Is platelet to lymphocyte ratio predictive of preeclampsia? A systematic review and meta-analysis

Abstract Background To evaluate the value of the platelet-to-lymphocyte ratio (PLR) in predicting preeclampsia (PE) in pregnant women. Methods PubMed, EMBASE and Web of Science databases were searched for observational studies (cohort, case-control or cross-sectional) that reported pre-treatment maternal PLR values in women with and without PE. The analysis was done using a random effects model. Pooled effect sizes were reported as weighted mean difference (WMD) with 95% confidence intervals (CIs). Newcastle-Ottawa Scale (NOS) was used to evaluate the risk of bias. Results Twenty-five studies with 7755 patients were included in this meta-analysis. PLR was comparable in patients with PE and healthy pregnant women (WMD −2.97; 95% CI: −11.95 to 6.02; N = 16). Patients with mild (WMD −3.00; 95% CI: −17.40 to 11.41; N = 12) and severe PE (WMD −5.77; 95% CI: −25.48 to 13.94; N = 14) had statistically similar PLR, compared to healthy controls. Conclusions Our findings show similar PLR in PE and healthy pregnancies. PLR, therefore, may not be used to differentiate between PE and normal pregnancy or for assessing the severity of PE. The majority of included studies were case-control, potentially introducing bias, and we identified evidence of publication bias as well. Plain Language Summary This study assessed the use of the platelet-to-lymphocyte ratio (PLR) for the prediction of preeclampsia (PE) in pregnant women. After reviewing 25 studies involving 7755 pregnant women, the researchers found that PLR values were similar in women with PE and those with healthy pregnancies. Additionally, PLR did not differ significantly in cases of mild or severe PE, compared to healthy pregnancies. This study concluded that the PLR may not a reliable predictor for PE in pregnant women, nor can it be used to assess the severity of PE. Further research is needed to identify more effective methods for predicting and managing PE in pregnant women.


Introduction
Preeclampsia (PE) is a condition that occurs exclusively during pregnancy and is characterised by the development of hypertension and proteinuria (Duley 2009, Ives et al. 2020).PE occurs in approximately 2-8% of all pregnancies and is considered a significant obstetric complication due to the associated maternal and neonatal complications (Duley 2009, Madazli et al. 2014, Wadhwani et al. 2020).Timely identification and accurate prediction of PE are crucial for appropriate management and improved outcomes for affected women and offspring.
Current research focuses on identifying new reliable biomarkers that can aid in the early prediction and assessment of PE (Petla et al. 2013, Danielli et al. 2022, MacDonald et al. 2022).
The platelet-to-lymphocyte ratio (PLR) has emerged as a potential predictor of various autoimmune and cardiovascular diseases (Gasparyan et al. 2019, Kurtul and Ornek 2019, Wang et al. 2023).Numerous studies attempted to examine the potential of PLR as a biomarker and evaluate its effectiveness in predicting the severity of the various diseases and associated complications (Zhou et al. 2019, Liu et al. 2022, Simadibrata et al. 2022).While some studies suggested that PLR may be considered a potential marker for assessing the presence of inflammation and the immune response in cases of PE, the research findings are inconclusive (Gezer et al. 2016, Sitotaw et al. 2018, Mannaerts et al. 2019, Elmaradny et al. 2022, Yakis ¸tiran et al. 2022, Ozkan et al. 2023).Some studies have indicated a notable increase in PLR in patients with pre-eclampsia compared to women with uncomplicated pregnancies (Gezer et al. 2016, Yakis ¸tiran et al. 2022).
This elevated PLR could potentially indicate an inflammatory response and an imbalance in the immune system, associated with pre-eclampsia.However, other studies report lack of significant differences or even a decrease in PLR levels in women with pre-eclampsia compared to healthy pregnancies (Sitotaw et al. 2018, Mannaerts et al. 2019, Elmaradny et al. 2022, Ozkan et al. 2023).It is important to note that additional factors like gestational age, maternal age and underlying health conditions might influence PLR values, and their impact should be considered when interpreting the results of these studies.
The main aim of this analysis was to summarise current evidence and to investigate whether there is a significant difference in PLR values between women with PE and those with normal pregnancies.Additionally, we aimed to assess whether PLR values show correlation with the severity of PE.Our results may provide an insight on whether PLR may be integrated into a routine clinical assessment for a comprehensive evaluation of pre-eclampsia.This may allow further aid in risk stratification, management decisions and potential interventions.

