Association of platelet-lymphocyte ratio with acute pancreatitis severity: systematic review and meta-analysis Authors: Mohamed Y Ibrahim, Mona M Alhaj, Abdelrahim Abdelrazig Ahmed Ibrahim, Malak Rabih, Maram Rabih Musa Rabih, Mohammed Elrayah Hassan Massaad, Abdelrahman K Elzubir, Mohammed Mansour Hassan Abdelrahman, Omer Afifeldin Elmusharf Abdelrahman, Sagad O O Mohamed Published: 2025 ABSTRACT AbstractBackgroundAcute pancreatitis (AP) is an inflammatory disease with a heterogeneous clinical course, ranging from mild symptoms to life-threatening complications. The platelet-to-lymphocyte ratio (PLR), an easily accessible inflammatory biomarker, has shown prognostic value in other conditions but has been understudied in AP. This review aimed to systematically evaluate the association between PLR and AP severity.MethodsA PRISMA-guided systematic review and meta-analysis was conducted, searching PubMed, Web of Science, Scopus, and WHO-VHL databases (earliest records to March 2025). Observational studies reporting PLR and AP severity were included. Pooled standardized mean differences (SMDs) were calculated using a random-effects model.ResultsAmong 20 included studies encompassing 5,920 patients, PLR was significantly higher in severe acute pancreatitis (SAP) versus non-SAP (SMD = 1.137, 95% CI: 0.752–1.702, P < .001). Subgroup analysis by SAP definition revealed significant differences (p = .011). In studies using the Revised Atlanta classification, PLR was markedly higher in SAP patients (SMD = 1.479, 95% CI: 0.729 to 2.231). The Egger’s regression test and funnel plot indicated no significant publication bias (p = .105) and the sensitivity analysis confirmed robustness of the findings.ConclusionPLR may serve as a low-cost biomarker for AP severity stratification, with higher levels indicating severe disease and poor outcomes. Its integration into clinical practice could enhance early risk assessment, particularly in resource-limited settings. Prospective studies are needed to validate optimal cutoff values and refine its prognostic utility.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12876-025-04556-w.Keywords: Platelet-lymphocyte ratio, Acute pancreatitis, Severity, Inflammation, Meta-analysis Introduction Acute pancreatitis (AP) is an inflammatory disease with multiple etiologies, most commonly gallstones, alcohol, and hypertriglyceridemia, followed by medication, autoimmune disease, structural and hereditary causes, as well as iatrogenic causes [1]. This damage to pancreatic parenchyma triggers a cascade of inflammation, which results in a heterogeneous clinical course. The presentation of AP ranges from mild and self-limited symptoms to life-threatening systemic complications, multiple organ failure, and death [1]. The global incidence of AP has been reported to have increased from 1990 to 2019, with the overall burden remaining high, as well as increasing mortality and disability-adjusted life years [2]. Given this heterogeneity and complexity, accurate and timely assessments of severity is crucial to initiate appropriate interventions that can lead to improved patient outcomes [3, 4]. Currently, multiple scoring systems are used to diagnose and classify the severity, such as the Revised Atlanta Classification, the Ranson score, APACHE II, BISAP, and Glasgow-Imrie criteria [5]. The scoring systems are based on multiple clinical and laboratory variables, making them complex and time-consuming to calculate [5]. On the other hand, platelet-to-lymphocyte ratio (PLR) is an affordable and easily obtained new inflammatory biomarker that has prognostic value in different inflammatory and critical conditions [6, 7]. It has been linked to multiple systemic inflammatory processes and rheumatological diseases [8, 9]. Its level has also been linked to heart failure, in-hospital and short-term mortality, as well as tumor growth, angiogenesis, and overall survival in advanced cancers [8, 10, 11]. Though several biomarkers have been extensively studied in AP [12], PLR’s role in predicting severity and death in AP has not been studied extensively. The aim of the current review is to systematically assess the association between PLR and AP severity, thereby providing supportive evidence for its potential clinical use. Methods Search approach and studies inclusion criteria This systematic review was conducted following the established guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement (Additional file 1) [13]. The protocol for this systematic review was registered on Prospero (CRD420251027202). To identify all relevant published research, we undertook an inclusive electronic literature search across PubMed, Web of Science, Scopus, and World Health Organization Virtual Health Library Regional Portal (WHO-VHL). The search strategy did not impose any limitations based on geographical location or publication date, aiming to capture the widest possible