Miscarriage Study Reveals Predictive Factors For Successful Management
Overview
Miscarriages occur in approximately 15.3% of pregnancies, with management options including expectant, medical, or surgical approaches. However, determining the best management strategy for individual patients is complex due to various influencing factors, making it difficult to navigate the existing literature when advising patients. This systematic review seeks to identify specific predictors that might influence the outcome of miscarriage management.
A comprehensive search was conducted across databases such as PubMed, Medline, and Google Scholar, covering all available literature up to April 2023. The review included studies that specifically examined predictors for the success of expectant or medical management of miscarriage. Studies that were of poor quality, review articles, trial protocols, and congress abstracts were excluded. Data collection followed PRISMA guidelines, and the quality of each study was evaluated using the QUIPS tool.
The review identified various predictors related to demographics, ultrasound findings, presenting symptoms, and biochemical markers. The 24 studies included in the review showed variability in how miscarriage was defined, the predictors that were reported, and the management outcomes assessed. Ten of these studies explored the influence of the type of miscarriage on the success of expectant or medical management. Most of these studies found that incomplete miscarriages had a higher success rate with expectant or medical management compared to missed miscarriages or anembryonic pregnancies.
The review concludes that it is possible to provide personalized miscarriage management advice based on specific case predictors. However, further research involving larger studies with consistent definitions of predictors, management strategies, and outcomes is necessary to enhance the support available to women during the decision-making process for miscarriage management.
Introduction
Miscarriage occurs in approximately 15.3% of all recognized pregnancies, translating to about 23 million miscarriages globally each year, or 44 every minute. Among those who experience miscarriage, around half have pregnancy tissue remaining in the uterus, known as a missed or incomplete miscarriage. These patients must choose between expectant, medical, or surgical management options.
Expectant management, recommended as the first-line approach by the National Institute for Health and Care Excellence (NICE), leads to complete miscarriage in roughly 50% of cases and can be continued as long as the patient prefers, provided there is no evidence of infection. Though expectant management is generally safe and cost-effective, its success rate varies widely, from 25% to 90% in different studies.
Medical management is the primary alternative to surgical intervention, with success rates ranging from 69% to 80% in various studies. One challenge in choosing between expectant and medical management is the lack of reliable clinical features to predict the likelihood of complete miscarriage.
Patient preference plays a crucial role in selecting the management approach unless surgical intervention is necessary due to complications such as hemorrhage or sepsis. The anticipated success rate significantly influences this decision. Research has shown that most women would opt for medical treatment if the success rate exceeds 65%.
In clinical practice, informed decision-making is essential, requiring that patients be provided with comprehensive information about the chances of success for each management option. Identifying clinical features that reliably predict the success of expectant or medical management could improve decision-making and potentially reduce the psychological burden associated with miscarriage.
This summary reviews the results of a systematic analysis of studies that investigate clinical features predictive of successful expectant or medical management of missed or incomplete miscarriages.
Method
This study adhered to the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) and was registered with the PROSPERO database under the number CRD42022333397. The only deviations from the initial protocol were the use of an alternative bias evaluation tool and an extension of the search date to April 2, 2023.
The literature search targeted articles published up to April 2, 2023, across several databases, including PubMed, Medline, and Google Scholar, to identify studies that evaluated predictors of miscarriage outcomes under expectant and medical management. PubMed served as the primary database, utilizing both controlled vocabulary (MeSH terms) and free-text keywords. This search was supplemented by a snowballing strategy, where references from primary sources and review articles were examined to ensure comprehensive coverage.
The study selection process involved a multi-stage screening by title, abstract, and full text, focusing on studies that explored defined predictors for successful miscarriage management. Only peer-reviewed human studies published in English were included, with exclusions made for studies of poor quality, review articles, trial protocols, and congress abstracts. The screening process was independently conducted by the first and second authors, with any discrepancies resolved through discussion until consensus was reached.
Data extraction was performed using a pre-developed spreadsheet, where the first author collected relevant data from the selected studies, which was then cross-verified by the second author. Due to the heterogeneity of the studies regarding predictors for miscarriage management outcomes, a meta-analysis was not feasible. Instead, the data were analyzed qualitatively and summarized in the results section. Each study’s quality was assessed using the QUIPS (Quality in Prognosis Studies) tool, with any disagreements on the extracted data resolved by consultation with a senior author.
This process ensured a rigorous and systematic approach to reviewing the literature on predictors of successful miscarriage management outcomes.
Result
The study results are organized across several tables for clarity: Table 1 provides an overview of the study duration, patient numbers, types of miscarriage, management strategies, objectives, success definitions, outcomes, and conclusions. Table 2 outlines the predictive factors for each management approach, while Table 3 details the inclusion and exclusion criteria. Table 4 presents the quality assessment of each included study using the QUIPS tool.
An initial search in PubMed, Medline, and Google Scholar yielded 2,311 records. Of these, 2,124 studies were excluded based on title alone. The remaining 187 abstracts were reviewed, with 124 excluded due to irrelevance, non-English language, or duplication. After further review of 63 full-text publications, 39 were excluded for not evaluating specific predictors of miscarriage management within a comparative cohort, leaving 24 studies for inclusion. These studies, published between 1999 and 2022, are summarized in a flow diagram.
