The comparison of myomectomy, UAE and MRgFUS in the treatment of uterine fibroids: a meta analysis

Abstract

Objective

To compare the re-intervention rates of myomectomy, uterine artery embolization (UAE) and magnetic resonance-guided focused ultrasound surgery (MRgFUS) for uterine fibroids (UFs) in different follow-up time.

Methods

Two investigators searched PubMed for clinical studies published in English from 1 Jan 2000 to 31 Dec 2020, and independently examined the paper to select qualified studies, extracted relevant information and assessed the risk of bias. Meanwhile, a meta-analysis of 31 studies containing totally 42103 patients was conducted to compare the re-intervention rate of myomectomy, UAE and MRgFUS.

Results

In the meta-analysis of 42103 patients, the 12-month re-intervention rates of myomectomy, UAE and MRgFUS for UFs were 0.06 (95%CI, 0.01–0.11), 0.07 (95%CI, 0.06–0.09), and 0.12 (95%CI, 0.04–0.20) respectively. The 24-month re-intervention rates were 0.10 (95%CI, 0.04–0.16), 0.08 (95%CI, 0.01–0.17), and 0.14 (95%CI, 0.07–0.21) respectively. The 36-month re-intervention rates were 0.09 (95%CI, 0.05–0.13), 0.14 (95%CI, 0.05–0.23), and 0.22 (95%CI, 0.11–0.32) respectively. Additionally, the 60-month re-intervention rates were 0.19 (95%CI, 0.15–0.24), 0.21 (95%CI, 0.17–0.25), and 0.49 (95%CI, 0.21–0.77) respectively.

Conclusions

The myomectomy has the lowest re-intervention rate of the three regimens in short time and long time while the MRgFUS has the highest. The rate of MRgFUS increased rapidly in the 60th month after the treatment.

Keywords:

• Uterine fibroids
• myomectomy
• UAE
• MRgFUS
• re-intervention

1. Introduction

Uterine fibroids (UFs) are considered as the most common benign uterine tumors in women of childbearing age, and a large number of women with UFs are probably to schedule a uterus-conserving treatment [1]. The patients who wish to preserve the uterus had the options of myomectomy, uterine artery embolization (UAE) and magnetic resonance-guided focused ultrasound (MRgFUS) [2]. Although there was a large body of literature on the efficacy of various uterine-sparing interventions, the outcome still remained controversial [3]. For example, the lowest 24-month re-intervention rate of myomectomy was reported by Michal Mara et al. as 3% and the highest was reported by S Subramanian et al. as 17% [4,5]. Therefore, it was of great significance to evaluate the short-time and long-time efficacy of the different regimens.

Disease status after uterine-sparing treatment is a crucial factor for clinical management of UFs, and the re-intervention risk of each option should be evaluated in different endpoints [1]. For example, the 12-month re-intervention rate of UAE was reported by Matthew R Davis et al. as 7% and the 36-month re-intervention rate was reported by Scott C Goodwin et al. as 14% [6,7]. However, the longitudinal data of the common uterine-sparing regimens still remained fragmentary [8]. As a result, we conducted this meta-analysis to 1) compare the re-intervention rate between myomectomy, UAE and MRgFUS groups and 2) estimate a clinically meaningful re-intervention rate for UFs in different follow-up time.

2. Methods

2.1. Search strategy

PubMed, Embase and Cochrane databases were systematically searched for relevant published articles with the following terms: (uterine fibroids OR fibroids OR UFs OR leiomyoma OR myoma) and (uterine artery embolization OR UAE OR magnetic resonance-guided focused ultrasound OR MRgFUS OR High-Intensity Focused Ultrasound OR HIFU OR myomectomy). The date were from 1 Jan 2000 to 31 Dec 2020. In addition, the search was restricted to human subjects and English-language studies.

