Comparison of misoprostol and manual vacuum aspiration in treating first-trimester incomplete miscarriage: a systematic review and meta-analysis of randomized controlled trials

doi:10.21203/rs.3.rs-3371957/v1

Background

To provide evidence of the comparative curative efficiency and prevalence of treatment-related adverse events following manual vacuum aspiration (MVA) or misoprostol in first trimester incomplete abortion.

Methods

We comprehensively searched international medical literature databases, including PubMed, Medline, Ovid, Embase, and Web of science for related articles published between 2005 and 2023. After article screening, seven articles were finally included in the meta-analysis after assessment for risk of bias. Important parameters of the included studies were strictly extracted. Stata 17MP was used to compare the proportion of complete uterine evacuation, necessity for additional MVA, prevalence of adverse effects (abdominal pain, bleeding, fever, chills, nausea and vomiting), patients’ subjective evaluation and publication bias. Meta-regression was also performed.

Results

Seven eligible studies of 1097 patients receiving misoprostol and 1079 patients receiving MVA were included, and MVA was found to have better performance than misoprostol in terms of complete uterine evacuation (RR = 0.972, P༜0.001), regardless of age, gestational age, parity, dosage of misoprostol, and marital status. Meanwhile, misoprostol was related to higher possibility for additional MVA (RR = 7.112, P༜0.001). In terms of adverse events, misoprostol medication led to more frequent bleeding (RR = 1.91, P༜0.001), chills (RR = 7.5, P = 0.018), fever (RR = 4.34, P = 0.001), nausea (RR = 3.13, P = 0.005), and vomiting (2.21, P = 0.008).

Conclusion

MVA outperformed misoprostol in terms of accomplishing complete uterine evacuation in patients with first-trimester incomplete miscarriage. Moreover, the occurrence rate of adverse events was lower in MVA group than that in misoprostol group. However, since the rate of complete uterine evacuation after MVA and misoprostol were both over 90%, misoprostol was also considered as an alternative for MVA in limited resources settings.

Trial registration:

The research was formally registered on International Platform of Registered Systematic Review and Meta-analysis Protocols (registration number: INPLASY202350113) before statistical analysis.

First-trimester incomplete miscarriage

misoprostol

MVA

complete uterine evacuation

adverse events

Pregnancy abnormality contributed to various kinds of complications, which could lead to adverse events, including maternal death ^[1]. Abortion was considered as an urgent situation accompanied by risks of bleeding, infection, secondary infertility, uterine perforation, etc, most of which needed quick medical response ^[2]. Among these, incomplete abortion with intrauterine remanent tissues, are more probable to cause severe infection if not identified in early phase and not managed in a standard process ^[3]. For incomplete abortion, medical or surgical removal of residual intrauterine pregnancy tissues was the ultimate pathway for safe recovery^[4]. First trimester was typically counted from the first day of pregnant female’s period through to the 12^th week, which was the earliest phase of pregnancy with profound internal biochemical and physiological instability. Management of first trimester incomplete abortion should be rigorous and careful.

Traditionally, surgical removal of remanent tissue relied on curettage which was both a diagnostic and therapeutic tool. In recent years, manual vacuum aspiration (MVA) gained popularity in first trimester incomplete abortion owing to less damage, better pain management, and efficient uterine evacuation ^[5-6]. However, propitious manipulation and application of MVA depended on complete surgical equipment and rigorously trained personnel ^[7-8]. In severe areas with low-income settings, where fully equipped surgical tools and perioperative tutelage, MVA was less frequently performed. Moreover, MVA was comparatively more expensive ^[9].

In cases when MVA was not accessible, drug medication to achieve discharge of intrauterine tissues was regarded as replacement. Misoprostol was an artificially synthesized compound functioning as prostaglandin E1 (PGE1) used for accelerating labor, abortion or gastric ulcers^[10]. As a cheaper and more accessible choice for incomplete abortion, it was believed to maintained satisfactory rate of complete uterine evacuation when used with agents like mifepristone or methotrexate^[11]. In the past few years, studies also concluded that single use of misoprostol could contribute to comparable success in uterine evacuation in first trimester incomplete abortion while controlling the prevalence of drug-related adverse events ^[12].

With a purpose to guide the choice between MVA and misoprostol in first trimester incomplete abortion, several high-quality prospective randomized controlled trials have been carried out to weigh their efficiency for complete uterine evacuation as well as safety concerns, while inconsistency came up among studies ^[13-19]. Particularly, the prevalence of adverse events following MVA or misoprostol use in first trimester incomplete abortion have varied greatly among different studies. Therefore, we performed this systematic review and meta-analysis based on original data from randomized controlled trials, hoping to provide more evidence for clinical decision making.

The research was formally registered on International Platform of Registered Systematic Review and Meta-analysis Protocols (registration number: INPLASY202350113) before statistical analysis.

Initial search for articles

HJ, MZ, and XY performed the initial search of related articles through PubMed, Medline, Embase, Ovid, and Web of Science with key search words of (“incomplete abortion” OR “incomplete miscarriage” OR “abortion” OR “miscarriage” OR “miscarry) AND (“misoprostol” OR “medical treatment” OR “medication” OR “MVA” OR “manual vacuum aspiration” OR “vacuum aspiration” OR “aspiration” OR “uterine aspiration”). The duration of publication time was set between 2005 and 2023 in order to include sufficient number of studies. Meanwhile, we only took into consideration studies written in modern English. A more experienced researcher XL verified and collated the articles.

Further screening of articles

HJ and MZ independently started the screening process of articles, the main inclusion criteria were: 1) having reported the outcomes (rate of complete uterine evacuation, occurrence rate of adverse events, patients’ subjective evaluation) of both misoprostol and MVA group; 2) patients received only one treatment of MVA or misoprostol and did not receive accessory treatment; 3) having reported adequate baseline characteristics of patients including age, parity, gestational age, marital status, etc; 4) gestational age less than 13 weeks (first-trimester incomplete miscarriage). Main exclusion criteria were: 1) having reported patients accompanied by pelvic infection, severe anemia, renal failure, etc; 2) having reported patients without detailed follow-up information; 3) gestational age over 13 weeks. In addition, studies which were case report, conference abstract\, correspondence were not considered. Moreover, studies with poor relevancy, full text unavailability, or duplicated studies were excluded. A more experienced researcher XL would participate and re-evaluate the screening process if disagreement occurred. In such circumstance, HJ, MZ and XL would make a final joint decision following discussion.

