Impact of medical, surgical and expectant management on spontaneous miscarriage/abortion on first trimester: A systematic review and meta-analysis of randomized, controlled trials

An increase in miscarriage in the first trimester of gestation and its associated complication is burden-some on the quality of life of a woman. Medical, surgical, and expectant care are carried out after the miscarriage to remove any remaining tissues in the uterus. Understanding the efficacy and safety of these interventions will raise awareness and be a deciding factor to choose an appropriate treatment plan. Present review aims to determine the efficacy and safety of medical, surgical, and expectant care of various medical and surgical methods for first-trimester miscarriage. This review included studies that allocated women to medical, surgical or expectant management in the first trimester. PubMed, Cochrane Library, MEDLINE, and Embase Library were searched for the literature. The primary outcome was the complete evacuation of products of conception. Data were independently reviewed, graded for evidence quality, and assessed for risk bias by using the guidelines of PRISAM (Preferred Report Items for Systematic Review and Meta-Analysis). 21 eligible articles were included in this systematic review, comprising of 7931 patients undergoing medical, surgical or expectant-management for early spontaneous-miscarriage. The success rate in surgical intervention was higher when compared with medical intervention (OR: 16.12 [9.11, 28.52]) and expectant management (OR: 2.78 [2.13, 3.61]). Whereas medical intervention had a high success rate when compared with expectant-management (OR: 4.29 [2.31, 7.97]). The review determines the effect of medical, surgical, and expectant-management procedures on women who have had spontaneous-miscarriages in their first-trimester. PROSPERO-International prospective register of systematic reviews–CRD42020154395


Introduction
A miscarriage is a common occurrence defined as a nonviable pregnancy with an empty/incomplete gestational sac, an embryo without cardiac action, or a gestational trophoblastic illness with molar placental degradation.It occurs in 15% to 20% of pregnancies, according to estimates.Approximately 80% of these spontaneous miscarriage pregnancies occur between the first and thirteenth weeks of gestation, with the risk decreasing after 12 weeks.Most patients are unaware of how frequently spontaneous miscarriages occur in the first trimester, which can lead to anxiety (30%), post-traumatic stress disorder (34%), and sadness (10%), all of which can disrupt mental harmony [1][2] [3] As a preventive measure for the evacuation of the retained products of conception in missed miscarriage and incomplete miscarriage, therapeutic alternatives such as surgical evacuation, expectant management, and medicinal management are used [4] .Vacuum aspiration is a type of surgical uterine evacuation that involves a vacuum source.It is also known as suction curettage, endometrial aspiration, or mini-suction.It is possible to utilize a handheld vacuum syringe or mechanical pump that is operated by foot (Manual Vacuum Aspiration) or electricity (Electric Vacuum Aspiration) [5] .Sharp metal curettage (also known as dilatation and curettage) is commonly performed in an operating room while the patient is sedated or under a general or regional anesthetic [6] .
Miscarriage medications typically involve synthetic prostaglandins such as Misoprostol, which is used primarily in incomplete miscarriages.Gemeprost and Dinoprost are less prevalent.Mifepristone, a progesterone antagonist, is used in conjunction with misoprostol to treat early miscarriage, particularly missed/silent miscarriage.Misoprostol, a safe and cheap medication, may allow for early POC ejection while avoiding complications [7][8] .The approach of expectant management allows the retained tissues of gestation to usually pass naturally, outside the hospital, and is an alternative to standard treatment with medication or surgery [9] .
Surgical procedure has a 95% success rate for missed abortion but an important unresolved issue is the cost of surgery and the risks associated with anesthesia [5] .Medical management of miscarriages has been demonstrated to be advantageous, particularly in women who have had a missed miscarriage or an empty sac.Misoprostol, on the other hand, is not approved for usage in all countries [10] .If a miscarriage is not handled, the fetal tissue will normally pass naturally, as it did for more than 65% of women who suffered a miscarriage.Unexpected hospitalizations and surgical curettage, on the other hand, occurred significantly more frequently during expectant and medicinal management than following surgical management [5][11] .
The main aim of this systematic review is to determine the efficacy and safety of medical, surgical, and expectant care of different medical and surgical methods for first-trimester miscarriage.

Methodology
The systematic review and meta-analysis were performed interpretation to the PRISMA and registered in Prospero CRD42020154395 [12][13] [14] .The PICO strategy (population, intervention, comparison, and outcome) was used to build the research question.Thus, this systematic review is required to clarify the safety, efficacy, and side effect of medical, surgical, and expectant management on first-trimester spontaneous miscarriage.

Eligibility
The review included original articles that evaluated the safety, efficacy, and side effect of pharmacological, surgical and expectant management on first-trimester spontaneous miscarriage.Studies that included patients do not receive medical, surgical and expectant interventions, review articles, letters to the editor; in vitro studies conference articles and case reports or series were excluded from this systematic review [15] .

a. Search strategy
A literature search on Medline/PubMed, Cochrane Library, MEDLINE, and Embase Library was performed using mesh terms mentioned in Supplementary material S1 and were searched [14][15] .Randomized case-control, cohort studies, and quasi-trials of women with first-trimester miscarriage were included, and directed a systematic review and meta- analysis generated both direct and mixed evidence on the effectiveness and side effects of medical, surgical, and expectant management.The selected articles through these databases were de-duplicated and the titles and abstracts of the articles were read independently by two of the authors using the software Rayyan.The studies which could potentially cover the inclusion criteria for this review were identified at this stage and accessed in their entirety.Cases of disagreement were resolved by consensus.