2.2. Eligibility criteria

Inclusion criteria were as follows: (1) The study participants were human beings. (2) Original research papers with 30 or more patients were written in English. (3) Studies provide sufficient data on uterine-sparing treatment, and re-intervention. The re-intervention was defined as all additional treatment necessary during the follow-up period to control persistent or recurrent fibroid-related symptoms (such as hysterectomy, myomectomy and endometrial ablation) [9]. (4) Studies with randomized controlled trials (RCTs), prospective or retrospective design. (5) Studies published between 1 Jan 2000 and 31 Dec 2020. (6) The patients underwent initial myomectomy, UAE or MRgFUS. (7) A follow-up period of 12, 24, 36 and (or) 60 months were required after treatment.

The studies were excluded according to the following criteria: (1) Abstracts, case reports, case series, meeting summaries, comments, letters, reviews or studies not related to human subjects. (2) Repeated studies including the same patients. (3) Insufficient sample size of less than 30 patients. (4) Studies with a lack of relevant outcome data and (5) Studies published before 1 Jan 2000 or after 31 Dec 2020.

2.3. Data extraction and quality assessment

Two investigators (Fan Xu and Lihong Deng) independently extracted the data from the studies. All data were confirmed by internal consistency, and disagreements were resolved by discussion between the two investigators. The present study was approved by the Ethics Committee of Nanchong Central Hospital, the Second Clinical Medical College, North Sichuan Medical University. The current meta-analysis was conducted in accordance with the Preferred Reporting Items for Systematic Review and Meta-Analysis (known as PRISMA) statement [10]. The Newcastle-Ottawa Scale (NOS) was employed to evaluate the study quality. Each study considered to be a suitable candidate for inclusion in the meta-analysis was awarded a ‘star’. A high-quality study was defined as a study with five or more stars in total.

2.4. Statistical analysis

All data were analyzed using STATA software version 12.0 (StataCorp, College Station, TX), with p < 0.05 being considered to be statistically significant. In the meta-analysis, the unit of analysis was the re-intervention rate of myomectomy, UAE or MRgFUS. Heterogeneity was assessed using the chi-square test and I² statistic. Heterogeneity was defined as a p value less than 0.10 or I² greater than 50%. A fixed-effects model was suitable for data with no or low heterogeneity (I²<50%). Otherwise, a randomeffects model was used. Publication bias was assessed with Egger's test. Sensitivity analysis was performed in order to identify the influence of an individual study on pooled estimates by removing one study at a time.

3. Results

3.1. Literature search

Initially, totally 4361 records were identified in the target database. Full texts of more than 300 articles were reviewed because the re-intervention risk was usually not a primary outcome in studies. A total of 31 studies were included in the final analysis (Figure 1). Among the included studies, six were randomized controlled studies [4,11–15] and 25 were cohort studies [5–7,9,16–36]. In addition, 18 from Europe (58.1%), 9 originated from USA (29.0%), 3 from Japan (9.7%), and 1 from Israel (3.2%). We evaluated the quality of the studies based on the Newcastle-Ottawa scale, and all studies had a score that ranged from 6 to 9 (Table 1).

Figure 1. Flow-chart showing the article-screening process.

Table 1. Characteristics of the included studies.