Assessment of quality of included studies

The checklist for risk of bias evaluation from Review Manager 5.4 was used for the assessment of study quality. HJ and MZ took the lead in this process under the supervision of XL.

Extraction of data

Firstly, HJ and MZ downloaded the original full texts of the included studies and extracted essential parameters like the name of the first author, the year of publication, the journal name, the place or country where the study was carried out, the duration of patients recruitment, etc. Meanwhile, important parameters of patients, including number, age, gestational age, parity, marital status and educational status of patients in misoprostol and MVA group were extracted. In addition, the main diagnosis and imaging diagnostic method were also collected. Moreover, the dosage, pharmaceutical company, and administration route of misoprostol were recorded. Besides the aforementioned data, follow-up information, including time of successful complete uterine evacuation, occurrence rate of adverse effects, patients’ subjective evaluation towards the treatment were all collected.

Statistical analysis

The pooling analytic process was performed using Stata 17MP (StataCorp LLC, Texas, The United States) based on original data. The main outcomes for comparison were rate of complete uterine evacuation, rate of supplementary MVA, prevalence of treatment-related adverse events, and patients’ subjective remarks. A subgroup analysis was carried out to look for influencing factors of RR using Stata 17MP.

Selection of studies

Through the initial search for relevant articles from international medical databases, including PubMed, Medline, Web of Science, Embase, Ovid, we found a total of 258 articles. Meanwhile, 24 studies which had been registered were found. Among the 282 studies, 36 studies were duplicated (32 removed by Endnote and 4 manually) and were subsequently ruled out, which left 246 studies. Relevancy analysis determined that 183 articles had poor relevancy, after which 63 studies remained for further screening. Among the 63 studies, 18 studies could not be retrieved. Furthermore, 19 studies were no longer considered due to irrelevant interventions. Six studies were deleted because of scare follow-up information. Eventually, seven studies were included in this meta-analysis after case reports, letters, etc were excluded. The flow diagram regarding the detailed articles inclusion and exclusion process was shown in Figure 1.

Basic characteristics of the included studies

After a strict screening process, seven original articles regarding the relative efficacy of complete uterine evacuation following misoprostol intake and MVA in first-trimester incomplete miscarriage were included in this systematic review and meta-analysis. Baseline characteristics of the seven studies including author, publication year, number of patients recruited, age of patients receiving misoprostol and MVA respectively, gestational age of patients receiving misoprostol and MVA respectively, patients’ marital status and educational level, patients’ parity, final diagnostic method of incomplete miscarriage, inclusion and exclusion criteria, dosage and administration route of misoprostol, diagnostic method of uterine evacuation, etc. were extracted and shown in Table 1. Two of the seven studies were published after 2020, two were published between 2010 and 2019, and three of them were published between 2005 and 2009. Three studies by Ani et al, Nwafor et al, and Ibiyemi et al were carried out in Nigeria, the remaining four studies were carried out in Egypt, Burkina Faso, Mozambique, and Uganda respectively. All seven studies were randomized controlled trials (RCT). A total of 1097 patients with first-trimester incomplete miscarriage receiving misoprostol treatment and 1079 patients with first-trimester incomplete miscarriage receiving MVA treatment were included in this systematic review and meta-analysis. The largest number of patients in one single study was recorded in the study by Dabash et al, which included 349 patients receiving misoprostol treatment and 348 patients receiving MVA treatment. There was no significant difference between misoprostol group and MVA group in all seven studies in terms of age, gestational age, marital status, and educational levels. However, we observed differences regarding marital status and educational level between different studies. The highest proportion of married female was 99.7% and 99.7% in misoprostol and MVA group in the study by Dabash et al, the lowest proportion of married female was 21.5% and 19.7% in misoprostol and MVA group in the study by Bique et al. Parity of patients receiving misoprostol ranged from 1.4 to 2.2±2.1 and parity of patients receiving MVA ranged from 1.5±1.5 to 2.1±2.1 respectively. Four studies eventually determined incomplete miscarriage by disease history, physical examination, speculum examination and ultrasound. The two studies by Bique et al and Weeks et al determined incomplete miscarriage relying on disease history, physical examination, speculum examination, without the help of ultrasound. Ani et al diagnosed incomplete miscarriage relying on ultrasound directly. The main inclusion criteria were evidence of incomplete miscarriage, uterine size less than 13 weeks’ gestation. The main exclusion criteria were known allergy to prostaglandins, profuse and uncontrolled vaginal bleeding, existence of intrauterine device, suspected ectopic pregnancy, signs of pelvic infection, uterine size over 13 weeks’ gestation, severe asthma, etc. All seven studies gave misoprostol through sub-lingual route. Five studies provided 600 μg misoprostol, one study (Ani et al) provided 400 μg misoprostol, and one study (Dabash et al) provided 2200 μg misoprostol. Five studies investigated the efficiency of treatment by ultrasound one week after treatment. The study by Weeks determined the efficiency of complete uterine evacuation mostly by bimanual examination.

Quality of included studies

We used Review Manager 5.4 (Cochrane Collaboration, Copenhagen, Demark) to evaluate the quality of the included studies as well as the risk of potential bias. Through our analysis, the seven studies we included all had unclear risks in some particular aspects, but that did not affect the fact that all seven studies were considered to be high-quality studies in terms of study design, participants recruitment, statistical analysis, etc. Therefore, all seven studies were included in the final meta-analysis. The quality assessment was shown in Figure 2.

