. of studies Pooled proportion (%) (95% CI) P value I², % No. of patients Pregnancy rate 10 54.2 (40.0—68.5) < 0.001 88.1 484 Delivery rate 8 40.0 (28.8—51.2) 0.001 70.2 343 Pregnancy loss rate 8 26.8 (11.0—42.6) < 0.001 88.7 227 Spontaneous miscarriage rate 8 19.6 (7.5—31.7) < 0.001 81.6 227 Rate of adverse pregnancy outcomes 6 23.2 (3.7—42.7) < 0.001 99.5 146 Preterm delivery rate 5 26.1 (0.1—52.0) < 0.001 99.6 133 IVF-ET conception rate 10 34.6 (21.7—47.6) < 0.001 99.4 291 Fig. 2 Histogram of meta-analysis for reproductive outcomes Fig. 3 Forest plot of pooled pregnancy rates after fertility-sparing treatments for adenomyosis Fig. 4 Forest plot of pooled delivery rates after fertility-sparing treatments for adenomyosis Fig. 5 Forest plot of pooled pregnancy loss rates after fertility-sparing treatments for adenomyosis Fig. 6 Forest plot of pooled spontaneous miscarriage rates after fertility-sparing treatments for adenomyosis Fig. 7 Forest plot of pooled rates of adverse pregnancy outcomes after fertility-sparing treatments for adenomyosis Fig. 8 Forest plot of pooled preterm delivery rates after fertility-sparing treatments for adenomyosis Fig. 9 Forest plot of pooled IVF-ET conception rates after fertility-sparing treatments for adenomyosis. (IVF-ET: in vitro fertilization and embryo transfer) Fig. 10 Forest plot of pooled cesarean section rates after fertility-sparing treatments for adenomyosis Results of Meta-analysis for reproductive outcomes after different fertility-sparing interventions UAE Uterine artery embolization, HIFU High-intensity focused ultrasound, RFA Radiofrequency ablation, MWA Microwave ablation, IVF-ET In vitro fertilization and embryo transfer Results of subgroup analysis for adenomyomectomy group UAE Uterine artery embolization, HIFU High-intensity focused ultrasound, RFA Radiofrequency ablation, MWA Microwave ablation, IVF-ET In vitro fertilization and embryo transfer Results of sensitivity analysis for studies with follow-up ≥ 36 months after adenomyomectomy Histogram of meta-analysis for reproductive outcomes Forest plot of pooled pregnancy rates after fertility-sparing treatments for adenomyosis Forest plot of pooled delivery rates after fertility-sparing treatments for adenomyosis Forest plot of pooled pregnancy loss rates after fertility-sparing treatments for adenomyosis Forest plot of pooled spontaneous miscarriage rates after fertility-sparing treatments for adenomyosis Forest plot of pooled rates of adverse pregnancy outcomes after fertility-sparing treatments for adenomyosis Forest plot of pooled preterm delivery rates after fertility-sparing treatments for adenomyosis Forest plot of pooled IVF-ET conception rates after fertility-sparing treatments for adenomyosis. (IVF-ET: in vitro fertilization and embryo transfer) Forest plot of pooled cesarean section rates after fertility-sparing treatments for adenomyosis The pregnancy rates were 50.1% (95% CI, 40.0% to 60.2%, P  < 0.001; I²=89.0%; 19 studies) following adenomyomectomy and 52.0% (95% CI, 32.4% to 71.6%, P  < 0.001; I ²=89.8%; 5 studies) following thermal ablation. The overall pregnancy rate after these treatments was 50.5% (95% CI, 41.9% to 59.2%, P  < 0.001, I ²=88.7%; 23 studies). Certainty of evidence: high. (Table  5 ; Fig.  3 ) The delivery rates were 39.5% (95% CI, 29.9% to 49.2%, P  < 0.001; I²=83.7%; 15 studies) in adenomyomectomy group and 32.5% (95% CI, 26.0% to 38.9%, P  = 0.367; I²=7.0%; 5 studies) in thermal ablation group. The overall delivery rate was 37.1% (95% CI, 29.7% to 44.5%, P  < 0.001, I²=79.9%; 19 studies) after these treatments. The certainty of evidence was low due to ongoing pregnancy reported in six studies (with missing results), preventing an exact calculation of the delivery rate. (Table  5 ; Fig.  4 ) The pregnancy loss rates were 19.8% (95% CI, 12.2% to 27.5%, P  < 0.001; I²=88.3%; 16 studies) for adenomyomectomy and 39.5% (95% CI, 13.8% to 65.1%, P  < 0.001; I²=98.9%; 9 studies) for thermal ablation. The overall pregnancy loss rate after these uterine-sparing treatments was 28.3% (95% CI, 17.2% to 39.4%, P  < 0.001, I²=97.5%; 24 studies). Certainty of evidence: medium. (Table  5 ; Fig.  5 ) The spontaneous miscarriage rates were 16.3% (95% CI, 9.7% to 22.9%, P  < 0.001; I²=84.4%; 16 studies) following adenomyomectomy