range of relevant studies. The search involved the published studies from the earliest records to March 2025, using the following search keywords: (“Pancreatitis”) AND (“Platelet-to-Lymphocyte” OR “platelet-lymphocyte” OR “platelet/lymphocyte” OR “PLR”) (Additional file 2). The references in the included articles were screened to make sure that no relevant studies have been missed. In addition, we conducted a search in Google Scholar along with our searches in major databases to identify potentially relevant grey literature and preprints. All publications were uploaded to Endnote software for initial screening of titles and abstracts as well as to remove duplicates. Inclusion and exclusion criteria Following the literature search, the selection procedure involved a two-step approach. Initially, four independent reviewers screened titles and abstracts of all identified articles to identify potentially relevant studies. Studies judged to be relevant subsequently underwent full-text evaluation to determine their final eligibility based on the pre-defined inclusion criteria. We included observational studies (cross-sectional, case-control, or cohort studies) that presented results on the association of platelet-lymphocyte ratio with acute pancreatitis severity. In addition, we excluded case reports, editorials, review articles, conference abstracts, and studies published in languages other than English. Quality assessment and data extraction The quality of included studies was assessed for potential bias using Joanna Briggs Institute critical appraisal checklists (https://jbi.global/critical-appraisal-tools). The tool aided in assessing the risk of bias related to study design, implementation, and data analysis. For each included study, the following data were extracted: author(s), publication year, geographic region, number of patients, age range of participants, mean age, outcome measures, and summary of the clinical outcome results related to AP analyses. Any discrepancies encountered throughout the quality assessment and data extraction were effectively resolved through discussion. Statistical analysis The analyses were performed utilizing meta package of R to calculate the pooled standardized mean difference (SMD) estimates and their associated 95% confidence intervals (CIs). We conducted random-effects meta-analyses, using the Restricted Maximum Likelihood (REML), to account for between-study variability. To obtain more reliable confidence intervals, especially given the small number of studies per comparison and the observed heterogeneity, the Hartung–Knapp–Sidik–Jonkman adjustment was implemented to account for heterogeneity among the reviewed publications. Heterogeneity among studies was evaluated by utilizing the I² statistic, which quantifies the variation proportion in effect estimates because of the true heterogeneity rather than chance. Publication bias, a potential source of bias because of the overrepresentation of publications with significant results, was assessed utilizing Egger’s tests and inspection of funnel plots [14–17]. The Duval and Tweedie trim-and-fill analysis was implemented to address publication bias caused by missing articles [15]. Results Studies characteristics The schematic flow of the study identification and selection process is presented in Fig. 1. The initial electronic database search yielded a total of 256 records. Following the removal of duplicate entries, 165 studies remained. These studies underwent a thorough title and abstract screening process. During this stage, 135 studies were excluded due to irrelevance. Full texts of the remaining records were screened. Ultimately, a total of 20 studies were selected for the systematic review [3, 5–7, 11, 12, 17–32] (Fig. 1). Study publications ranged from 2010 to 2024 and included patients, with a total of 5,920 patients with acute pancreatitis. All of the included studies assessed adult population. The main features of the selected studies, as well as risk of bias assessment, are presented in Table 1 and (Additional file 3). Flow chart for studies selection process Table 1Baseline characteristics of the studies and summary of the and findingsStudyYearCountrySample sizePatients withsevere APMale%Age (mean)Quality assessmentMeasurement timing (within n hours of admission)Main findingsAk et al.2022Turkey5142247.1557/848 hHigher PLR values in SAP group, based on the revised Atlanta classification, compared to othersAkdur et al.2022Turkey1713939.868.028/8n/aHigher PLR values in SAP group those with higher Modified Glasgow Prognostic scores compared to others. PLR demonstrated a predictive value lower to that of NLR and CRP. The discriminative performance based on ROC curve analysis was (AUC = 0.667) for diagnosing SAP (cut-off > 168.05, Sensitivity = 75%, and Specificity = 42.38%).Cho et al.2018South Korea2432568.359.38/81 hHigher PLR values in SAP group, based on the