Successful complete uterine evacuation rate by misoprostol and MVA

The pooled analysis of the seven studies we recruited showed treatment by misoprostol had an incidence rate of complete uterine evacuation of 93.04%, while MVA manifested with complete uterine evacuation rate of 98.70% with an RR of 0.972 [95% CI 0.959-0.984 P＜0.001], which showed that MVA outperformed misoprostol shown by higher successful complete uterine evacuation rate. The highest complete uterine evacuation rate of misoprostol treatment was 98.3% and the lowest was 81.3% in the studies included. Regarding patients treated by MVA, the highest rate of complete uterine evacuation was 100% (in 2 studies), and the lowest was 91.5% in the articles we included. Among the seven studies, five studies observed a better performance and efficiency of MVA in terms of complete uterine evacuation rate manifested by RRs less than 1, and the other two studies found a more promising treatment outcome by misoprostol. Particularly, the study by Dabash et al in 2010 had the highest number of included patients, thus its relative weight was comparatively high in this meta-analysis, which were 73.32% and 20.37% in fixed and random model respectively. Of the seven articles we included, the work by Weeks et al had an RR over 1.00, the rest 6 articles all had RRs less than 1.00, which were 0.863 [95% CI 0.798-0.935 P＜0.001], 0.849 [95% CI 0.732-0.986 P=0.032], 0.839 [95% CI 0.766-0.918 P＜0.001], 0.986 [95% CI 0.971-1.001 P=0.058], 0.953 [95% CI 0.921-0.987 P=0.007], 0.911 [95% CI 0.857-0.968 P=0.003]. The forest plot of complete uterine evacuation was shown in Figure 3A.

Meanwhile, according to results of pooled analysis by five studies, the pooled percentage of patients needing additional MVA was 10.3% in misoprostol treatment group and was 0.9% in MVA treatment group. The pooled RR for necessity for additional MVA was 7.112 [95%CI 2.817-17.958 P＜0.001]. Of all five studies, the results were consistent with the pooled data, while one study found an RR of 8.196 [95%CI 0.447-150.370 P=0.156], which was not statistically significant. The forest plot of necessity for additional MVA was shown in Figure 3B.

Subgroup analysis

We performed a meta-analysis to determine the pooled RR (misoprostol vs MVA) of complete uterine evacuation in patients with first-trimester incomplete miscarriage after misoprostol and MVA treatment to reflect their relative curative efficiency. The pooled RR of complete uterine evacuation was 0.93 [95% CI 0.88-0.98 P=0.013] which indicated that MVA achieved higher efficiency of complete uterine evacuation in patients with first-trimester incomplete miscarriage in general. With a purpose to explore the RR (misoprostol vs MVA) in different stratification in terms of age, gestational age, parity, etc., we performed a subgroup analysis to determine whether misoprostol was able to outperform MVA in some subgroups. We chose age, gestational age, parity, marital status, and misoprostol dosage as the variables. We stratified age into three subgroups: less than 24 years old, between 24 and 26 years old, and over 26 years old. According to Figure 4A, the pooled RR for patients less than 24 years old was 1.05 [95% CI 0.98-1.14], the pooled RR for patients between 24 and 26 years old was 0.90 [95% CI 0.85-0.95 P=0.359], and the pooled RR for patients over 26 years old was 0.92 [95% CI 0.84-0.99 P＜0.001]. The result indicated that patients over 24 years old were more probable to benefit from MVA in achieving complete uterine evacuation. Although the pooled RR for patients less than 24 years old was 1.05, there was only one study in this subgroup and the difference was not statistically significant, we did not conclude that patients less 24 years old were more suitable to choose misoprostol other than MVA. Based on Figure 4B, the pooled RRs in terms of different misoprostol dosage were 0.86 [95% CI 0.80-0.93], 0.93 [95% CI 0.87-0.99], and 0.99 [95% CI 0.97-1.00] when the doses of misoprostol were 400 μg, 600 μg, and 2200 μg, which indicated that MVA led to higher percentage of complete uterine evacuation than misoprostol when the latter was given with doses less than 2200 μg. In terms of gestational age, the RRs for patients with gestational age less than 9 weeks and over 9 weeks were 0.84 [95% CI 0.78-0.91] and 0.95 [95% CI 0.78-1.16] respectively, which illustrated that MVA was related with better outcomes of complete uterine evacuation regardless of gestational age (Figure 4C). According to Figure 4D, the proportion of married women was divided into three groups, less than 30%, between 30% and 60%, and over 60%, the RRs of which were 0.91 [95% CI 0.86-0.97], 0.92 [95% CI 0.81-1.03], and 0.90 [95% CI 0.74-1.08] respectively. This result demonstrated that MVA was recommended as a prior choice regardless of the marital status. The RRs of different parity were also calculated. According to Figure 4E, in misoprostol group, the pooled RR was 0.85 [95% CI 0.73-0.99] when parity＜1.5, 0.92 [95% CI 0.82-1.04] when 1.5≤parity＜2, 0.95 [95% CI 0.92-0.99] when parity≥2. In the meantime, Figure 4F found out that in MVA group, pooled RR was 0.93 [95% CI 0.83-1.03] when 1.5≤parity＜2 and 0.91 [95% CI 0.82-1.01] when parity≥2. These data illustrated that MVA was more recommended to achieve complete uterine evacuation regardless of parity. Generally speaking, subgroup analysis demonstrated that MVA was more efficient than misoprostol regardless of age, gestational age, parity, dosage of misoprostol, and marital status. Summary of the subgroup analysis was shown in Table 2.

Occurrence of adverse effects

All studies we included reported at least three kinds of adverse events, and the most frequently mentioned ones were abdominal pain, bleeding, chills, fever, nausea, and vomiting. The studies by Ani et al and Bique et al reported the general occurrence rate of adverse events. According to their results, 88.2% (Ani et al) and 98.2% (Bique et al) of patients treated by misoprostol experienced adverse events and for patients treated by MVA, the numbers were 57.4% (Ani et al) and 97.0% (Bique et al), with RRs of 1.5 [95%CI 1.28-1.84 P＜0.001] (Ani et al), and 1.01 [95%CI 0.97-1.06] (Bique et al) respectively.

Abdominal pain was recorded in five studies, the pooled analysis of which yielded that a general 67.8% of patients treated by misoprostol and 71.7% of patients treated by MVA expressed experience of abdominal pain. The RR was 0.80 [95%CI 0.63-1.01 P=0.063], indicating that the difference of occurrence rate of abdominal pain was not statistically significant (Figure 5A). Moreover, in the pooled analysis of abdominal pain, the study by Bique et al had a high relative weight of 86.71%, which might have a comparatively decisive influence on the final analysis. Meanwhile, two of the included studies found a significantly higher rate of abdominal pain following MVA with RRs of 0.566 [95%CI 0.389-0.822 P=0.003] and 0.337 [95%CI 0.254-0.447 P＜0.001]. However, one study came out with a higher probability to experience pain after misoprostol rather than MVA with an RR of 1.156 [95%CI 1.073-1.244 P＜0.001]. Six studies provided patients’ own grading or rating of the pain. In the pooled analysis, 76.9% of patients receiving misoprostol classified the pain as mild-tolerable, while 65.6% of patients following MVA treatment categorized the pain as mild-tolerable. The pooled RR was 1.06 [95%CI 0.73-1.53 P=0.77] which indicated a not significant lower occurrence rate of mild-tolerable pain after misoprostol intake.