and 27.1% (95% CI, 8.1% to 46.1%, P  < 0.001; I²=99.0%; 9 studies) following thermal ablation. The overall spontaneous miscarriage rate after these treatments was 21.4% (95% CI, 12.5% to 30.3%, P  < 0.001, I²=97.6%; 24 studies). Certainty of evidence: medium. (Table  5 ; Fig.  6 ) The rates of adverse pregnancy outcomes were 21.4% (95% CI, 7.5% to 35.3%, P  < 0.001; I²=99.0%; 11 studies) in adenomyomectomy group and 1.0% (95% CI, −1.6% to 3.7%, P  = 0.022; I²=62.1%; 6 studies) in thermal ablation group. The overall rate of adverse pregnancy outcomes was 15.3% (95% CI, 7.9% to 22.7%, P  < 0.001, I²=98.4%; 16 studies) after these treatments. Certainty of evidence: medium. (Table  5 ; Fig.  7 ) The preterm delivery rates were 18.4% (95% CI, 2.9% to 33.9%, P  < 0.001; I²=99.0%; 10 studies) for adenomyectomy and 0.3% (95% CI, −1.0% to 1.7%, P  = 0.404; I²=1.9%; 6 studies) for thermal ablation. The overall preterm delivery rate after these treatments was 13.5% (95% CI, 5.0% to 22.1%, P  < 0.001, I²=98.4%; 16 studies). Certainty of evidence: medium. (Table  5 ; Fig.  8 ) The IVF-ET conception rates were 40.5% (95% CI, 28.8% to 52.1%, P  < 0.001; I²=99.2%; 14 studies) after adenomyomectomy and 27.5% (95% CI, −17.3% to 72.2%, P  < 0.001; I²=99.7%; 5 studies) after thermal ablation. The overall IVF-ET conception rate after these treatments was 37.5% (95% CI, 25.2% to 49.8%, P  < 0.001, I²=99.5%; 18 studies). Certainty of evidence: medium. (Table  5 ; Fig.  9 ) The cesarean section rates were 99.6% (95% CI, 98.3% to 100.8%, P  = 0.038; I²=46.6%; 12 studies) in adenomyomectomy group and 44.6% (95% CI, 13.4% to 75.9%, P  < 0.001; I²=99.1%; 9 studies) in thermal ablation group. The overall cesarean section rate was 73.5% (95% CI, 61.2% to 85.7%, P  < 0.001, I²=99.6%; 20 studies) after these treatments. Certainty of evidence: medium. (Table  5 ; Fig.  10 ) We performed subgroup analysis when the results indicated significant heterogeneity (I² statistic > 50%). However, subgroup analyses were limited to the adenomyomectomy group due to the scarcity of studies in the UAE and thermal ablation groups. The subgroup analyses were conducted according to surgical approach, whether participants received adjuvant hormone therapy, age of participants, type of adenomyosis, size of adenomyosis, and year of publication. The subgroup analysis results demonstrated a significant reduction in heterogeneity (I² statistic ≤ 50%) for the following outcomes: pregnancy rates for laparotomy and diffuse adenomyosis, pregnancy loss rates for adenomyomectomy without adjuvant hormone therapy and for diffuse adenomyosis, spontaneous miscarriage rates for adenomyomectomy without adjuvant hormone therapy and for diffuse adenomyosis, adverse pregnancy outcomes rates for average diameter of adenomyosis lesion < 50 mm and ≥ 50 mm, preterm delivery rate for average diameter of lesion ≥ 50 mm, IVF-ET conception rates for adenomyomectomy without adjuvant hormone therapy and for lesion 50%) persisted in the remaining results of the subgroup analysis. (Table  6 ) We performed sensitivity analyses for studies with a follow-up duration ≥ 36 months by excluding those with a follow-up duration < 36 months in the adenomyomectomy group. The results indicated that the pregnancy rate was 54.2% (95% CI, 40.0% to 68.5%, P  < 0.001; I²=88.1%; 10 studies). The delivery rate was 40.0% (95% CI, 28.8% to 51.2%, P  = 0.001; I²=70.2%; 8 studies). The pregnancy loss rate was 26.8% (95% CI, 11.0% to 42.6%, P  < 0.001; I²=88.7%; 8 studies). The spontaneous miscarriage rate was 19.6% (95% CI, 7.5% to 31.7%, P  < 0.001; I²=81.6%; 8 studies). The rate of adverse pregnancy outcomes was 23.2% (95% CI, 3.7% to 42.7%, P  < 0.001; I²=99.5%; 6 studies). The preterm delivery rate was 26.1% (95% CI, 0.1% to 52.0%, P  < 0.001; I²=99.6%; 5 studies). The IVF-ET conception rate was 34.6% (95% CI, 21.7% to 47.6%, P  50%). We did not perform a sensitivity analysis for the cesarean section rate because the result exhibited low heterogeneity (I² statistic ≤ 50%). (Table  7 )