revised Atlanta classification, compared to others. PLR demonstrated a predictive value significantly superior to that of CRP. The discriminative performance based on ROC curve analysis was (AUC = 0.638) for predicting presence of persistent organ failure in patients with gallstone acute pancreatitis (cut-off > 229, Sensitivity = 70.4%, and Specificity = 50.3%).Cifci et al.2020Turkey55713143.458.668/8At admissionHigher PLR values in those with higher Ranson scores and one-year mortality compared to others. Higher PLR was significantly associated with ICU stay. The discriminative performance based on ROC curve analysis was (AUC = 0.628) for predicting one-year mortality (cut-off > 215, sensitivity = 56%, and specificity = 75%).Duru et al.2023Turkey1043759.666.56/8At admissionNo significant association between PLR and CT Severity Index and Ranson scoreJunare et al.2021India1602657.537.338/824 hHigher PLR values in SAP group, based on the revised Atlanta classification, compared to others. PLR demonstrated a predictive value higher than APACHII score and lower to that of NLR The ROC curve analyses showed good discriminative performances for predicting ICU admission (AUC = 0.883), mortality (AUC = 0.867), organ failure (AUC = 0.874), and need for intervention (AUC = 0.862) (cut-off > 208.71).Kaplan et al.2018Turkey1421340.861.66/8n/aPLR was significantly correlated with mortality, Ranson score, and BISAP scoreKarabuga et al.2022Turkey50011249.455.688/8At admissionHigher PLR values in those with higher BISAP score compared to others. PLR demonstrated a predictive value lower to that of NLR. The discriminative performance based on ROC curve analysis was (AUC = 0.682) for diagnosing SAP (cut-off > 263.5, sensitivity = 55.36%, and specificity = 38.5%).Khan et al.2021Pakistan154676148.478/824 hHigher PLR values in those with higher CT Severity Index score compared to others.Kobler et al.2018Poland95868486/824 hHigher PLR values in those with higher MMSS compared to others.Li et al.2024China253606445.988/8At admissionHigher PLR values in SAP group, based on the revised Atlanta classification, compared to othersLiu X. et al.2018China1012163548/872 hHigher PLR values in SAP group, based on the revised Atlanta classification, compared to others. The ROC curve analysis showed a moderate discriminative performance (AUC = 0.877, cut-off = 195.08) for diagnosing SAP. Higher PLR was significantly associated with presence of persistent organ failureLiu G. et al.2021China2791862.452.978/8At admissionHigher PLR values in SAP group, based on the revised Atlanta classification, compared to others participants (combined MSAP and SAP group). PLR demonstrated a predictive value higher to that of NLR and BISAP. The ROC curve analyses showed a good discriminative performance (AUC = 0.877, cut-off = 173.13) for predicting persistent organ failurePetrescu et al.2020Romania33758n/an/a8/8n/aHigher PLR values in all patients with complications compared to others (cut-off > 123.82, sensitivity = 72.4%, and specificity = 47%).Tahir et al.2024Pakistan22519261.7848.326/8n/aHigh diagnostic precision of PLR in calculating the severity of acute biliary pancreatitis based on the CT Severity Index.Vo et al.2023Vietnam1312167.9467/8At admissionHigher PLR values in SAP group, based on the revised Atlanta classification, compared to others. The ROC curve analysis showed a moderate discriminative performance (AUC = 0.72) for diagnosing SAP (cut-off > 202.7, sensitivity = 66.7%, and specificity = 71.8%).Wang et al.2019US1101074.639.57/848 hHigher PLR values in SAP group, based on presence of persistent organ failure, compared to others.Xu et al.2024China77117860.449.677/824 hHigher PLR values in SAP group, based on the revised Atlanta classification, compared to others. The ROC curve analysis showed a weak discriminative performance (AUC = 0.569) for predicting presence of persistent organ failure.Yalcin et al.2017Turkey6677132.977.887/848 hHigher PLR values in non-survivors compared to survivors groupZhou et al.2018China4065660577/848 hHigher PLR values in SAP group, based on the revised Atlanta classification, compared to others. The ROC curve analysis showed a moderate discriminative performance (AUC = 0.693) for predicting 28-day mortality (cut-off > 169.285, sensitivity = 64.3%, and specificity = 61.7%) Baseline characteristics of the studies and summary of the and findings PLR in patients with severe acute pancreatitis vs. non-severe acute pancreatitis We conducted meta-analyses to assess the difference in the mean levels of the PLR between patients with mild acute pancreatitis (MAP), moderately severe acute pancreatitis (MSAP), severe acute pancreatitis (SAP), and combined groups of MAP and MSAP (non-SAP). The meta-analyses results indicated that the PLR levels were significantly higher in SAP patients compared to the other patients’ groups (MAP, MSAP, and non-SAP). The meta-analysis revealed