Article Text
Abstract
Introduction Preeclampsia is an important cause of maternal and fetal morbidity and mortality. Although the diagnostic and prognostic values of circulating placental growth factor (PlGF) have been extensively studied, urinary PlGF represents an excellent alternative to facilitate sample collection in the follow-up of pregnant women. The aim of this study is to determine whether urinary PlGF levels throughout pregnancy can predict severe maternal, fetal/placental and neonatal complications in women with hypertensive disorders in pregnancy.
Methods and analysis Studies that evaluated pregnant women with hypertensive disorders and at least one measurement of urinary PlGF will be included. Studies that measure urinary PlGF after the occurrence of the complications will be excluded. The main outcome will consist of severe maternal complications in women with hypertensive disorders in pregnancy. Secondary outcomes will consist of severe fetal/placental and neonatal complications as defined by the International Collaboration to Harmonize Outcomes for Pre-eclampsia. Prospective cohort studies and case–controls studies reporting original data will be included. Studies will be identified by searching MEDLINE and SCOPUS databases. The first literature search was conducted on 2 March 2020, and another search will be performed before analyses. All eligible studies will be assessed for risk of bias with a standardised 10-items study quality assessment tool adapted from the Study Quality Assessment tools developed by the National Institutes of Health (NIH). Summary of ORs and 95% CIs will be reported to evaluate the association between urinary PlGF levels and hypertensive disorders in pregnancy and its complications. A random-effect meta-analysis will also be performed.
Ethics and dissemination Review by an ethics committee will not be required for this systematic review. This study will follow the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines and will be submitted for publication in a peer-reviewed journal as well as for presentation at conferences targeting different stakeholders, including researchers, physicians and patients.
PROSPERO registration number CRD42020186313.
- hypertension
- maternal medicine
- physiology
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
Strengths and limitations of this study
This systematic review and meta-analysis will offer a synthesis and a comprehensive understanding of the work done to this day on urinary testing of placental growth factor (PlGF) for prediction of outcomes in hypertensive disorders in pregnancy.
This prospectively registered study will be rigorously conducted with the contributions of all authors sharing expertise in methodology and context.
A small number of studies and the heterogeneity of the data may be a limitation of this study.
This study may identify gaps in knowledge that could be addressed by future studies aimed at investigating the use of urinary PlGF in pregnancy.
Introduction
Preeclampsia is the most severe hypertensive disorder in pregnancy occurring in 2%–5% of all pregnancies around the globe.1 It is a leading cause of fetal and maternal morbidity and mortality and may cause severe complications.1 To help combine and compare the results of research on preeclampsia, the International Collaboration to Harmonize Outcomes for Pre-eclampsia (iHOPE) established a core outcome set that defines the maternal and fetal complications of preeclampsia.2
Even if the underlying causes of preeclampsia remain unknown, preeclampsia may be caused by a defect of circulating angiogenic factors leading to endothelial cells dysfunction.3 4 As such, it has been shown that women diagnosed with preeclampsia present increased circulating blood levels of the anti-angiogenic factor soluble fms-like tyrosine kinase 1 (sFlt-1).5 sFlt-1 normally binds angiogenic factors such as the vascular endothelial growth factor (VEGF) as well as the placental growth factor (PlGF). Increased circulating levels of sFlt-1 lead to a reduction in free PlGF and VEGF and prevents their action on endothelial cells generating endothelial dysfunction.6
Previous observational retrospective study suggested that the maternal blood levels of angiogenic and anti-angiogenic factors may be altered as soon as 5 weeks prior to the clinical manifestation of preeclampsia,7 or even as soon as the first trimester in pregnancy.5 Likewise, several investigators conducted clinical studies to evaluate the prognostic and diagnostic values of circulating angiogenic factors in hypertensive pregnant women.8–10 Circulating levels of PlGF can predict the occurrence of preeclampsia and its complications. It has been shown that circulating levels of PlGF are lower in patient with preeclampsia compared with healthy controlled patient (61.3±28.1 vs 122.4±81.0 pg/ml (p<0.001)).9 Current efforts have focused on the assessment of circulating levels of angiogenic factors. However, as these routine blood tests during pregnancy are time consuming, expensive and certainly uncomfortable to pregnant women, the development of a urinary test may be more convenient and acceptable.