This study protocol was registered in the international prospective register of systematic reviews (PROSPERO) online database (ID: CRD42020199586). The meta-analysis was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and MOOSE guidelines. Electronic databases including Medline, Embase, Cochrane Library, Web of Science, ClinicalTrials.gov, and Google Scholar were searched from January 2000 to July 2023. The following search strategy was employed on Medline: (adenomyosis [Title/Abstract]) AND ((Reproduction [MeSH]) OR (pregnancy [Title/Abstract]) OR (fertility [Title/Abstract]) OR (conceive [Title/Abstract]) OR (gestation [Title/Abstract]) OR (fetation [Title/Abstract])). Reference lists of relevant review articles and all identified articles were reviewed to identify additional studies. We considered studies of any design (randomized controlled trials, cohort studies, case-control studies, and cross-sectional studies) that reported reproductive outcomes following fertility-sparing interventions (adenomyomectomy, UAE, and image-guided thermal ablation) for symptomatic adenomyosis. However, case reports, case series, review articles, animal studies, commentaries, and meeting reports were excluded. Only articles reporting the number of participants desiring fertility, rather than the total number of participants (including those with no desire for children), were included. In studies comparing the aforementioned fertility-sparing interventions with other treatments, data from the single fertility-sparing intervention group were included. Repeated studies and those not reporting any reproductive outcomes were excluded. We only included articles written in English or Chinese. Articles written in other languages were also excluded. According to a standardized core outcome set and outcome definitions for research on uterus-sparing treatments of adenomyosis (based on an international multistakeholder-modified Delphi consensus study) [ 18 ], the following reproductive outcomes after interventions were recorded: numbers of pregnancy, delivery, pregnancy loss, spontaneous miscarriage, intentional termination of pregnancy, stillbirth, complications or adverse pregnancy outcomes during pregnancy or delivery, gestational age at delivery, birthweight, mode of conception, time to pregnancy leading to live birth, and delivery method. The pregnancy rate was defined as the percentage of participants who achieved a live and correctly sited (eutopic) pregnant following treatments. Similarly, the delivery rate was defined as the percentage of participants who successfully delivered (live birth) after treatments. The pregnancy/delivery rate = number of participants who became live and correctly sited pregnant/number of participants with live births ÷ total number of participants preparing for pregnancy after treatments × 100%. If any ongoing pregnancy was reported in the study, we did not calculate the delivery rate. Pregnancy loss included ectopic pregnancy, spontaneous miscarriage, stillbirth, and intentional termination of pregnancy. The pregnancy loss rate/spontaneous miscarriage rate = number of participants with pregnancy loss/with spontaneous miscarriage ÷ number of participants who became pregnant after treatments (including ectopic pregnancy) × 100%. If any ongoing pregnancy was reported in the study, we did not evaluate the pregnancy loss rate. Complications or adverse outcomes during pregnancy and delivery encompass neonatal death, major congenital anomaly, abnormal placentation (including placenta accreta, placenta increta, placenta previa, retained placenta, and placenta insufficiency), postpartum hemorrhage, threatened preterm birth, fetal distress, malpresentation, premature rupture of membranes, gestational hypertension, gestational diabetes, pre-eclampsia, spontaneous uterine rupture, cervical incompetence and so on. Live births are categorized based on the gestational age at delivery into term (≥ 37 weeks) and preterm (< 37 weeks) deliveries. The modes of delivery are classified into cesarean section and vaginal delivery. The rate of adverse pregnancy outcomes/preterm delivery rate/cesarean section rate = number of participants with complications or adverse pregnancy outcomes/with preterm delivery/with cesarean section ÷ number of participants with live births after treatments × 100%. If any ongoing pregnancy was reported in the study, we did not evaluate the preterm delivery rate. Modes of conception were categorized into spontaneous conception and assisted reproductive technology (i.e., in vitro fertilization