First author	Year	Source of patients	No. of patients	Clinical period (year)	Follow up (month)	Design	NOS score
Richard D Edwards	2007	UK	157	2000–2004	12	RCT	9
Isaac T Manyonda	2012	UK	163	NA	24	RCT	8
Michal Mara	2006	Czech	74	2001–2004	24	RCT	9
J G Moss	2011	UK	157	2000–2005	12	RCT	8
Sanne M van der Kooij	2010	Netherlands	177	2002–2004	60	RCT	8
Annefleur M de Bruijn	2016	Netherlands	177	2002–2004	120	RCT	9
Giovanna Tropeano	2012	Italy	176	2001–2010	36	cohort	7
Francis Joffre	2004	France	85	2000	18	cohort	8
Marlijne E Ikink	2014	Netherlands	119	2010–2013	12	cohort	8
Scott C Goodwin	2008	USA	1278	NA	36	cohort	7
Paul N M Lohle	2008	Netherlands	93	1998–2002	54	cohort	8
Yutaka Morita	2008	Japan	48	2005–2006	12	cohort	8
Christian Scheurig-Muenkler	2013	Germany	380	2000–2007	60	cohort	8
Miyuki Sone	2010	Japan	33	2004–2006	21	cohort	7
Elizabeth A Stewart	2002	USA	65	NA	83	cohort	8
W J Walker	2006	UK	172	NA	60	cohort	9
V Froeling	2013	Germany	77	2002–2009	60	cohort	8
Stevo Duvnjak	2016	Denmark	350	2001–2011	93	cohort	6
K Funaki	2009	Japan	91	2004–2008	12	cohort	7
Irene Mindjuk	2015	Germany	252	2010–2012	18	cohort	8
Anand Narayan	2010	USA	247	2000–2002	60	cohort	8
Martin Popovic	2009	Austria	53	2004	36	cohort	7
S D Quinn	2014	UK	280	2003–2010	60	cohort	8
M Sami Walid	2010	USA	41	1999–2009	12	cohort	7
Rebecca Flyckt	2016	USA	374	1995–2009	96	cohort	8
Roopina Sangha	2016	USA	118	2005–2009	24	cohort	7
S Subramanian	2001	USA	820	NA	24	cohort	6
Krzysztof R Gorny	2014	USA	138	2005–2011	48	cohort	7
Matthew R Davis	2018	USA	35631	2009–2013	12	cohort	8
Inez M Verpalen	2020	Netherlands	123	2010–2017	34	cohort	8
Aya Mohr-Sasson	2018	Israel	154	2012–2017	36	cohort	9

RCT: randomized controlled trial; NOS: Newcastle-Ottawa Scale; NA: not available.

Due to the re-intervention data, the patients treated by ultrasound image-guided focused ultrasound surgery (USgFUS) were not involved. Some studies presented more than one treatment. Therefore, the total number was not the same as the sum of subgroups. These studies compared the re-intervention case following myomectomy, UAE and MRgFUS in the treatment of UFs. The main outcome was the re-intervention rate of these regimens. Table 1 lists the baseline characteristics of the patients included in these studies.

3.2. Primary outcome

Meta-analysis of the 12-month, 24-month, 36-month and 60-month re-intervention rate of MRgFUS, UAE and myomectomy for UFs was shown in Table 2. Among the short-time efficacy of this regimens, for 12-month re-intervention rate, estimations of the pooled rates of MRgFUS, UAE, and myomectomy were 0.12 (95%CI, 0.04–0.20; I²=89.1%; p = 0.000), 0.07 (95%CI, 0.06–0.09; I²=14.2%; p = 0.324), and 0.06 (95%CI, 0.01–0.11; I²=95.1%; p = 0.000), respectively. In terms of the 24-month re-intervention rate, estimations of the pooled rates of MRgFUS, UAE, and myomectomy were 0.14 (95%CI, 0.07–0.21), 0.08 (95%CI, 0.01–0.17; I²=75.7%; p = 0.016), and 0.10 (95%CI, 0.04–0.16; I²=76.0%; p = 0.002), respectively.

Table 2. Meta-analysis of the 12-month, 24-month, 36-month and 60-month re-intervention rate of MRgFUS, UAE and myomectomy for UFs.