Diagnostic and prognostic values of circulating angiogenic factors have been studied extensively in the past, but less is known about urinary PlGF diagnostic and prognostic values. PlGF is a low molecular weight protein of ~30 kDa11, which is filtered by the kidneys. Urinary levels of this protein highly correlate with its circulating blood levels (r=0.934),12 thus making urinary PlGF an interesting alternative for routine testing. This systematic review and meta-analysis will offer a synthesis and understanding of the work done to this day on urinary testing of PlGF for prediction of outcomes in hypertensive disorders in pregnancy.
Objective
The main objective of this study is to determine whether urinary PlGF levels throughout pregnancy can predict severe maternal, fetal/placental and neonatal complications in women with hypertensive disorders in pregnancy.
Method and analysis
Protocol and registration
The Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols13 were followed for the elaboration of this protocol (see online supplemental file 1). This systematic review was registered prospectively.
Supplemental material
Participants
The review will target studies that include pregnant women with hypertensive disorders in pregnancy and at least one measurement of urinary PlGF during pregnancy. Studies with measurements performed after the occurrence of the complications and postpartum will not be included. No exclusion criteria based on underlying maternal medical conditions will be applied.
Outcome measurement
Patient and public involvement
A panel of four experts in maternal and fetal medicine as well as a patient with lived experience of preeclampsia were consulted to determine and prioritise the outcomes of the study. A medical student was involved in reviewing the protocol for completeness and to ensure its clarity to non-expert readers.
The main outcome will consist of severe maternal complications in women with hypertensive disorders in pregnancy (as defined below). Secondary outcomes will consist of severe fetal/placental, neonatal complications.
The core outcomes set for preeclampsia developed by the iHOPE2 will be used as follow:
Maternal core outcomes as: maternal mortality, severe morbidity (eg, eclampsia, stroke, cortical blindness, retinal detachment, pulmonary oedema, acute kidney injury, liver capsule haematoma or rupture, placental abruption, postpartum haemorrhage, raised liver enzymes, low platelets, admission to intensive care unit required, intubation and mechanical ventilation (not for childbirth)).
Offspring outcomes as: stillbirth, gestational age at delivery (defined as delivery <37 weeks of gestation), birth weight, small-for-gestational-age (defined as a ≤10 growth centile), neonatal mortality, neonatal seizures, admission to neonatal unit required, respiratory support.
Considering the heterogeneity of the outcomes used in the different studies, we will also examine the outcomes defined in the Pre-eclampsia Integrated Estimate of RiSk14 study or their equivalent, including hemolysis, elevated liver enzymes and a low platelet count syndrome, disseminated intravascular coagulation, intrauterine growth restriction, abnormal fetal Doppler and oligohydramnios. Definitions of small-for-gestational-age will include the Gordijn and Beune definition as per the Delphi procedure.15 This suggestion was put forth by our expert panel for determination and prioritisation of outcomes for also applying more stringent criteria for placental insufficiency.
The following definitions will be used to classify hypertensive disorders in pregnancy16:
Gestational hypertension will be defined as de novo persistent hypertension after 20 weeks of pregnancy (blood pressure ≥140 mm Hg systolic or ≥90 mm Hg diastolic).
Chronic hypertension will be defined as hypertension (blood pressure ≥140 mm Hg systolic or ≥90 mm Hg diastolic) predated to pregnancy or discovered before 20 weeks of gestation.
Preeclampsia will be defined as gestational hypertension accompanied by one or more new-onset conditions among proteinuria as a protein/creatinine ratio ≥0.3 g/g, protein ≥1 g/L, 24 hours collection ≥0.3 g/day or one dipstick measurement ≥1+, acute kidney injury, liver involvement, neurological complications, haematological complications or uteroplacental dysfunction.