Besides abdominal pain, bleeding was acknowledged as one of the most common adverse events following miscarriage. In this meta-analysis, all seven studies have reported the incidence of mild to severe bleeding. The pooled incidence of bleeding were 50.9% for misoprostol and 30.4% for MVA with an RR of 1.91 [95% CI 1.43-2.55 P＜0.001], indicating a statistically significant lower probability to experience bleeding if treated by MVA (Figure 5B). The highest RR in terms of bleeding was 5.94 [95%CI 0.73-48.47 P=0.096] reported by a study recruiting 102 patients and the lowest RR in terms of bleeding was 1.29 [95%CI 1.12-1.49 P＜0.001] reported by a study recruiting 327 patients. Comparatively obvious difference in the occurrence rate of chills after treatment of misoprostol and MVA was observed according to results of four studies. The pooled occurrence rate of chills was 23.2% for misoprostol and 1.6% for MVA with an RR of 7.50 [95%CI 1.41-39.83 P=0.018] (Figure 5C). All four studies which have reported the incidence of chills concluded an RR over or equal to 1, indicating higher vulnerability of chills following misoprostol treatment. The highest RR in terms of chills was 172.13 [95% CI 10.83-2736.34] recorded by Bique et al with patients’ number of 111 and 101 respectively. Fever was also recorded as one of the most frequent adverse events, especially after oral intake of misoprostol. Based on the pooled analysis, fever occurred to 16.8% of patients taking misoprostol and 2.9% receiving MVA. The pooled RR was 4.34 [95% CI 1.82-10.36 P=0.001] (Figure 5D). All four studies with data of post-treatment fever provided an RR over 1, which portended high occurrence rate of fever after misoprostol treatment. Nausea and vomiting were also noted as important adverse events. The pooled analysis indicated lower occurrence rate of nausea after MVA treatment than misoprostol treatment (21.5% for misoprostol and 11.3% for MVA). The pooled RR was 3.13 [95% CI 1.41-6.92 P=0.005]. The pooled analysis indicated lower occurrence rate of vomiting after MVA treatment than misoprostol treatment (7.7% for misoprostol and 3.3% for MVA). The pooled RR was 2.21 [95% CI 1.23-3.94 P=0.008].

Generally speaking, there was no significant difference in terms of the occurrence rate of abdominal pain or mild-tolerable pain between patients of first-trimester incomplete miscarriage receiving misoprostol and MVA treatments. However, we found a statistically lower occurrence rate of bleeding, chills, fever, nausea, and vomiting in patients of first-trimester incomplete miscarriage receiving MVA compared with those receiving misoprostol, which indicated a generally lower incidence rate of adverse events following MVA other than misoprostol.

Subjective evaluation from the patients

The studies we included also collected some subjective evaluation indicators from patients who have received either misoprostol or MVA. These subjective evaluations included general satisfactory rate, possibility of recommending misoprostol or MVA to relatives or friends, the worst adverse event, etc. In terms of general satisfactory rate, 90.9% of patients receiving misoprostol and 87.0% of patients receiving MVA expressed their feeling as satisfactory. The RR rate for general satisfaction (misoprostol vs MVA) was 0.989 [95%CI 0.969-1.010 P=0.319], which indicated there was no significant difference regarding patients’ satisfactory degree between misoprostol and MVA group. Among the five studies which have reported the satisfactory rate, the study by Dao et al found a statistically higher satisfactory rate in patients who received MVA with an RR of 0.919 [95%CI 0.848-0.995 P=0.036]; the study by Bique et al found a statistically higher satisfactory rate in patients who received misoprostol with an RR of 2.361 [95%CI 1.808-3.083 P＜0.001]. An approximate 89.4% of patients receiving misoprostol and 73.7% of patients receiving MVA would recommend corresponding treatment to their relatives and friends respectively, with an RR of 1.146 [95%CI 1.105-1.189 P＜0.001], which indicated a higher probability of recommendation in patients receiving misoprostol. When asked about the worst adverse event, 22.3% of patients receiving misoprostol and 24.6% of patients receiving MVA considered pain as the worst adverse event with an RR of 0.925 [95%CI 0.756-1.132 P=0.448]. Therefore, the proportion of patients considering pain as the worst treatment-related adverse event did not differ significantly between misoprostol and MVA group. Meanwhile, 11.8% of patients receiving misoprostol and 3.6% of patients receiving MVA considered bleeding as the worst adverse event with an RR of 3.073 [95%CI 2.006-4.708 P＜0.001]. Based on the data, we found a higher proportion of patients receiving misoprostol considered bleeding as the worst adverse event compared with patients receiving MVA.

Meta-regression

We performed a meta-regression to determine the potential heterogeneity among included studies and to look for items which might have contributed to the general heterogeneity in terms of complete uterine evacuation by misoprostol and MVA in first-trimester incomplete miscarriage. Indicators including year of publication, country where the study was carried out, age, gestational age, marital status and parity of patients receiving misoprostol or MVA, as well as dosage of misoprostol were considered. The meta-regression manifested that publication year (t=1.53, P=0.16), country where the study was initiated (t=2.16, P=0.08), age (t=-1.14, P=0.037), gestational age (t=1.13, P=0.375), percentage of patients who were married (t=-0.14, P=0.895), parity of patients in misoprostol group (t=0.94, P=0.415), parity of patients in MVA group (t=0.32, P=0.767) and dosage of misoprostol (t=1.26, P=0.263) were probably not potential contributors to the general heterogeneity of RR of complete uterine evacuation in treating first-trimester incomplete miscarriage. The details of meta-regression analysis were shown in Table 3.