The 24 studies included exhibited significant variation in their approach to evaluating predictors, management methods, and outcomes defined as successful. Among these, only one was a randomized controlled trial; the others were retrospective or prospective cohort studies. Expectant management was evaluated in 14 studies, and medical management in 11, with one study covering both. Follow-up durations ranged from 3 days to 8 weeks, and medical management regimens varied across studies, with patient numbers ranging from 31 to 451.
Demographic and Patient History Variables
Four studies examined maternal age as a predictor, with only Casikar et al. finding younger maternal age statistically significant in predicting the success of expectant management. Schreiber et al. found no significant link between ethnicity and miscarriage management outcomes, although women of Hispanic ethnicity were less likely to succeed with a single dose of misoprostol. Sonalkar et al. and Hamel et al. also found no significant differences across racial groups. Five studies analyzed gestational age by the last menstrual period, finding it had no impact on miscarriage resolution outcomes.
Casikar et al. and Fernlund et al. found no correlation between previous delivery types (vaginal or cesarean) and miscarriage management outcomes. Two studies found a significant relationship between higher parity and lower odds of successful medical management, while three others found no significant correlation. Three studies found that miscarriage history was not a significant predictor of management outcomes. Previous termination of pregnancy was found not to be a significant predictor in two studies. Conflicting results were found regarding smoking status; one study found it irrelevant, while another found non-smoking status to be a significant predictor of successful medical management.
Ultrasound Features
Seven studies looked at ultrasound predictors, with mixed results regarding gestational sac measurements and other parameters like retained pregnancy tissue diameter and endometrial cavity measurements. Ten studies analyzed outcomes based on miscarriage type. Some studies found a significant correlation between miscarriage type and management success, while others did not, likely due to heterogeneity in definitions and methods. Six studies evaluated blood flow as a predictor of outcomes. Some found significant correlations with vascularity, while others did not.
Presenting Symptoms
Nine studies examined vaginal bleeding, with two finding a significant relationship with expectant management success and five finding no significant correlation. Seven studies analyzed lower abdominal pain, with one study reporting a significant relationship with successful medical management, while the others did not.
Biochemical Markers
Five studies analyzed progesterone levels, with four finding significant differences between successful and unsuccessful outcomes, while one did not. Three studies found lower HCG levels to be significantly associated with successful outcomes, while one did not. Some studies analyzed additional markers like 17-hydroxy-progesterone and inhibin A, with mixed results regarding their predictive value.
In conclusion, the studies reviewed provide a comprehensive overview of the various factors that may predict the outcomes of different miscarriage management strategies, though results are often conflicting due to differences in study design, definitions, and methods.
Conclusion
This systematic review is the first to comprehensively analyze predictors of success in expectant and medical management of miscarriage. The review identifies several key variables that influence outcomes, categorized into four main areas: (1) Demographics and patient history, (2) Ultrasound findings, (3) Presenting symptoms, and (4) Biochemical markers.
A notable strength of this review is its broad scope, encompassing a wide range of previously investigated predictors. The methodology for assessing bias was robust, and the review included a substantial number of participants. However, the analysis revealed significant heterogeneity due to variations in definitions of miscarriage management success, management parameters, and predictor descriptions across studies, making meta-analysis unfeasible.
Despite this variability, the review highlights that certain variables can predict management outcomes under specific conditions. The high variability in study definitions emphasizes the need for international consensus on definitions for future research.
Among the predictors analyzed, ultrasound features were the most frequently studied. The review found inconsistencies in outcomes related to ultrasound parameters, such as the timing of scans and the classification of miscarriage types. For instance, incomplete miscarriages and missed miscarriages may show different outcomes depending on the point at which the pregnancy ceased development.
Demographic factors, including age, ethnicity, and smoking status, as well as a history of pregnancy terminations, were also explored. These factors potentially influence management success, with prior cervical interventions possibly affecting future treatment outcomes. Presenting symptoms like vaginal bleeding and pain can indicate the likelihood of spontaneous resolution, influencing management decisions.
Biochemical markers such as progesterone and HCG have established roles in early pregnancy and miscarriage prediction. Falling HCG levels can indicate miscarriage timing and management success, while other markers like Inhibin A and ADAM 12 might also be relevant. The multifactorial nature of miscarriage suggests that a combination of biomarkers could offer better predictive insights.
A tool that provides personalized predictions for miscarriage management could significantly benefit patients and healthcare systems. The economic impact of miscarriage is substantial, with an estimated short-term cost of £471 million annually in the UK. Additionally, the emotional toll of miscarriage and decision-making stress highlights the need for improved personalized counseling.
In conclusion, the review confirms that various predictors—encompassing demographics, ultrasound features, presenting symptoms, and biochemical markers—are associated with miscarriage management outcomes. However, the inconsistent definitions and study variations underscore the need for standardized approaches in future research to enhance personalized management and support for women experiencing miscarriage.
Oncology Related Tools
Other
Latest Research