Variables	Number of studies	Number of patients	reintervention rate (95%CI)	I- squared (%)	P-value
12 month
MRgFUS	4	442	0.12 (0.04–0.20)	89.1	0.000
UAE	6	5013	0.07 (0.06–0.09)	14.2	0.324
Myomectomy	3	14665	0.06 (0.01–0.11)	95.1	0.000
24 month
MRgFUS	1	91	0.14 (0.07–0.21)	–	–
UAE	3	98	0.08 (0.01–0.17)	75.7	0.016
Myomectomy	5	485	0.10 (0.04–0.16)	76.0	0.002
36 month
MRgFUS	3	329	0.22 (0.11–0.32)	86.3	0.002
UAE	4	1523	0.14 (0.05–0.23)	94.7	0.000
Myomectomy	3	184	0.09 (0.05–0.13)	0.0	0.508
60 month
MRgFUS	3	336	0.49 (0.21–0.77)	96.5	0.000
UAE	11	6051	0.21 (0.17–0.25)	84.1	0.000
Myomectomy	5	14084	0.19 (0.15–0.24)	53.7	0.071

MRgFUS: magnetic resonance-guided focused ultrasound; UAE: uterine artery embolization; UFs: uterine fibroids.

Among the long-time efficacy of this regimen, for the 36-month re-intervention rate, estimations of the pooled rates of MRgFUS, UAE, and myomectomy were 0.22 (95%CI, 0.11–0.32; I²=86.3%; p = 0.002), 0.14 (95%CI, 0.05–0.23; I²=94.7%; p = 0.000), and 0.09 (95%CI, 0.05–0.13; I²=0.0%; p = 0.508), respectively. Regarding the 60-month re-intervention rate, estimations of the pooled rates of MRgFUS, UAE, and myomectomy were 0.49 (95%CI, 0.21–0.77; I²=96.5%; p = 0.000), 0.21 (95%CI, 0.17–0.25; I²=84.1%; p = 0.000), and 0.19 (95%CI, 0.15–0.24; I²=53.7%; p = 0.071), respectively. Figure 2 presents the longitudinal data of re-intervention rate.

Figure 2. The re-intervention rate for MRgFUS, UAE and myomectomy.

3.3. Publication bias and sensitivity analyses

The publication bias analyses were quantified with Egger’s test. No obvious asymmetry or p > 0.05 was detected, suggesting that there was no evidence of publication bias. A sensitivity analysis was employed to assess the reliability of the results. When one study was deleted at a time, the heterogeneity and the results still presented no difference.

4. Discussion

In recent decades, success has been achieved in the management of UFs after uterine-sparing treatment. Nevertheless, the real efficacy of this regimen still remained unknown [37]. The meta-analysis of re-intervention rate after treatment for UFs was conducted to demonstrate that, the efficacy of the common uterine-sparing treatment including myomectomy, UAE, and MRgFUS, was satisfactory in the 12th month, the 24th month and the 36th month. However, the re-intervention rate of MRgFUS increased rapidly in the 60th month after treatment. To the best of our knowledge, this is the first study to report re-intervention rate meta-analysis after the treatment for UFs.

UFs are the extremely common benign tumors of female genital tract, causing substantial physical and psychological damages to patients. For women desiring uterine conservation, the re-intervention risk after treatment was the main factor affecting the clinical decision making [38]. However, the short-time and long-time re-intervention rate of the uterine-sparing regimens was still debated [1]. Some previous studies have reported the efficacy of different regimens. For example, Isaac T Manyonda et al. reported that [12], both UAE and myomectomy contributed to significant and equal improvements in QoL. Meanwhile, UAE allowed a shorter hospital stay but with a higher rate of reintervention. In our meta-analysis, the re-intervention rate of UAE was higher than myomectomy but lower than MRgFUS.

Myomectomy, including laparoscopic, hysteroscopic, abdominal or transvaginal myomectomy, is a widely used procedure for removing leiomyomas [23]. A longitudinal retrospective cohort study including 35631 patients demonstrated that the re-intervention rate of myomectomy was 4% [6]. In our meta-analysis, estimations of the pooled 12-month, 24-month, 36-month and 60-month re-intervention rates of myomectomy were respectively 6%, 10%, 9% and 19%, which were the lowest of the three regimens. According to the results, less than one fifth patients with UFs suffered re-intervention during the five years following myomectomy. The reason of myomectomy with the lowest re-intervention rate may be related to the surgery, while in myomectomy procedure, the tumors had removed. Therefore, the re-intervention event happened only due to new fibroids development.