Superimposed preeclampsia will be defined as any of the maternal organ dysfunction of preeclampsia in a woman with chronic hypertension.
Type of studies
Prospective cohort studies and case–controls studies reporting original data will be included in this review. As the main objective is to search for a predictor, prospective studies are best designed for that purpose and though some cross-sectional studies may fulfil the requirement, it will be very difficult for retrospective studies to serve data for a predictive analysis. Literature reviews, case studies and case reports, letter to the editors, comments on article and editorials will be excluded. Study focusing only on animal research, on PlGF quantification in serum or plasma or study unrelated to pregnancy will also be excluded. Only studies published in French or in English will be included. No time restriction will be applied.
Search strategy
MEDLINE and SCOPUS will be searched for the systematic review and another search will be planned closer to the publication of the results. The first search was conducted on 2 March 2020, and another search will be performed before the analyses. The reference lists of the included studies will also be reviewed manually. The complete search strategy was reviewed and validated by a librarian and is provided as online supplemental file 2.
Supplemental material
Study records
Study selection
Inclusion criteria were fixed by two reviewers (CF, A-MC). Studies will be imported in Zotero V.5.0.89 and duplicates will be removed. Studies will be independently screened by two reviewers according to title and abstract (CF, JM). Full text will be assessed if the sorting is still unclear. Disagreement between the two reviewers will be resolved with the opinion of a third reviewer (A-MC). The sorting will be compiled in an Excel spreadsheet.
Data collection
The data will be extracted independently in duplicate by two reviewers using a prepiloted standardised data extraction form. The following data will be extracted from the publications: author, publication year, study group, laboratory, aim of the study, population, gestational age, definition of hypertensive disorders in pregnancy, preeclampsia, identified outcomes, laboratory method for determination of urinary PlGF, the brand of the PlGF assay kit, urinary PlGF results and proposed cut-off, if the urinary PlGF results were standardised according to the urinary creatinine, summary test characteristics (sensitivity, specificity, likelihood ratios, area under the receiver-operating characteristic (ROC) curve), OR. Disagreement between two reviewers will be resolved with the opinion of a third reviewer (A-MC). Missing data will be reported as well and the quality of the article will be assessed accordingly. The data will be compiled in an Excel spreadsheet.
Risk of bias assessment
A standardised 10-items study quality assessment tool adapted from the Study Quality Assessment tools developed by the National Institutes of Health (NIH) (https://www.nhlbi.nih.gov/health-topics/study-quality-assessment-tools) will be used by two independent observers. Each study included will be classified as either good, fair or poor quality. The strength of evidence between studies will be determine using the Grading of Recommendations, Assessment, Development and Evaluations (GRADE) system.17
Summarising data
Primary analysis will be conducted using only urinary PlGF standardised for urinary creatinine. Summary of OR and 95% CIs will be reported to evaluate the association between urinary PlGF levels and hypertensive disorders in pregnancy and its complications. If there is a sufficient number of studies, we will pool the results with a random-effects meta-analysis due to a priori concerns that not all included studies will assess comparable populations. Pooled effect sizes will be obtained using Mantel-Haenszel method, a more reliable approach than inverse-variance to determine study weights. Since a small number of studies presenting OR is expected, we decided to focus our analyses on effect size. However, studies presenting Relative Risk (RR) and Likelihood ratio (LR) could be considered for potential comparisons. Furthermore, if the number of valid studies allows it, hierarchical summary receiver-operating characteristic models will be presented as pooled accuracy measures and 95% CIs. Summary of sensitivity and specificity with 95% CIs will also be reported. For both pooled effects presented, heterogeneity will be assessed by Higgins’ and Thomson’s I2 index. Analyses will be performed using the meta package from R Software V.4.0.0 (R Core Team (2020). R Foundation for Statistical Computing, Vienna, Austria).