and embryo transfer, IVF-ET). The IVF-ET conception rate = number of participants who conceived with assisted reproductive technology ÷ number of participants who became live and correctly sited pregnant after treatments × 100%. Study selection was conducted by three independent reviewers (LL, HT, and YL) according to the eligibility criteria. Each article was assessed by a minimum of two reviewers. The screening of titles and abstracts for each article was performed, and the potentially eligible full texts were subsequently assessed for eligibility. Any disagreement on the inclusion of an article was resolved by consulting with the corresponding author (WC). Three reviewers (LL, YL, and TW) independently extracted clinical data from each identified article using pretested data collection forms. We collected the following study characteristics: the first author, year of publication, type of intervention, adjuvant treatment, surgical approach, number of participants desiring fertility, study design, follow-up period, age of the participants, type of adenomyosis, subfertility history, size of uterus and adenomyosis, and reproductive outcomes. The reproductive outcomes included pregnancy rate, delivery rate, pregnancy loss rate, spontaneous miscarriage rate, rate of adverse pregnancy outcomes, preterm delivery rate, IVF-ET conception rate, and cesarean section rate following treatments. Two reviewers (LL and WC) independently assessed the risk of bias and evidence quality of each article using the Newcastle-Ottawa Scale quality assessment tool for observational non-randomized studies, including case-control studies, cohort studies, and cross-sectional studies. All included studies were classified as high (7–9 stars), medium (4–6 stars), or low (0–3 stars) evidence quality according to the following domains: selection, comparability, and outcome. In case of disagreement, another reviewer (HT) was consulted to obtain a consensus. (Supplement 1) We used Stata version 15.0 (Stata Corp) for meta-analysis. The reproductive outcomes from individual articles were pooled according to different types of fertility-sparing interventions (adenomyomectomy, UAE, or image-guided thermal ablation). The pregnancy rate, delivery rate, pregnancy loss rate, spontaneous miscarriage rate, rate of adverse pregnancy outcomes, preterm delivery rate, IVF-ET conception rate, and cesarean section rate were presented as the frequency with percentage, and the pooled proportions were calculated with a 95% confidence interval (CI) using the Freeman–Tukey Double arcsine transformation. The pooled prevalence of reproductive outcome was generated using the metan package, a procedure in Stata used for meta-analysis of binary and continuous data with fixed-effects or random-effects model for pooling aggregate data. Forest plots with 95% CI were also generated using this procedure. Heterogeneity originates from variations among studies. The I² statistic and its 95% CI were used to examine the between-study heterogeneity. I² statistic of ≤ 50% indicated low heterogeneity, warranting the use of a fixed-effect model for meta-analysis. Conversely, I² statistic of > 50% signified significant heterogeneity, necessitating the application of a random-effect model for meta-analysis. In instances where heterogeneity was substantial (I² statistic > 50%), subgroup analysis and sensitivity analysis were conducted to investigate potential sources of heterogeneity. Statistical significance was further evaluated using Egger’s linear regression asymmetry test, with a threshold set at P -values < 0.05. Five types of subgroup analyses were performed: (1) according to the surgical approach of adenomyomectomy (laparotomy or laparoscopy); (2) according to whether participants received perioperative or postoperative adjuvant hormone therapy; (3) according to the average/median age of participants ( 35 years); (4) according to the type of adenomyosis (focal or diffuse); (5) according to the size of adenomyosis (average diameter < 50 mm or ≥ 50 mm); (6) according to the year of publication (before or after 2012). Additionally, a sensitivity analysis was performed for studies with a follow-up duration of ≥ 36 months by excluding studies with shorter follow-up duration (< 36 months) in the adenomyomectomy group.