Search strategy
PubMed, Embase and Web of Science databases were comprehensively searched for the relevant English language studies, published up to 15 May 2023, using combinations of keywords: ('Platelet-to-Lymphocyte Ratio' OR 'PLR' OR 'blood parameters' OR 'blood count' OR 'platelets' OR 'lymphocytes') AND ('preeclampsia' OR 'eclampsia' OR 'gestational hypertension').Manual searches of reference lists and relevant review articles were also conducted for any missed eligible studies.

Study selection criteria
We included: (1) studies that investigated the link between pre-treatment maternal PLR and PE; (2) studies that included PE patients and healthy pregnant women as a control group; (3) English-language observational studies (cohort, case-control or cross-sectional) from peer-reviewed journals.
Reports not meeting the inclusion criteria, studies with incomplete or insufficient data to calculate effect sizes (mean differences), conference abstracts, case reports, editorials, reviews and animal studies were not considered for inclusion.

Selection of studies
Following the execution of the search strategy in the three designated databases, duplicate studies were removed.Two reviewers then independently searched titles and abstracts for relevant publications for further evaluation.In the subsequent stage, the full text of the chosen studies was carefully assessed to determine whether they met the criteria for inclusion in our meta-analysis.This comprehensive assessment involved a detailed examination of the methodology and the results sections of each study.Any discrepancies or disagreements regarding the inclusion of studies were resolved through in-depth discussions among the study authors.

Risk of bias assessment and data extraction
We used Newcastle-Ottawa Scale (NOS) (Well et al. 2023) to evaluate the risk of bias.Data extraction from the final set of included studies was carried out independently by two authors using a standardised data extraction form.Any discrepancies during this process were resolved by discussion.A third senior author was consulted to provide additional insight and expertise when needed to facilitate consensus among the authors.In studies where the effect sizes were reported as median with range (minimum-maximum), we adopted the method by Hozo et al. to derive mean and standard deviation (Hozo et al. 2005).

Statistical analysis
We decided to employ a random-effects model for all analyses.This choice was motivated by the observed variations in study characteristics, including baseline characteristics of the patients, study settings, sample size and severity of PE considered.These variations had the potential to introduce substantial heterogeneity in the reported findings.By using a random-effects model, we aimed to address this expected heterogeneity.Pooled effect sizes were reported as weighted mean difference (WMD), with corresponding 95% confidence interval (CI).The I-squared statistic was utilised to assess the presence of statistical heterogeneity.Higher values of I2 (>40%) indicated greater heterogeneity in the included studies (Cochrane Handbook for Systematic Reviews of Interventions 2023).An evaluation of publication bias was conducted by Egger's test and funnel plots (Egger et al. 1997).A p value less than .05indicated statistical significance.We conducted subgroup analysis based on the country of the study, total sample size, sample size in the PE group and NOS quality score obtained by the included studies.

Adherence to relevant guidelines
This study was conducted according to the PRISMA guidelines (Page et al. 2021, PRISMA 2023), and the protocol of the study was registered in the International Prospective Register of Systematic Reviews (PROSPERO).The registration number assigned to this study protocol in PROSPERO is CRD42023429260.

Results
Our search strategy identified 2948 studies (Figure 1).After removal of duplicates, titles and abstracts of the remaining 2057 unique studies were evaluated.Additional 2006 reports did not meet eligibility criteria and were excluded.Of the remaining 51 studies, 26 did not meet the necessary criteria and were excluded.Finally, 25 studies with 7755 patients were selected for our analysis (Figure 1) (Yavuzcan et al. 2014, Kirbas et al. 2015, Gezer et al. 2016, Toptas et al. 2016, Jeon et al. 2017, Y€ ucel and Ustun 2017, C¸intesun et al. 2018, Kim et al. 2018, Sitotaw et al. 2018, Gogoi et al. 2019, Kholief et al. 2019, Mannaerts et al. 2019 et al. 2023et al. , Ozkan et al. 2023)).Table 1 summarises the specific details of the included studies.
The quality assessment scores for the studies varied between 6 and 8 on the NOS, which has a maximum possible score of 9. On average, the included studies received a mean NOS score of 7.1.Most of the studies (n ¼ 13) were conducted in Turkey.Two studies were carried out in Indonesia, the Republic of Korea and Egypt.One study was conducted in China, Georgia, Saudi Arabia, Belgium, India and Ethiopia.Regarding the study designs, most of the studies (n ¼ 11) had a case-control design.Five studies were cross-sectional, while eight studies were retrospective and used data from registries or medical records.Only one study adopted a prospective cohort design.