Prespecified subgroup analyses
We will consider clinical heterogeneity and if there is sufficient power, we will conduct subgroup analyses according to type of hypertensive disorder in pregnancy (preeclampsia de novo vs superimposed preeclampsia, preterm preeclampsia <37 and <34 weeks), trimester in pregnancy at urinary collection and method of PlGF measurement, as well as maternal medical conditions. Variables of interest were chosen for subgroup analyses considering their clinical relevance in leading to a potential gain of information. Heterogeneity will be quantify using I2 index and a χ2 test will be performed to assess for homogeneity between subgroups with statistical significance at 5%.
Sensitivity analyses
Sensitivity analyses will be conducted by study epoch and study quality if numbers are sufficient.
Assessment of reporting bias
Considering that studies with inconclusive or non-significant results often remain unpublished, we will assess reporting bias using a funnel plot if at least 10 studies are included in the meta-analysis.18 In the absence of publication bias, all studies will lie symmetrically around the calculated pooled odd ratios. Since the interpretation of the funnel plot is subjective, arcsine test for dichotomous outcomes measured as ORs will be presented to quantifies asymmetry considering a statistical significance of 5%.19
Interpretation of results
All authors will contribute to the interpretation of the results with their respective input of expertise for methodology and statistical analyses (M-CB, SL-P) and for content and context (CF, JM, A-MC, M-ER-L) and a virtual meeting will be planned to discuss and debate the results and reach consensus for interpretation and presentation of results. We will assess the quality of evidence using the GRADE framework.17
Protocol amendments
Any amendments to the protocol will be reported, justified and dated.
Ethics and dissemination
Review by an ethics committee will not be required since no data from participants will be collected. The results of this systematic review and meta-analysis will follow the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines20 and will be submitted for publication in a peer-reviewed journal. The results will also be submitted for presentation at conferences targeting different stakeholders, including researchers, physicians and patients.
Discussion
Preeclampsia remains an important cause of maternal, fetal and neonatal mortality and morbidity. Testing for angiogenic factors can improve identification and management of women with hypertensive disorders in pregnancy. Urinary testing is easier to obtain than blood samples and as a non-invasive test, it is more acceptable to most women during antenatal care. Thus, urinary PlGF represents a great alternative in the evaluation of pregnant women and to facilitate longitudinal follow-up. Moreover, serial urinary samples may allow for better interpretation of biomarker and clinical trajectory of women at risk or developing an hypertensive disorder in pregnancy. Finally, in low-resource settings and remote areas, as well as for telemonitoring, urinary testing for PlGF may eventually be more accessible to guide management and medical decisions at lower costs.
Strengths and limitations
This prospectively registered study provides a search strategy, well-defined outcomes and analysis plan. The systematic review and meta-analysis will be rigorously conducted with the contributions of all authors sharing expertise in methodology and context. The GRADE framework will be used to report our results. Considering the interest for urinary PlGF is recent, the small numbers of studies and the heterogeneity of the data may be a limitation of this study. However, this systematic review may identify future areas of study on the use of urinary PlGF in pregnancy.
Acknowledgments
We thank Dr Annabelle Cumyn, Dr Annie Ouellet, Dr Evelyne Rey and Dr Evelyne Raîche, expert panel in maternal and fetal medicine, Marie-Noëlle Richard, patient with lived experience of preeclampsia, Josée Toulouse, librarian and Anne-Marie Lafaille-Hebert, medical student, for their contribution to the protocol. Dr Cumuy also kindly revised the manuscript for linguistic review.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Contributors All authors contributed to the preparation of the protocol, reviewed and approved the final version of the protocol. A-MC and CF conceived the research question. A-MC, CF, M-CB, JM and SL-P contributed to the methodological aspects and analysis sections of the protocol. M-ER-L contributed to expert content and outcomes definitions. A-MC is the guarantor of this review.
Funding CF is a recipient of a Postgraduate Scholarship from the Faculty of Medicine and Health Sciences of Université de Sherbrooke. A-MC’s research program is supported by funding from the Canadian Institutes of Health Research Research (MFM-46445).
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.