This meta-analysis revealed promising reproductive outcomes following fertility-sparing interventions for women with adenomyosis who desire fertility, as successful pregnancies and deliveries were observed in all treatment groups, including those undergoing UAE. A total of 17 articles reported on participants’ subfertility history prior to treatments, with reproductive outcomes significantly improving post-treatment in nearly all studies. These improvements included relatively high rates of pregnancy and delivery. This study demonstrated encouraging overall pregnancy rates (exceeding 50%) and delivery rates (exceeding 32%) following adenomyomectomy and image-guided thermal ablation. Additionally, since only one article on UAE met the inclusion criteria, we could not conduct the meta-analysis for UAE. This study reported that among the six women who underwent UAE, both the pregnancy rate and delivery rate were 83.3%. Despite significant improvements in pregnancy rates following treatments, some patients still did not achieve successful conception. This suggests that other factors or conditions besides the adenomyosis may influence pregnancy rates. Firstly, age is an important factor affecting pregnancy rates. The younger the average age of the patients, the higher the postoperative pregnancy rate. Secondly, some patients presented with infertility factors other than adenomyosis before treatments, such as endometrial diseases, reduced endometrial receptivity, and decreased ovarian function. Thirdly, adenomyosis frequently coexists with uterine fibroids, which may also impact pregnancy rates. Adenomyomectomy is the most commonly used fertility-sparing intervention for women with adenomyosis. It can be classified into complete excision (primarily for focal adenomyosis) and partial excision of the lesion (mainly for diffuse adenomyosis), and it can be performed through either laparotomy or laparoscopic approach [ 51 ]. Our results revealed favorable reproductive outcomes following treatment (pregnancy rate: 50.1%, delivery rate: 39.5%), which are consistent with previous studies [ 14 , 51 ]. Compared to medical therapy alone (such as Gonadotrophin-releasing hormone agonist, GnRH-a), adenomyomectomy provides better reproductive performance and extended pregnancy prospects post-treatment [ 22 , 24 ], suggesting that surgical removal of lesions can mitigate the adverse effects of adenomyosis on fertility. However, we observed that the cesarean section rate was as high as 99.6%, and the preterm delivery rate (18.4%) and adverse pregnancy outcomes rate (21.4%) were relatively elevated compared to thermal ablation. Although a few studies [ 33 , 36 ] have reported successful vaginal deliveries post-surgery, an elective caesarian delivery is generally considered preferable to prevent uterine rupture. Only one included article [ 29 ] reported two cases of spontaneous uterine rupture during pregnancy: one occurred at 37 weeks (resulted in a stillbirth), and the other at 32 weeks (survived). However, a review [ 52 ] clarified that pregnancies following adenomyomectomy have higher rates of uterine rupture (>1.0%) compared to other uterine surgeries, including myomectomy (0.26%) and cesarean Sect. (0.27%–0.7%). As adenomyotic tissues exhibit varying degrees of myometrial infiltration, resulting in an unclear boundary of the lesion, complete excision remains challenging and often involves the removal of normal myometrium. This invasive procedure can compromise the integrity of the uterine wall and reduce the uterus’s tensile strength, potentially increasing the risk of uterine rupture during pregnancy and delivery [ 51 ]. The potential factors related to uterine rupture include the skill of the operator, the extent of the myometrium defect, reconstruction of the uterine wall and uterine cavity, postoperative infection and hematoma formation, and the period of contraception before postoperative pregnancy [ 52 ]. To enhance the strength of the uterine wall for future pregnancy, various surgical management strategies and suture techniques have been developed to restore myometrial thickness, anatomical alignment of the uterine layers, and endometrial integrity, such as robot-assisted surgery [ 13 , 37 , 53 ]. Otsubo et al. [ 54 ] reported that the optimum uterine wall thickness for successful conception and preventing uterine rupture during pregnancy may range from 9 to 15 mm. Preterm delivery may be associated with a thinning of the uterine wall and reduced uterine capacity following surgery. Hijazi et al. [ 37 ] reported that the mean thickness of the uterine wall decreased by 41% (from 4.02 ± 1.11 cm preoperatively to 2.37 ± 0.84 cm postoperatively), and the preterm delivery rate reached 33.3% after treatment. Therefore, surgical operators must strike a balance between excessive excision of normal myometrium and incomplete removal of lesions. Changes in the anatomical integrity and function of the uterus, such as pelvic or intrauterine adhesions, uterine deformities, reduced uterine capacity, and decreased blood flow in the postoperative field, may contribute to adverse pregnancy outcomes. The primary adverse pregnancy outcomes were placental abnormalities, including placenta accreta, placenta increta, retained placenta, and placenta previa. These complications may be associated with disruptions of endometrial integrity and postoperative ischemia in the myometrium or endometrium. Kishi et al. [ 30 ] reported two cases of placenta accreta, both resulting from severe adenomyosis affecting a broad range of the submucosal myometrium. To completely remove the lesion, endometrial perforation was difficult to avoid during surgery, potentially leading to placental invasion into the outer myometrium through the defected submucosal layer. The postoperative field may be observed as an ischemic area within six months after surgery [ 52 ]. If the embryo