PLR in patients with preeclampsia and healthy controls
PLR was comparable in patients with PE and healthy pregnant women (WMD −2.97; 95% CI: −11.95 to 6.02; I 2 ¼ 86.3%; N ¼ 16) (Figure 2).Egger's test and funnel plots did not show the presence of publication bias (p > .05)(Supplementary Figure 1).Subgroup analysis also did not show any significant distinction in the PLR between women with PE and healthy pregnancies across the different subgroups (Table 2).

Discussion
The results of this study show no significant difference in the PLR between women with PE and healthy pregnant women.Similar findings were reported for the analysis based on the severity of PE.This implies that PLR may not serve as a reliable biomarker for distinguishing between PE and normal pregnancies.One possible explanation for the lack of significant difference in PLR between PE and healthy pregnant controls could be related to the complex pathophysiology of PE.
It is widely accepted that PE is a multifactorial disorder involving various mechanisms, including inflammation, endothelial dysfunction, oxidative stress and immune dysregulation (Uzan et al. 2011, Gathiram andMoodley 2016).PLR, as a marker of systemic inflammation, may be influenced by several factors other than PE, such as maternal age, gestational age and other comorbidities (Akgun et al. 2017, Hershko Klement et al. 2018).These confounding factors might mask the potential association between PLR and PE.
Our analysis revealed comparable PLR in women with mild PE and women with healthy pregnancies.Similarly, comparable PLR was observed in women with severe PE and healthy pregnant women.This suggests that PLR may not be a useful marker for distinguishing between different stages of PE or assessing disease severity.Alternative biomarkers, including placental growth factor (PlGF), soluble fms-like tyrosine kinase-1 (sFlt-1) and neutrophil-lymphocyte ratio (NLR) have demonstrated stronger associations with the severity of PE and may offer more accurate prognostic information (Chau et al. 2017, Kang et al. 2020, Leaños-Miranda et al. 2020).The heterogeneity among the included studies should also be considered when interpreting the results of this meta-analysis.Variations in study design, patient populations, sample sizes and laboratory techniques may have     While our study did not show a significant finding, it does indicate the need for sequential measurement of PLR throughout pregnancy to identify the most appropriate gestational age for sampling and to capture potential dynamic changes in PLR that might be linked with the development of PE.Furthermore, PLR should be evaluated in conjunction with conventional PE markers, such as blood pressure measurements, proteinuria and inflammatory markers.Combining PLR with established markers in predictive models may enhance the overall efficiency and accuracy of PE prediction.Future research should focus on constructing robust prediction models, conducting large prospective studies, and validating the findings in diverse populations to fully explore the potential of PLR as a predictive marker for PE.

Limitations
It is important to acknowledge some limitations of this metaanalysis.First, although we conducted a thorough quality assessment of the included studies, potential biases and limitations inherent in the original studies cannot be completely ruled out.These factors may have influenced the overall effect estimate.Second, the majority of reports in our analysis were case-control studies that are susceptible to selection bias, particularly with respect to the choice of controls.Finally, we found statistical evidence of publication bias.

Conclusions
Our findings suggest a lack of significant difference in PLR between PE and healthy pregnancies.PLR, therefore, is not a useful biomarker for distinguishing between PE and normal pregnancies or assessing disease severity.Further research is needed to identify more reliable and specific biomarkers for the early detection and management of PE.Large-scale prospective cohort studies are warranted to accurately assess the role of PLR in PE.These studies would provide more robust evidence of the association between PLR and PE, overcome limitations of small sample sizes, account for potential confounding variables, and offer longitudinal data to evaluate PLR dynamics throughout pregnancy.Such studies may also have the potential to enhance our understanding of PLR as a biomarker for PE and its clinical utility in risk prediction and management strategies.

Figure 1 .
Figure 1.Selection process of studies included in the review.

Figure 2 .
Figure 2. Weighted mean difference (WMD) of PLR between cases with preeclampsia and healthy controls.

Figure 3 .
Figure 3. Weighted mean difference (WMD) of PLR between cases with mild preeclampsia and healthy controls.

Figure 4 .
Figure 4. Weighted mean difference (WMD) of PLR between cases with severe preeclampsia and healthy controls.

Table 1 .
Studies included in the meta-analysis and their characteristics.

Table 2 .
Findings of the subgroup analysis between preeclampsia and the healthy controls.
contributed to the observed heterogeneity.Different studies may have employed varying techniques for measuring PLR, leading to inconsistent results.Variations in laboratory protocols, such as blood collection methods, platelet and lymphocyte counting techniques, and timing of sample collection, can introduce variability and affect the accuracy and comparability of PLR measurements.Standardising measurement techniques across studies is, therefore, crucial to reduce variability.