Publication bias

We generated funnel plots to intuitively observe the potential publication bias based on the symmetry of the funnel plot. The funnel plots of RRs regarding complete uterine evacuation (Figure 6A) and necessity for additional MVA (Figure 6B) were considered roughly symmetrical, which reflected low possibility of publication bias. Meanwhile, we applied Egger’s test to investigate the existence of publication bias in a quantitative manner. The result of Egger’s test showed there was no sign of publication bias with t=-2.18 and P=0.081 in terms of RR of complete uterine evacuation. Egger’s test also indicated low probability of publication bias in terms of RR of necessity for additional MVA with t=1.08 and P=0.359. The plots for Egger’s tests were shown in Figure 6C-D. Therefore, we generally considered that there was no publication bias regarding RRs of both complete uterine evacuation and necessity for additional MVA.

Although it is widely recommended that MVA is the prioritized choice in patients with incomplete miscarriage for better performance of complete uterine evacuation, standard treating procedures, and low risks of adverse events, the promotion of MVA is still limited in regions with restricted resources, insufficient medical staff, and equipment. Misoprostol is a promising medical treatment option as an alternative of MVA for its satisfactory rate of complete uterine evacuation, and tolerable adverse events. So far, several studies have been launched to appraise the feasibility of misoprostol as an alternative of MVA in conditions where medical equipment like ultrasonography or well-trained surgeons were not available, but the results of these studies varied from one another ^[13–19]. Consequently, we carried out this systematic review and meta-analysis to evaluate the relative efficacy, safety concerns and patients’ subjective evaluation of misoprostol and MVA.

In terms of efficacy in treating first-trimester incomplete miscarriage, our study found out higher efficiency of complete uterine evacuation after MVA (98.70%) than misoprostol (93.04%) with an RR of 0.972. Though MVA was superior to misoprostol in terms of success of complete uterine evacuation, we found the rates of complete uterine evacuation were both over 90% either after MVA or misoprostol treatment, which demonstrated that in case of limited resources settings, misoprostol could be used as an alternative of MVA on condition that post-treatment adverse events like bleeding or pain could be well managed. In a study carried out in Myanmar, the success rate of complete uterine evacuation after misoprostol intake in treating incomplete abortion was 75%, which was lower than the pooled rate of complete uterine evacuation we reported, this could be attributed to the dose, timing, and giving routes of misoprostol, and the post-abortion care provided ^[20]. Some experts launched studies to evaluate possible combination with misoprostol and other drugs to induce complete uterine evacuation. Shimels et al conducted a systematic review and meta-analysis to compare the relative efficiency in treating incomplete miscarriage by misoprostol plus mifepristone or by misoprostol alone, which demonstrated that combination of misoprostol and mifepristone achieved better outcomes ^[21]. Another study focused on the relative efficacy of misoprostol in treating second trimester abortion in terms of regimens. The study found out that the success rate of delivery at 12, 24, 36, and 48 hours were not statistically different in patients who received misoprostol (400 mcg) intravenously every four hours or orally every four hours, which indicated that various routes of giving misoprostol seemed to be able to achieve similar outcomes ^[22]. In 2022, a study performed in Uganda compared the outcomes of misoprostol in treating incomplete second-trimester abortion under the guidance by physicians or by midwives. The result indicated that the model-based risk difference was − 2.3% (midwife vs physician) ^[23]. In general, misoprostol could be a satisfactory alternative to MVA to treat first-trimester incomplete miscarriage in low resource settings under the guidance of obstetrical physicians despite that MVA did achieve higher rate of complete uterine evacuation.

This systematic review and meta-analysis also compared the occurrence rate of several common adverse events after misoprostol and MVA treatment, including abdominal pain, bleeding, chills, fever, nausea, and vomiting. Through our analysis, the occurrence rate of abdominal pain was not statistically different between misoprostol and MVA group. However, the occurrence rate of bleeding, chills, fever, nausea, and vomiting was comparatively lower in MVA group. Through the pooled analysis, over 50% of patients receiving misoprostol experienced bleeding in the course of 6 hours to 7 days, which was probably the result of exfoliation of endometrium and gestation sac ^[24]. As a result, strict surveillance of bleeding and routine evaluation by ultrasound was greatly recommended after misoprostol use to avoid unpredicted massive bleeding ^[25]. Besides bleeding, fever and chills were also commonly observed following misoprostol intake. According to previous report, 75% of patients would experience fever and chills 30 min after misoprostol intake ^[26]. The temperature of 97% of patients experiencing fever could be relieved within three hours after medication of antiallergic treatment or antipyretic, analgesic agents ^[27]. Therefore, intake of ibuprofen (600 mg) 1 hour before the use of misoprostol could be helpful in reducing both fever and abdominal pain ^[28]. Previous reports also indicated that ibuprofen (600 mg) could be taken every eight hours in severe cases to avoid high temperature and abdominal pain ^[29–31]. In addition, since misoprostol was a prostaglandin E1 analogue, it might have benefits in relieving gastric or duodenal ulcers ^{[32, 33]}. Based on previous reports, overdose of misoprostol could lead to diarrhea, vomiting, mydriasis, tremor, and high fever ^[34–37]. In conclusion, the occurrence rate of adverse events including bleeding, chills, fever, nausea, and vomiting was comparatively lower after MVA treatment than after misoprostol treatment. However, the adverse events could be prevented or well managed under routine surveillance or through combined therapy with other drugs.

As is widely recommended, MVA has several advantages. First of all, the negative pressure of MVA was only 80 mmHg, while the negative pressure of electric vacuum aspiration (EVA) was often as high as 400–500 mmHg ^{[38, 39]}. As a result, MVA caused less damage to uterine tissues and was associated with less bleeding. Secondly, plastic straw was applied in MVA procedure so that cervical dilation was not necessary. On the contrary, we needed to perform cervical dilation before EVA procedure, which might cause potential damage ^[40]. Moreover, according to published works, MVA could bring benefits of shorter operation time, less intraoperative bleeding, lower VAS score, lower occurrence rate of induced abortion syndrome ^[41–43]. According to our study, MVA outperformed misoprostol in achieving a higher rate of complete uterine evacuation in patients with first-trimester incomplete miscarriage, and a lower incidence rate of adverse events like bleeding, chills, fever, nausea, and vomiting. Therefore, it is still recommended that MVA be regarded as the first-line treatment given that the medical resources were abundant. However, in terms of patients’ subjective evaluation, the rate of giving satisfactory remarks did not differ between MVA and misoprostol group, but a statistically higher proportion of patients receiving misoprostol tended to recommend the treatment to their friends.