was planted in this ischemic region before blood flow resumes, the placenta may invade the myometrium deeply to obtain more blood supply, potentially leading to placental abnormalities. Subgroup analyses were conducted based on different surgical approaches (laparotomy and laparoscopy). The findings suggested that laparoscopic approaches resulted in better reproductive outcomes compared to laparotomy. This was evidenced by higher pregnancy and delivery rates, coupled with lower rates of pregnancy loss, adverse pregnancy outcomes, and preterm delivery associated with the laparoscopic technique. Laparoscopic surgery offers the benefits of reduced invasiveness, expedited recovery, and a diminished incidence of postoperative complications. These advantages positively influence postoperative reproductive outcomes and have led to its escalating application in clinical practice. Image-guided thermal ablation, as a minimally invasive intervention, is increasingly being used in the treatment of adenomyosis. This procedure induces instant coagulative necrosis of the target lesion by focusing thermal energy on the designated area, with or without an electrode [ 55 ]. By minimizing damage to the surrounding normal myometrium, it has less impact on the anatomical integrity and function of the uterus, eliminating the need for abdominal and uterine incision, suturing, and reconstruction during thermal ablation. Another significant advantage is the rapid recovery post-surgery, which facilitates early postoperative pregnancies. For instance, when a patient desires to conceive soon after treatment, it is recommended that she begin trying to conceive 1–6 months following the procedure [ 48 ]. Our results demonstrated encouraging postoperative reproductive outcomes (pregnancy rate: 52.0%, delivery rate: 32.5%) with low rates of cesarean Sect. (44.6%), preterm delivery rate (0.3%), and adverse pregnancy outcomes (1.0%). No cases of uterine rupture were reported in the included studies, and most women gave birth vaginally after treatment, indicating that the less invasive intervention not only preserves fertility but also allows for safe childbearing. Only one study [ 33 ] directly compared the reproductive outcomes of laparoscopic adenomyomectomy and HIFU in patients with adenomyosis. The results showed that HIFU achieved better postoperative reproductive outcomes (pregnancy rate: 52.0%, delivery rate: 36.0%) than adenomyomectomy (pregnancy rate: 30.2%, delivery rate: 27.9%), with a shorter time to pregnancy after the procedure. However, as this study was a retrospective design rather than a randomized controlled trial (RCT), the patients’ fertility desire and baseline fertility potential might differ. Additionally, this study [ 33 ] found that patients with diffuse adenomyosis had a lower pregnancy rate compared to those with focal adenomyosis after treatment (40% vs. 70%). The probable reason was that diffused lesions required larger ablation areas, leading to more severe endometrial burning. Yang et al. [ 50 ] reported no evidence of ovarian function decline following MWA for adenomyosis, demonstrating that thermal ablation does not have an observable adverse effect on the ovary within one year after the procedure. However, our results showed that the pregnancy loss rate (39.5%) and spontaneous miscarriage rate (27.1%) after thermal ablation were relatively high compared to those following adenomyomectomy. A significant proportion of pregnancy losses (42.4%) were due to intentional termination for personal reasons. Spontaneous miscarriage may be associated with endometrial damage during the ablation procedure. Due to the ill-defined endometrial–myometrial boundaries, complete ablation of adenomyotic lesions may be accompanied by thermal injury to the endometrium. Nam [ 48 ] proposed that excessive ablation of lesions might damage the integrity of the myometrium and affect the endometrium, potentially leading to intrauterine adhesions and decreased endometrial receptivity. To preserve fertility, the operator performed RFA with a reduced ablation range, ensuring at least a 1-cm thickness of the myometrium remained both from the endometrium and the serosa. Given the limited high-quality evidence, further studies are needed to compare the reproductive outcomes following thermal ablation for adenomyosis with other fertility-sparing interventions. Uterine artery embolization is a well-established minimally invasive technique for treating benign uterine conditions such as uterine fibroids and adenomyosis. This procedure involves injecting embolization agents into the uterine arteries, leading to ischemic necrosis of the targeted lesions [ 56 ]. The available studies on reproductive outcomes after UAE for adenomyosis are severely limited due to the controversial impact of UAE on fertility and the lack of endorsement from various reputable learned societies regarding the use of UAE for women who desire to conceive [ 57 – 60 ]. Given the inadequate evidence specific to adenomyosis, we could draw upon existing literature on uterine fibroids as a reference. Several studies have indicated the detrimental impact of UAE on ovarian function [ 61 – 64 ]. It appears that UAE is associated with inferior fertility outcomes, including reduced rates of successful pregnancy and delivery, as well as elevated rates of miscarriage and obstetrical complications, when compared to other interventions [ 17 , 65 – 70 ]. Potential anastomoses between uterine and ovarian arteries can facilitate the entry of embolization agents into the ovarian artery, thereby impeding ovarian function through diminished blood supply and direct tissue injury [ 70 – 73 ]. Moreover, the possibility of ischemic injury within the myometrium or endometrium cannot be overlooked, as it may potentially impact endometrial receptivity [ 74 ]. Additionally, the radiation dose administered during