Data Extraction
Randomized trials, quasi-randomized studies, cohort study and case-control studies that evaluated medical treatment, surgical treatment and expectant treatment management of first-trimester miscarriage that was defined as a spontaneous loss of a non-viable intrauterine pregnancy between 0 and 13 th weeks gestation were included.Studies that evaluated combination of two treatment options (e.g.medical, expectant and surgical management) were included.Studies with multiple comparison arms were also included.We manually extracted data, using a excel sheet on: year and author, country of study, sample size, age, confounding factors, type of intervention, pre-outcomes and outcomes: success rate, bleeding, abdominal pain, and infection rate [14][15] .

Assessment of risk of bias in included studies
The risk of bias for the chosen studies was evaluated with Joanna Briggs Institute (JBI) criteria [16] .Two reviewers independently will decide whether there is a "High risk", "Low risk" or "unclear risk" of bias.The risk of bias will be ranked high when the study reached up to 49% of yes, moderate when it is (50-69) % and low when it is above or equal to 70%.

Statistical Analysis
The meta-analyses were performed for suitable outcomes using Review Manager Software 5.4.1.The odds Ratio (OR) was used as an effective measure for dichotomous variable outcomes in the study such as success rate, surgery required abdominal pain, blood diffusion, infection rate, nausea, and vaginal bleeding.The weighted mean difference was used for vaginal bleeding in days.The heterogeneity between the medical, surgical, and expectant studies was verified by the inconsistency test (I [2] ).I [2] values lower than 25% were considered low heterogeneity among the studies; values between 25 and 49% were considered moderate heterogeneity and values greater than 50% were considered high heterogeneity.
Statistical analyses were performed with Review Manager (RevMan) software version 5.4.1, and Comprehensive Meta-Analysis (CMA) software trial version (www.meta-analysis.com).

Risk of bias assessment
The risk of bias was estimated using the JBI scale; most studies showed low to moderate risk of bias.The lowest risk of bias was seen in study by Demetroulis [25] et al., and highest risk of bias was seen among Fernlund [26] et al.Most studies did not conduct statistical analysis for confounding factors.Blinding of participants and clinicians was not possible due to the type of intervention.The results of the quality assessment of the studies are shown in the Supplementary Table S2.

Meta-analysis
The results of meta-analysis for the outcomes are presented as forest plots in Figure 2. The forest plot indicated that the odds of success in surgical intervention was higher when compared with medical intervention (N= 4274, OR: 16.12 [9.11, I² = 0%).The studies showed that risk of abdominal pain was higher in medical when compared to surgical (OR:  Egger's regression method [21] (Egger test, P=0.621).

Discussion
Among the 21 selected studies, eleven studies compared medical intervention with surgical, three compared medical management with expectant management and eight studies compared surgical with expectant management for the management of spontaneous miscarriage in the first trimester.From the studies, it was observed that the success of complete abortion was higher in medical when compared to expectant whereas the medical treatment was inferior in comparison to surgical treatment.The reason for failure of abortion in medical vs surgical is due to the remaining residual sac which would require surgical evacuation [26][39] .
Though a higher success was observed in surgical trials, however the results of the trial a greater risk of infection following a surgical management with requirement for hospitalization when compared to medical or expectant management.There were no studies that compared the infection rate between medical management and expectant management.
Most common side effect observed was the risk of vaginal bleeding and abdominal pain among the patients before and after the management of miscarriage.The studies included collected history of vaginal bleeding and abdominal pain through self-report interviews or questionnaire.The pooled result of all the studies showed that the risk of vaginal bleeding was higher in the expectant group as this group needs to wait for the expulsion of the gestation tissue.The risk of abdominal pain was higher in the misoprostol group when compared to other intervention [43][44] .
The risk of bias assessment of all the studies included in the systematic review was generally low to moderate.Blinding of participants and clinicians was not possible in most of the studies.There was no clarity regarding the selective reporting bias as the trial protocols were not assessed.Loss to follow-up and exclusions after randomization were low [45] .
In present study we tried to minimize bias by assigning two independent reviewers to assess the eligibility for inclusion data extraction and assessed risk of bias independently.Data extraction was undertaken by one review author and checked by another.However, due to subjective assessments there might be some risk of bias.

Conclusion
Although it would be critical to have more data, the current evidence suggests medical treatment is superior to expectant care in terms of success rate and less frequent side effects and can be an alternative to surgery management of first trimester miscarriage.Study has identified high risk of abdominal pain with the use of medical intervention, vaginal bleeding requiring blood transfusion in expectant management and higher infection rate in surgical group requiring hospitalization or antibiotic regimen.These side-effects should be explained to the women during treatment counselling.
Further studies are required to compare the medical with expectant care.Future trials should consider women's views and quality of life measures alongside the clinical outcome.