However, we did acknowledge there existed several shortcomings in this study. Firstly, the number of articles included for further analysis was relatively small. If more clinical trials were launched, we should be able to include more studies to reach a more comprehensive and well-rounded conclusion. Secondly, most of the studies we included came from countries in the same region, which might have contributed to potential bias. Therefore, in the future we’d like to incorporate more high-quality clinical trials from a diverse region to evaluate the performance of MVA and misoprostol more comprehensively in first-trimester incomplete miscarriage.

Through this systematic review and meta-analysis, we found the efficiency to achieve complete uterine evacuation in first-trimester incomplete miscarriage by MVA (98.70%) was superior to that by misoprostol (93.04%) with lower occurrence rate of bleeding, fever, chills, nausea, and vomiting. In terms of subjective evaluation, there was no statistical difference regarding the general satisfactory rate of misoprostol and MVA, while more patients receiving misoprostol seemed to be willing to recommend the treatment to their friends and more patients receiving misoprostol considered bleeding as the worst adverse event following treatment. In addition, given the high rate of complete uterine evacuation (༞90%) of both misoprostol and MVA group, as well as the controllability of adverse events, we supposed that misoprostol could serve as an alternative when MVA was unavailable in low medical resources settings.

This research was approved by the Ethics Committee of West China Second University Hospital, Sichuan University. This research was funded by National Key Research and Development Program of China (2018YFC1004603), Sichuan Science and Technology Program (2020YFQ0006) to Dr. Xinghui Liu. All authors agreed with the publication of this article. None of the authors declared any conflict of interest. All data generated or analyzed during this study are included in this published article and its supplementary information files.

Author contributions

(I) Conception and design: Hongyu Jin, Man Zhang, Xinghui Liu.

(II) Administrative support: Hongyu Jin, Xinghui Liu.

(III) Provision of study materials or patients: Hongyu Jin, Xu Yang, Muhelisa Muhetaer.

(IV) Collection and assembly of data: Hongyu Jin, Man Zhang, Jianhong Liu, Yali Chen, Yujie Wu.

(V) Data analysis and interpretation: Man Zhang, Muhelisa Muhetaer, Yali Chen.

(VI) Manuscript writing: Hongyu Jin, Man Zhang.

(VII) Final approval of manuscript: Xinghui Liu.