the procedure can exert detrimental effects on the ovaries and endometrium [ 75 ]. In contrast, an increasing number of recent studies have demonstrated more satisfactory reproductive outcomes following UAE treatment than previously reported. Several studies have indicated that UAE does not lead to ovarian dysfunction in young women (< 40 years old), and these women can exhibit a greater capacity for recovery even if the ovaries are temporarily damaged [ 76 – 79 ]. Some reviews [ 80 , 81 ] have shown that UAE can be offered to women seeking pregnancy with fertility and miscarriage rates comparable to those of the age-matched general population. The authors proposed that patient age is a major confounding factor and suggested that the primary reason for poor reproductive outcomes after UAE in many previous studies was the inclusion of participants over 40 years. Moreover, improvements in the UAE technique and embolization agents have also enhanced pregnancy outcomes [ 81 – 83 ]. Therefore, whether the results of previous studies apply to current technology remains a question. In the future, more high-quality randomized controlled trials (RCTs) regarding UAE treatment for adenomyosis are needed to strengthen the evidence base, particularly for younger women. Previous literature suggested that adenomyomectomy and thermal ablation do not have significant adverse effects on ovarian reserve or that any impact is recoverable in a short period [ 84 ]. However, the impact of UAE on ovarian reserve remains controversial as previously mentioned. Compared to endometriosis and the surgical techniques, the excision of endometriosis and subsequent hemostasis are considered to have detrimental effects on ovarian reserve [ 85 ]. The impact of surgery on ovarian reserve can be reduced by improving surgical techniques [ 86 ]. The treatments for adenomyosis and endometriosis can both potentially have adverse effects on ovarian reserve, but the mechanisms of impact differ. Since the lesions of adenomyosis are located in the uterus rather than the ovaries, the impact of treatments on ovarian reserve is primarily through indirect mechanisms such as thermal damage (thermal ablation) or ischemic injury (UAE). In contrast, the lesions of endometriosis are located in the ovaries, and the excision of these lesions directly affects ovarian reserve. Because the I 2 values in most results demonstrated significant heterogeneity, we performed subgroup and sensitivity analyses to explore potential causes of the heterogeneity within the adenomyomectomy group. Heterogeneity might stem from differences in baseline participant characteristics (especially reproductive potential, such as age and subfertility history), the extent of adenomyosis, treatment protocols, duration of follow-up, or other study-specific characteristics, given that none of the included articles were RCTs. Subgroup analyses were conducted based on whether participants received adjuvant hormone therapy, participants’ age, type of adenomyosis, size of adenomyosis, and year of publication. The results showed that heterogeneity was significantly reduced (I² statistic ≤ 50%) in several subgroup analyses based on whether participants received adjuvant hormone therapy, type of adenomyosis, and size of adenomyosis, indicating these factors could partly explain the heterogeneity. We also performed sensitivity analyses for studies with a follow-up ≥ 36 months, and the heterogeneity remained significant, indicating that the duration of follow-up was not the source of heterogeneity. However, due to data limitations, we were unable to explore all potential sources of heterogeneity. The strengths of this meta-analysis include a comprehensive systematic analysis of reproductive outcomes following fertility-sparing interventions for adenomyosis according to a standardized core outcome set [ 18 ] and detailed subgroup analyses compared to previous meta-analysis. Despite these strengths, there are some limitations. Firstly, the absence of RCTs and the reliance on single-arm clinical studies or observational non-RCTs mean that participant characteristics and study designs varied significantly. In particular, participants’ reproductive potential at baseline and their fertility desire after treatments differed markedly across studies, potentially affecting outcome accuracy. Secondly, only one article regarding UAE was available, and most physicians intentionally do not select patients desiring fertility for UAE treatment. Thirdly, studies investigating interventions for adenomyosis have used many different outcome measures, and most articles were not primarily designed to explore reproductive outcomes, making it challenging to provide clinical practice guidelines. Although Tellum et al. developed a standardized core outcome set for research on uterus-sparing treatments of adenomyosis in 2022 [ 18 ], and we conducted this meta-analysis strictly following the standards and definitions of reproductive outcomes outlined in that consensus, the included articles did not adhere to it because they were published before the consensus was established. Fourth, the length of follow-up was a confounding factor, as the assessment of reproductive outcomes improved with a longer duration of follow-up. A short follow-up period (< 24 months) in 21.9% of articles was not sufficient to observe reproductive outcomes adequately. Fifth, significant heterogeneity was observed in most of the outcome measures, and we could not explore all sources of heterogeneity. Sixth, only 25.0% of the included articles were classified as high quality, with the majority categorized as poor-to-medium quality. Seventh, our search was limited to articles written in English or Chinese. To address these limitations, future research should employ standardized outcome measures and include more well-designed RCTs, particularly studies focusing on UAE.