that PLR levels were significantly higher in SAP group compared to non-SAP group (SMD = 1.138, 95% CI: 0.573 to 1.703, p <.001, I² = 96.0%) (Fig. 2). The funnel plot and Egger’s regression test indicated no significant publication bias (p =.105) (Additional file 4). Forest plot for the pooled SMD of PLR between patients with SAP vs. non-SAP Comparison of the PLR between SAP, MAP, and MSAP groups showed that PLR levels were higher in SAP group than in MAP group (SMD = 2.22, 95% CI: 1.088–3.354, p <.001, I² = 98.2%) and MSAP group (SMD = 1.280, 95% CI: 0.089–2.471, p =.035, I² = 96.2%) (Figs. 3 and 4) Evidence of publication bias was detected for the SAP vs. MAP analysis based on Egger’s regression test (p =.032) and funnel plot inspection, whereas no significant bias was observed for the SAP vs. MSAP comparison (Additional file 4). Forest plot for the pooled SMD of PLR between patients with SAP vs. MAP Forest plot for the pooled SMD of PLR between patients with SAP vs. MSAP In addition, three studies by Cifci et al., Yalcin et al., and Zhou et al. assessed the difference in the mean levels of the PLR between non-survivors and survivors of acute pancreatitis. Of them, two studies reported in-hospital mortality, while Cifci et al. investigated one-year mortality. These studies were consistently showing that PLR levels were significantly higher in the non-survivors groups than in survivors groups. Lastly, four studies stratified patients with acute pancreatitis based on varied sets of clinical (Table 1). These different scoring protocols included assessing early organ failure via the Modified Marshall system score (MMSS), Computed Tomography (CT) Severity Index, C-reactive protein (CRP), Ranson score, and BISAP score. However, all of the studies (except Duru et al.) were consistently showing that PLR levels were significantly higher in patients with higher risk or complications compared to the other groups (Table 1). Meta-regression and sensitivity analyses Meta-regression analyses were conducted to analyze the effect of several variables on the heterogeneity in the meta-analysis of SAP vs. non-SAP comparison. The results showed that none of these variables had a significant moderating effect on the outcome: measurement time (coefficient = 0.394, P =.568), quality assessment score (coefficient = 0.212, P =.611), number of patients (coefficient = −0.001, P =, 0.264), mean age (coefficient = −0.052, P =.115), male ratio (coefficient = 0.005, P =.869), and publication year (coefficient = 0.15, P =.298). Subgroup analysis by SAP definition revealed significant differences (p =.011). In studies using revised Atlanta classification, PLR was markedly higher in SAP patients (SMD = 1.479, 95% CI: 0.729 to 2.231). In contrast, the other studies used other methods showed a smaller but significant elevation (SMD = 0.421, 95% CI: 0.100 to 0.742). Conversely, subgroup analysis by country indicated no significant differences (p =.170). The effect size was higher in studies from Asian countries (SMD = 1.421, 95% CI: 0.588 to 2.254) compared to other studies (SMD = 0.707, 95% CI: 0.118 to 1.297). Furthermore, a leave-one-out sensitivity analysis was conducted to assess the robustness of the meta-analysis results for PLR in SAP versus non-SAP. The results showed that all re-analyses, where each study was removed one at a time, yielded a statistically significant pooled effect estimate, indicating that the effect size from each individual study was always consistent and did not cause excessive change to the pooled estimates, with SMD ranging from 0.983 to 1.240 (Additional file 4). Meta-regression and sensitivity analyses were not performed for the other comparisons due to the limited number of studies. Association of PLR with clinical outcomes Some investigations have evaluated the discriminative performance of PLR, which was assessed via receiver operating curve (ROC) curve analysis. The diagnostic performance of PLR as a biomarker for AP severity and outcomes showed variability across studies. area under the curve (AUC) values for PLR in predicting severity and various outcomes show a wide range, from 0.521 to 0.877. Sensitivity and specificity also vary significantly across studies. PLR often outperformed or was comparable to markers like NLR, CRP, BISAP, or APACHE II for outcomes like SAP diagnosis, organ failure, and mortality (Table 1). Studies reported that PLR correlated well with established scoring systems like BISAP score, Ranson score, CT Severity Index, and MMSS. the presence of persistent organ failure, ICU Studies reported that PLR was associated with persistent organ failure (POF), increased mortality, and the occurrence of complications (Table 1). PLR showed moderate to good performance for diagnosing SAP, with AUCs indicating fair (0.667–0.72) to good (0.877) discriminative ability (Table 1). Sensitivity ranged from 55 to 75%, and specificity from 38 