Kim C, Barnard S, Neilson JP, et al. Medical treatments for incomplete miscarriage [J]. Cochrane Database Syst Rev. 2017;1(1):Cd007223.
Ng BK, Annamalai R, Lim PS, et al. Outpatient versus inpatient intravaginal misoprostol for the treatment of first trimester incomplete miscarriage: a randomised controlled trial [J]. Arch Gynecol Obstet. 2015;291(1):105–13.
Strelow M, Maissiat J, Savaris MS, et al. Lower and extended dosage of misoprostol for cervical ripening in 1st trimester miscarriage (MISO200): A randomized clinical trial [J]. Eur J Obstet Gynecol Reprod Biol. 2022;269:30–4.
Linehan LA, San Lazaro Campillo I, Hennessy M, et al. Reproductive outcomes following recurrent first-trimester miscarriage: a retrospective cohort study [J]. Hum Reprod Open. 2022;2022(4):hoac045.
Chung JPW, Li Y, Law TSM, et al. Ultrasound-guided Manual Vacuum Aspiration is an optimal method for obtaining products of conception from early pregnancy loss for cytogenetic testing [J]. Int J Biochem Cell Biol. 2022;147:106226.
Hayes-Ryan D, Meaney S, Byrne S, et al. Womens experience of Manual Vacuum Aspiration: An Irish perspective [J]. Eur J Obstet Gynecol Reprod Biol. 2021;266:114–8.
Choobun T, Khanuengkitkong S, Pinjaroen S. A comparative study of cost of care and duration of management for first-trimester abortion with manual vacuum aspiration (MVA) and sharp curettage [J]. Arch Gynecol Obstet. 2012;286(5):1161–4.
Huchon C, Drioueche H, Koskas M, et al. Operative Hysteroscopy vs Vacuum Aspiration for Incomplete Spontaneous Abortion: A Randomized Clinical Trial [J]. JAMA. 2023;329(14):1197–205.
Kakinuma T, Kakinuma K, Kaneko A, et al. Safety and efficacy of manual vacuum aspiration under local anesthesia compared to general anesthesia in the surgical management of miscarriage: a retrospective cohort study [J]. Patient Saf Surg. 2022;16(1):16.
Pongsatha S, Suntornlimsiri N, Tongsong T. Effectiveness and adverse effects of vaginal misoprostol as a single agent for second trimester pregnancy termination: the impact of fetal viability [J]. Arch Gynecol Obstet; 2023.
Reischer T, Limbach I, Catic A et al. Factors Influencing the Duration of Termination of Pregnancy for Fetal Anomaly with Mifepristone in Combination with Misoprostol [J]. J Clin Med, 2023, 12(3).
Johnson DM, Michels-Gualtieri M, Gomperts R, et al. Safety and effectiveness of self-managed abortion using misoprostol alone acquired from an online telemedicine service in the United States [J]. Perspect Sex Reprod Health. 2023;55(1):4–11.
Ani VC, Enebe JT, Dim CC, et al. Sublingual misoprostol versus manual vacuum aspiration for treatment of incomplete abortion in Nigeria: a randomized control study [J]. Pan Afr Med J. 2022;41:90.
Nwafor JI, Agwu UM, Egbuji CC, et al. Misoprostol versus manual vacuum aspiration for treatment of first-trimester incomplete miscarriage in a low-resource setting: A randomized controlled trial [J]. Niger J Clin Pract. 2020;23(5):638–46.
Ibiyemi KF, Ijaiya MA, Adesina KT. Randomised Trial of Oral Misoprostol Versus Manual Vacuum Aspiration for the Treatment of Incomplete Abortion at a Nigerian Tertiary Hospital [J]. Sultan Qaboos Univ Med J. 2019;19(1):e38–e43.
Dabash R, Ramadan MC, Darwish E, et al. A randomized controlled trial of 400-µg sublingual misoprostol versus manual vacuum aspiration for the treatment of incomplete abortion in two Egyptian hospitals [J]. Int J Gynaecol Obstet. 2010;111(2):131–5.
Dao B, Blum J, Thieba B, et al. Is misoprostol a safe, effective and acceptable alternative to manual vacuum aspiration for postabortion care? Results from a randomised trial in Burkina Faso, West Africa [J]. BJOG. 2007;114(11):1368–75.
Bique C, Ustá M, Debora B, et al. Comparison of misoprostol and manual vacuum aspiration for the treatment of incomplete abortion [J]. Int J Gynaecol Obstet. 2007;98(3):222–6.
Weeks A, Alia G, Blum J, et al. A randomized trial of misoprostol compared with manual vacuum aspiration for incomplete abortion [J]. Obstet Gynecol. 2005;106(3):540–7.
Mary M, Tin KN, Maung TM, et al. Introducing misoprostol for treatment of incomplete abortion: A feasibility and acceptability study in secondary-level health facilities in Myanmar [J]. Sex Reprod Healthc. 2023;36:100825.
Shimels T, Getnet M, Shafie M, et al. Comparison of mifepristone plus misoprostol with misoprostol alone for first trimester medical abortion: A systematic review and meta-analysis [J]. Front Glob Womens Health. 2023;4:1112392.
Pongsatha S, Suntornlimsiri N, Tongsong T. Outcomes of Pregnancy Termination of Dead Fetus in Utero in Second Trimester by Misoprostol with Various Regimens [J]. Int J Environ Res Public Health, 2022, 19(19).
Atuhairwe S, Byamugisha J, Kakaire O, et al. Comparison of the effectiveness and safety of treatment of incomplete second trimester abortion with misoprostol provided by midwives and physicians: a randomised, controlled, equivalence trial in Uganda [J]. Lancet Glob Health. 2022;10(10):e1505–e13.
Podolskyi V, Gemzell-Danielsson K, Maltzman LL, et al. Effectiveness and acceptability of home use of misoprostol for medical abortion up to 10 weeks of pregnancy [J]. Acta Obstet Gynecol Scand. 2023;102(5):541–8.
Liu N, Ray JG. Short-Term Adverse Outcomes After Mifepristone-Misoprostol Versus Procedural Induced Abortion : A Population-Based Propensity-Weighted Study [J]. Ann Intern Med. 2023;176(2):145–53.
Macnaughton H, Nothnagle M, Early J. Mifepristone and Misoprostol for Early Pregnancy Loss and Medication Abortion [J]. Am Fam Physician. 2021;103(8):473–80.
Durocher J, Aguirre JD, Dzuba IG, et al. High fever after sublingual administration of misoprostol for treatment of post-partum haemorrhage: a hospital-based, prospective observational study in Argentina [J]. Trop Med Int Health. 2020;25(6):714–22.
Cooper SA, Cowan A, Tallarida RJ, et al. The Analgesic Interaction of Misoprostol with Nonsteroidal Anti-Inflammatory Drugs [J]. Am J Ther. 1996;3(4):261–7.
Munger MA, Nelson SD, Teng CC, et al. Reduced risk of NSAID-Induced adverse events with concomitant use of misoprostol (MICRO study) [J]. Pharmacotherapy. 2022;42(7):540–8.
Dragoman MV, Grossman D, Nguyen MH, et al. Two prophylactic pain management regimens for medical abortion ≤ 63 days' gestation with mifepristone and misoprostol: A multicenter, randomized, placebo-controlled trial [J]. Contraception. 2021;103(3):163–70.
Fiala C, Agostini A, Bombas T, et al. Management of pain associated with up-to-9-weeks medical termination of pregnancy (MToP) using mifepristone-misoprostol regimens: expert consensus based on a systematic literature review [J]. J Obstet Gynaecol. 2020;40(5):591–601.
Abdel Hamid AS, El Zeneiny H, Fathy A, et al. A pilot study to compare propranolol and misoprostol versus misoprostol and placebo for induction of labor in primigravidae; a randomized, single-blinded, placebo-controlled trial [J]. BMC Pregnancy Childbirth. 2023;23(1):226.
Taha AS, McCloskey C, McSkimming P, et al. Misoprostol for small bowel ulcers in patients with obscure bleeding taking aspirin and non-steroidal anti-inflammatory drugs (MASTERS): a randomised, double-blind, placebo-controlled, phase 3 trial [J]. Lancet Gastroenterol Hepatol. 2018;3(7):469–76.
Maior M, Souza ASR, Souza GFA, et al. Comparison between 200 µg and 800 µg of vaginal misoprostol for cervical ripening before operative hysteroscopy: A randomized controlled trial [J]. Int J Gynaecol Obstet. 2022;158(1):205–12.
Abdelhakim AM, Gadallah AH, Abbas AM. Efficacy and safety of oral vs vaginal misoprostol for cervical priming before hysteroscopy: A systematic review and meta-analysis [J]. Eur J Obstet Gynecol Reprod Biol. 2019;243:111–9.
Shakir-Reese JM, Ye PP, Perritt JB, et al. A factorial-design randomized controlled trial comparing misoprostol alone to Dilapan with misoprostol and comparing buccal to vaginal misoprostol for same-day cervical preparation prior to dilation & evacuation at 14 weeks 0 days-19 weeks 6 days gestation() [J]. Contraception. 2019;100(6):445–50.
Kaiser J, Royer PA. Profound Hyperthermia After Postpartum Rectal Misoprostol Administration [J]. Obstet Gynecol. 2016;127(6):1067–9.
Ansari A, Abbas S. Manual Vacuum Aspiration (MVA) - A safe option for evacuation of first trimester miscarriage in cardiac patients [J]. J Pak Med Assoc. 2017;67(6):948–50.
Grentzer J, McNicholas C, Eisenberg DL, et al. Comparison of procedure time between manual and electric vacuum aspiration for pregnancy termination between 10–14 weeks: A randomized trial [J]. Contraception. 2022;113:108–12.
Kakinuma T, Kakinuma K, Sakamoto Y, et al. Safety and efficacy of manual vacuum suction compared with conventional dilatation and sharp curettage and electric vacuum aspiration in surgical treatment of miscarriage: a randomized controlled trial [J]. BMC Pregnancy Childbirth. 2020;20(1):695.
Creinin MD. Manual Compared With Electric Vacuum Aspiration for Treatment of Molar Pregnancy [J]. Obstet Gynecol. 2018;132(1):217–8.
Dean G, Colarossi L, Porsch L, et al. Manual compared with electric vacuum aspiration for abortion at less than 6 weeks of gestation: a randomized controlled trial [J]. Obstet Gynecol. 2015;125(5):1121–9.
Tasnim N, Mahmud G, Fatima S, et al. Manual vacuum aspiration: a safe and cost-effective substitute of electric vacuum aspiration for the surgical management of early pregnancy loss [J]. J Pak Med Assoc. 2011;61(2):149–53.

