Settings
This systematic review and metanalysis focused on related studies on calcium supplementation for improved pregnancy outcomes in hypertensive pregnancies in developing countries. This review was conducted according to the preferred reporting items for systematic reviews and meta-analyses (PRISMA). PICO framework was adopted to underpin the literature search. The studies reviewed were solely published RCTs in which pregnant women received calcium supplementation as compared to placebo or no intervention. This review was carried out on studies conducted in developing countries. “A developing country (or a low and middle-income country) is a country with a less developed industrial base and a low human development index (HDI ≤ 0.450) relative to other countries” (32). Even though not all developing countries are the same, the populations involved in the trials selected for this review had generally low calcium intake at baseline. Studies from developing countries were thus selected because of the high-risk population of pregnant women in these settings.
Searches
Study identification was done in two phases;
In phase one, a comprehensive search of PubMed, Cochrane library, Hinari, and Scopus database was carried out using different search terms to identify relevant keywords contained in the title, abstract and subject descriptors. The search terms and strategies were constructed based on PICO (Patient, Intervention, Comparator, and Outcome) framework as described in detail in appendix A. Synonyms of these search terms were constructed and used to perform an extensive search for literature in the various databases listed above. These strategies were modified to suit each search engine where appropriate. Reference lists of the articles found were screened for more relevant articles. The next stage was exclusion of articles that were not relevant to the topic as well as duplicate articles. The articles left were assessed for eligibility by reading their abstracts. Articles that were not significant for the study were excluded from the fully screened/hand searched articles. In the final stage, published articles that met the specific inclusion criteria were used for the meta-analyses.
In phase two, all previous systematic reviews on calcium supplementation in pregnant women published from 2000 to October 2020 were identified. Individual RCTs included in these previous reviews were selected. Then all individual RCTs on the same topic published between 1985 to October 2020 identified. The reference lists of the retrieved studies were also checked to identify more relevant publications. If more than one study was published by the same authors or institution, the publication year, study quality and sample size were considered, and one study was selected. A library was created for this review using Mendeley desktop (version 1.19.5) referencing software. The primary investigator conducted a comprehensive search and screening of the study titles from the abovementioned databases. All studies with eligible titles were exported to Mendeley library. After that all duplicates were removed before abstract screening. Two reviewers independently conducted the abstract screening. Full texts were retrieved if decisions could not be reached from information provided in the abstract using standardized tools, with guidance from the eligibility criteria. Disagreements regarding selection were resolved by consensus or discussion with a third reviewer. Reporting of findings were done using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement.
Study selection procedure
The review eligibility criteria were defined based on the Participants, Interventions/Exposure, Comparator, and Outcome (PICO) framework. This is shown below;
Population: Thereview focused on pregnant women in developing countries due to their high risk of hypertensive disorders during pregnancy. The study therefore included only RCTs involving pregnant women (especially those in areas with low dietary calcium intake) in developing countries.
Intervention: The primary interest was calcium supplementation or in combination with other supplements among pregnant women. The supplementation must have been administered orally before 32 weeks of gestation to ensure its effectiveness in preventing PIH.
Comparison: The comparator involved pregnant women who received placebo or no calcium supplementation.
Outcome: Improved maternal and perinatal outcomes in hypertensive pregnancies (pre-eclampsia, eclampsia, gestational hypertension or PIH, preterm delivery, low birth weight, and maternal and neonatal mortality).
Eligibility Criteria
Inclusion criteria
The inclusion criteria were based on PICO as follows:
- Only RCTs involving healthy pregnant women.
- All types of RCTs (being open, single-blind or double-blind).
- RCTs in which calcium supplementation and, or any other co-supplement started before 32 weeks of pregnancy at the latest.
- RCTs in which calcium supplementation was compared with placebo or no intervention.
- RCTs which presented evidence on at least one of the primary outcomes of interest (i.e. pre-eclampsia, eclampsia, gestational hypertension or PIH) in the presence of the intervention.
- RCTs which were conducted in developing countries and published in English language between 1985 and October 2020.
- RCTs accessible for the reviewers by full-text or abstract.
Exclusion criteria
The following exclusion criteria were used:
- RCTs in which the intervention was given as dietary calcium intake.
- RCTs which enrolled pregnant women diagnosed with pre-existing hypertension before the trial began.
- RCTs which were inaccessible neither by abstract nor full-text for the reviewers.
- All ongoing and unpublished RCTs.