UAE is less invasive, and associated with a shorter hospital stay and a quicker recovery regimen for UFs. Additionally, the currently available data on the likelihood of failure after UAE are inconsistent [39]. For example, the lowest 60-month re-intervention rate of UAE was reported by Giovanna Tropeano et al. as 7% and the highest was reported by Annefleur M de Bruijn et al. as 35% [15,16]. In our meta-analysis, estimations of the pooled 12-month, 24-month, 36-month and 60-month re-intervention rates of UAE were respectively 7%, 8%, 14% and 21%, which were lower than MRgFUS but higher than myomectomy. The results proved that one fifth patients with UFs suffered re-intervention during the five years following UAE. Additionally, the reason of UAE having a higher re-intervention rate than myomectomy also lies in the procedure that the uterine fibroids has not been removed, and thus the fibroids left in suit may still have the risk of recurrence [40].

High intensity focused ultrasound (HIFU), including MRgFUS and USgFUS, is a newly thermal ablation technique for UFs treatment, resulting in reduced toxicity compared with other ablation techniques [41]. It is a no-invasive and low-cost procedure, which can cause minimal pain [42]. However, the short-time and long-time efficacy still remained unknown [43]. In our study, the re-intervention was defined as all additional intervention necessary after primary intervention (such as hysterectomy, myomectomy and UAE) [9]. Due to the lack of re-treatment data of USgFUS, we only conducted a meta-analysis of MRgFUS. Estimations of the pooled 12-month, 24-month, 36-month and 60-month re-intervention rates of MRgFUS were respectively 12%, 14%, 22% and 49%, which were the highest of the three regimens. Based on the obtained results, MRgFUS was a satisfactory alternative to surgery or UAE for fibroids with a re-intervention lower than 30%. However, the rate increased rapidly in the 60th month. It can be explained by the theory that the efficacy of this technique was associated with none perfusion volume (NPV) ratio of fibroids after treatment. For MRgFUS procedure, the mean NPV% was 48.4 ± 25.0% [35]. This part of fibroids without ablation may keep growing, which can cause the re-intervention.

However, the present study still had some limitations. Firstly, the main limitation of the present meta-analysis was the observed substantial heterogeneity. We are aware of that a lot of factors such as age, location of fibroids and menopause status might influence the re-intervention. However, further sub-analysis by this factor was not possible, because the information can not be obtained from the relevant studies. Nevertheless, this meta-analysis provides insights into current re-intervention rate and the obtained findings can be directly applicable in daily practice for counseling patients that are often eligible for over one treatment option. Secondly, there was not a sufficient number of RCTs for inclusion, which may have led to selection bias. In addition, our study has not included the patients with UFs who were treated by USgFUS due to the lack of data. The mean NPV% of USgFUS reported by Zhao WP et al. [44] was 68.5 ± 24.1%, which was more than that of MRgHIFU, indicating that further studies are required to evaluate the re-intervention of USgFUS.

In conclusion, myomectomy, UAE, and MRgFUS are effective for the treatment of UFs. The re-intervention rates of myomectomy, UAE, and MRgFUS are lower than 30% those during the 12 months, 24 months and 36 months. However, the rate of MRgFUS increased fast (49%) in the 60th month.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the Bureau of Science and Technology Nanchong City Program [No: 16YFZJ0029; 18SXHZ0126]; the Primary Health Development Research Center of Sichuan Province Program [No: SWFZ17-Y-25; SWFZ17-C-69]; the Science and Technology Department of Sichuan province Program [No: 2015JY0056]; the Nanchong Federation of Social Science Associations Program [No: NC2016B017]; Yuzhong District Research Program of Basic Research and Frontier Technology of Chongqing [No. 20202723].