Table 1 is available in the Supplementary Files section.

Table 2. Summary of subgroup analysis.

Covriates	Subgroup	No. of studies	Meta-analytic summary estimate
Covriates	Subgroup	No. of studies	Pooled RR	I²	P value
Age	Less than 24 years	1	1.05 [95% CI 0.98-1.14]	/	P=0.013
	24-26 years	2	0.90 [95% CI 0.85-0.95]	0.0%
	Over 26 years	4	0.92 [95% CI 0.84-0.99]	93.7%
	Overall	7	0.93 [95% CI 0.88-0.98]	89.2%
Gestational age	Less than 9 weeks	2	0.84 [95% CI 0.78-0.91]	0.0%	P=0.090
	Over 9 weeks	2	0.95 [95% CI 0.78-1.16]	92.0%
	Overall	4	0.90 [95% CI 0.80-1.02]	85.0%
Marital status	Married%＜30%	1	0.91 [95% CI 0.86-0.97]	/	P=0.007
	30≤Married%＜60	2	0.92 [95% CI 0.81-1.03]	63.8%
	60≤Married%≤100	3	0.90 [95% CI 0.74-1.08]	96.3%
	Overall	6	0.91 [95% CI 0.85-0.97]	92.1%
Parity of patients of misoprostol group	Parity＜1.5	1	0.85 [95% CI 0.73-0.99]	/	P=0.015
	1.5≤Parity＜2	3	0.92 [95% CI 0.82-1.04]	93.7%
	Parity≥2	1	0.95 [95% CI 0.92-0.99]	/
	Overall	5	0.92 [95% CI 0.87-0.98]	89.8%
Parity of patients of MVA group	1.5≤Parity＜2	3	0.93 [95% CI 0.83-1.03]	89.7%	P=0.015
	Parity≥2	2	0.91 [95% CI 0.82-1.01]	83.0%
	Overall	5	0.92 [95% CI 0.87-0.98]	89.8%
Dosage of misoprostol	400 μg	1	0.86 [95% CI 0.80-0.93]	/	P=0.013
	600 μg	5	0.93 [95% CI 0.87-0.99]	78.7%
	2200 μg	1	0.99 [95% CI 0.97-1.00]	/
	Overall	7	0.93 [95% CI 0.88-0.98]	89.2%

Table 3. Results of meta-regression analysis showing the potential heterogeneity among studies.

Covriates	Subgroup	No. of studies	Meta-analytic summary estimate
Covriates	Subgroup	No. of studies	Pooled RR	t value	P value
Publication year	2000-2009	3	1.12 [95% CI 0.91-1.40]	t=1.53	P=0.16
	2010-2019	2	1.07 [95% CI 0.85-1.36]
	2020-now	2	0.86 [95%CI 0.72-1.03]
Country	Nigeria	3	0.81 [95% CI 0.66-0.99]	t=2.16	P=0.08
	Egypt	1	0.94 [95% CI 0.79-1.11]
	Burkina Faso	1	0.91 [95% CI 0.75-1.09]
	Mozambique	1	0.87 [95% CI 0.70-1.07]
	Uganda	1	1.05 [95% CI 0.89-1.24]
Age of patients of misoprostol group	Less than 24 years	1	1.18 [95%CI 0.91-1.54]	t=-1.14	P=0.307
	24-26 years	2	0.89 [95% CI 0.76-1.04]
	Over 26 years	4	1.03 [95% CI 0.86-1.25]
Age of patients of MVA group	Less than 24 years	1	1.18 [95%CI 0.91-1.54]	t=-1.14	P=0.307
	24-26 years	2	0.89 [95% CI 0.76-1.04]
	Over 26 years	4	1.03 [95% CI 0.86-1.25]
Gestational age of patients of misoprostol group	Less than 9 weeks	2	0.88 [95% CI 0.55-1.41]	t=1.13	P=0.375
Gestational age of patients of misoprostol group	Over 9 weeks	2	0.95 [95% CI 0.70-1.31]	t=1.13	P=0.375
Gestational age of patients of MVA group	Less than 9 weeks	2	0.88 [95% CI 0.55-1.41]	t=1.13	P=0.375
Gestational age of patients of MVA group	Over 9 weeks	2	0.95 [95% CI 0.70-1.31]	t=1.13	P=0.375
Marital status of patients of misoprostol group	Married%＜30%	1	0.91 [95% CI 0.69-1.20]	t=-0.14	P=0.895
	30≤Married%＜60	2	1.00 [95% CI 0.71-1.42]
	60≤Married%≤100	3	0.99 [95% CI 0.72-1.36]
Marital status of patients of misoprostol group	Married%＜30%	1	0.91 [95% CI 0.69-1.20]	t=-0.14	P=0.895
	30≤Married%＜60	2	1.00 [95% CI 0.71-1.42]
	60≤Married%≤100	3	0.99 [95% CI 0.72-1.36]
Parity of patients of misoprostol group	Parity＜1.5	1	0.85 [95% CI 0.56-1.30]	t=0.94	P=0.415
	1.5≤Parity＜2	3	1.09 [95%CI 0.69-1.72]
	Parity≥2	1	1.12 [95% CI 0.68-1.86]
Parity of patients of MVA group	1.5≤Parity＜2	3	1.02 [95%CI 0.84-1.24]	t=0.32	P=0.767
Parity of patients of MVA group	Parity≥2	2	0.91 [95% CI 0.79-1.06]	t=0.32	P=0.767
Dosage of misoprostol	400 μg	1	0.88 [95% CI 0.63-1.22]	t=1.26	P=0.263
	600 μg	5	0.94 [95% CI 0.73-1.20]
	2200 μg	1	0.99 [95% CI 0.79-1.23]

No competing interests reported.

Table1.docx

Comparison of misoprostol and manual vacuum aspiration in treating first-trimester incomplete miscarriage: a systematic review and meta-analysis of randomized controlled trials

Status:

Version 1

Abstract

Figures

Background

Methods

Results

Discussion

Conclusion

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Version 1