Data Extraction
Data was extracted following a standard protocol and using standard data extraction forms by two reviewers. Following assessment of methodological quality, the papers were based on quantitative studies only. A data charting table was used to extract background information and process the information from each utilized study. The data extraction form was developed specifically for quantitative research data extraction based on the work of the Cochrane collaboration and the centre for reviews and dissemination as described in detail in appendix B. Figure 1 explains the multistep process for study selection and inclusion. All variables that focus on answering the research question were included. The research team independently conducted a trial data extraction and later discuss as a group to determine consistency of the data extraction approach with the research questions and objectives. The data extraction form was continually reviewed and updated in an iterative process. This improved the quality applicability, validity and consistency of the chart. Once this was completed, the primary author conducted the data extraction which was reviewed by the other authors. Any and all discrepancies were discussed and agreed upon in the final interpretation. All articles reviewed were assigned a unique code to help track all articles reviewed and those that were excluded during the data extraction/charting process. The data extraction form holds information: 1) General information about the study (author, year, country, study design, sample size, and study period); 2) Intervention and control used in the study (calcium and placebo); 3) General baseline characteristics of participants (age, population of disease risk, dosage of daily calcium intake, mean gestational age at starting calcium, etc.); 4) Outcomes of interest that relate to the meta-analyses (pre-eclampsia, eclampsia, and gestational hypertension). (See appendix D for more details on data extraction form).
The meta-analyses were performed using Stata (Version 16.0) with guidance from high-quality articles from published journals which provided framework for the statistical analysis. The primary outcomes assessed were pre-eclampsia, eclampsia and gestational hypertension while the secondary outcomes were maternal and neonatal mortality, preterm birth, and low birth weight, all due to HDP. Pooled analyses were conducted where data were available from more than one study for an outcome. Log risk ratio (RR) along with its variance and the 95% confidence interval (CI) were estimated for each study. Dichotomous outcomes were analysed using the relative risk (RR) with 95% CI, and continuous outcomes were analysed using the mean difference (MD) with 95% CI. If the literature only provided the median, range, and sample size, the algorithms were utilized to estimate the mean deviation and standard deviation.
The risk ratios were then directly pooled across studies using fixed-effects model (inverse-variance method) if heterogeneity was absent, otherwise a random-effects model was used to pool the data. Statistical heterogeneity was evaluated using Chi-square test (p<0.10 was considered as providing evidence of heterogeneity) and I2 statistic (values greater than or equal to 50% indicating significant heterogeneity). If heterogeneity was present (p<0.1 or I2 ≥50%), a source of heterogeneity was explored by fitting characteristics of subjects (mean age, mean gestational age), clinical data (dosage, and duration of supplement), and methodological characteristics (definition of outcome measurements, and setting of the study) in a meta-regression model one by one. Sensitivity analysis by excluding the outlier studies and/or a subgroup analysis according to that factor was performed.
Bearing in mind the association between the intervention (calcium supplement) and primary outcomes (pre-eclampsia and gestational hypertension) may be modified by different clinical characteristics, the following subgroup analyses were performed where data were available: calcium dose (high dose defined as ≤2.0 g/day and >1.2 g/day; moderate dose defined as ≤1.2 g/day and ≥0.6 g/day; low dose defined as <0.6 g/day); the mean gestational age at which calcium supplementation began and whether combined with other drugs or co-supplements; and risk of disease in population. Study participants at high risk of HDP were defined by the trial authors’ definitions and included as being teenagers, women with previous history of pre-eclampsia, women with increased sensitivity to angiotensin II, women with chronic hypertension, primiparous women, nulliparous women, or obese women. The sub-group analyses were based on evidence that calcium supplementation during pregnancy would be more effective for improved outcomes in hypertensive pregnancies. Publication bias was assessed using Egger’s test. P-values < 0.05 were considered as statistically significant, except for the test of heterogeneity where p < 0.10 was used. Subgroup analyses were limited to the primary outcomes.
Quality appraisal and data management processes
All the included trials were assessed for methodological quality and outcomes of interest using a standardized form. Data were abstracted for study design, study site, methods of sequence generation, allocation concealment, attrition and primary outcomes of interest. Individual studies were evaluated independently for methodological validity by two reviewers according to the Cochrane collaboration’s tool for assessing risk of bias in RCTs (33). Any disagreement that arises between the reviewers was resolved through discussion and with the assistance of a third reviewer where required.
Risk of bias and quality procedures
Study quality was independently assessed by two reviewers using the Cochrane collaboration tool for assessing risk of bias in RCTs (33). The following seven domains were evaluated; 1) selection bias (random sequence generation) 2) allocation concealment), 3) performance bias (blinding of participants and assessors), 4) detection bias (blinding of outcome assessment), 5) attrition bias (incomplete outcome data), 6) selective outcome reporting, and 7) other bias. Each item was classified as low, high, or an unclear risk of bias (if there was insufficient information) (see Appendix B for more details on risk of bias assessment).