This study highlights the promising reproductive outcomes after fertility-sparing interventions for women with adenomyosis who desire pregnancy. The cesarean section rate, preterm delivery rate, and adverse pregnancy outcomes rate were higher after adenomyomectomy compared to thermal ablation. Conversely, the pregnancy loss rate and spontaneous miscarriage rate were higher after thermal ablation than after adenomyomectomy. The limited available evidence, potential selection bias, and heterogeneity among included articles are confounding factors that might influence the assessment of outcomes. Despite the limitations, our study provides valuable insights for clinical decision-making regarding reproductive outcomes after fertility-sparing interventions for adenomyosis, particularly for women who wish to preserve their fertility.

Adenomyosis is a common benign condition defined as the presence of endometrial glands and stroma within the myometrium [ 1 ]. The prevalence of adenomyosis varies significantly across different studies due to challenges in designing valid epidemiologic studies. The estimated prevalence among patients undergoing hysterectomy has ranged from 8.8% to 61.5% [ 2 ]. Recent studies have shown that nulligravid young women can also suffer from adenomyosis, suggesting that the disorder may develop earlier in reproductive life than previously thought [ 3 , 4 ]. Adenomyosis appears to be associated with subfertility and adverse reproductive outcomes. Previous studies have demonstrated significant differences in clinical reproductive outcomes between women with and without adenomyosis [ 5 – 7 ]. Women with adenomyosis seem to have significantly lower implantation rates, clinical pregnancy rates, and live birth rates but higher spontaneous miscarriage rates and adverse obstetric outcomes (e.g., preterm birth, small for gestational age, pre-eclampsia, postpartum hemorrhage) compared to those without adenomyosis [ 8 – 11 ]. At the molecular level, the presence of ectopic endometrium within the myometrium disrupts the endometrial cellular, hormonal, and immunological environment, adversely affecting decidualization, placentation, and embryonic developmental programming [ 9 ]. As the incidence of adenomyosis rises among nulligravid women and childbearing increasingly occurs later in life in many countries, patients are more frequently opting for fertility-sparing interventions over hysterectomy, particularly those who wish to preserve their fertility. However, there is no consensus on the extent to which these treatments can improve reproductive outcomes in patients with adenomyosis [ 12 ]. The fertility-sparing interventions discussed in this study include adenomyomectomy, uterine artery embolization (UAE), and image-guided thermal ablation. Image-guided thermal ablation encompasses high-intensity focused ultrasound (HIFU), radiofrequency ablation (RFA), and microwave ablation (MWA) which utilize different thermal energy sources guided by ultrasound or magnetic resonance imaging. Uterine-sparing surgery has been proven effective for infertility likely caused by adenomyosis [ 13 ]. Encouraging reproductive outcomes have been reported following adenomyomectomy and thermal ablation [ 14 , 15 ]. In contrast, UAE is considered controversial for women who wish to conceive due to studies reporting low pregnancy rates and high rates of obstetric complications, such as miscarriage and postpartum hemorrhage, following UAE [ 16 , 17 ]. Currently, the exact impact of different interventions on reproductive outcomes in adenomyosis-associated subfertility has not been fully clarified. Reproductive outcomes following fertility-sparing interventions for adenomyosis are clinically significant. However, there is currently no consensus on the most appropriate management for women seeking fertility with adenomyosis, and no direct randomized comparison of these interventions has been reported. This highlights the challenges faced by gynecologists in managing these patients. To provide valuable evidence for clinical decision-making, we aimed to conduct a meta-analysis of reproductive outcomes following various fertility-sparing interventions in women with adenomyosis.

Supplementary Material 1. Supplementary Material 1.