to 72%, with varying cut-off thresholds (168.05–263.5). PLR’s performance was often comparable to or better than neutrophil-to-lymphocyte ratio (NLR) or C-reactive protein (CRP). Likewise, PLR’s performance for predicting persistent organ failure varied widely. AUCs ranged from weak (0.569) to good (0.877), with cut-offs between 173.13 and 229. PLR outperformed NLR and BISAP in some studies but showed inconsistent reliability. Junare et al. PLR demonstrated good discriminative performance (AUC = 0.883) for predicting ICU admission, the need for intervention. Some studies showed that combining PLR with other markers like NLR revealed higher AUC for predicting AP prognosis and superior diagnostic discrimination [19, 30]. Discussion This systematic review and meta-analysis provided robust evidence that the PLR is significantly elevated in patients with SAP compared to those with milder forms of AP. Elevated PLR was also associated with adverse clinical outcomes, including increased mortality, ICU admission, persistent organ failure, and higher severity scores. These findings are consistent with the pathophysiological understanding that platelets are key mediators of inflammation and thrombosis, while lymphocytes are central to immune regulation. Platelets stimulate the cascade of inflammation through the secretion of cytokines, endothelial-cell interaction, and connection with leukocytes, thus leading to tissue damage and thrombosis. Lymphocytes, on the other hand, are an indicator of immune competence; a reduced lymphocyte level is an indicator of immune suppression or dysregulation. Thus, a high platelet-lymphocyte rate may indicate high-grade inflammatory processes and immune repression, which are precisely those circumstances that promote clinical degradation and manifestation of complications of acute pancreatitis of the severe type. An increased PLR may thus reflect an amplified inflammatory response and immune dysregulation, both of which are hallmarks of severe AP and its complications [8–11]. Beyond acute pancreatitis, PLR has demonstrated prognostic value in various other inflammatory and critical conditions. In cardiovascular diseases, elevated PLR correlates with adverse events such as myocardial infarction and heart failure, reflecting systemic inflammation and endothelial dysfunction. Similarly, in malignancies like breast, lung, and colorectal cancer, PLR has been shown to predict tumor aggressiveness and patient survival, linking inflammation to cancer progression. Furthermore, PLR has been associated with disease activity in autoimmune disorders and infectious diseases, highlighting its potential as a broadly useful, inexpensive biomarker of systemic inflammation that can guide clinical decisions across a range of pathologies [8–11]. This is in line with prior research linking PLR to poor outcomes in other inflammatory and critical conditions, such as cardiovascular diseases, rheumatological disorders, and malignancies [8–11]. PLR is an affordable, easily accessible biomarker derived from routine blood counts, reflecting the systemic inflammatory response that is central to the pathogenesis and progression of acute pancreatitis. Unlike complex severity scores such as Ranson, APACHE II, and BISAP, which require multiple clinical and laboratory variables and can be time-consuming, PLR offers a rapid and cost-effective assessment of inflammation and immune status. Therefore, PLR can be used a supplementary biomarker alongside established severity scores to yield timely diagnostic information, an advantage especially useful in low-resource environments. Combining PLR with other clinical and laboratory parameters could enhance risk prediction models for AP severity and outcomes. Moreover, investigating dynamic changes in PLR during the disease course may provide insights into its utility for monitoring progression and response to therapy. Elevated PLR mirrors the imbalance between the pro-inflammatory and anti-inflammatory responses, which is a hallmark of severe disease and complications in AP [8–11]. Its integration into clinical practice could facilitate early identification of high-risk patients and guide timely interventions, especially in resource-limited settings. Our findings align with previous literature evaluating other inflammatory markers in AP, such as CRP and NLR, but PLR has been less extensively studied in this context [6–8]. The current review addresses this gap by synthesizing data from diverse studies and populations, reinforcing the relevance of PLR as a marker of systemic inflammation and prognosis in AP. Despite these strengths, several limitations should be considered. Changes in the platelets and lymphocytes count may occur because of factors unrelated to the basic pathological process, especially infections, drugs, hematological diseases, and stress. This is due to the fact that the cut-off values of the PLR were inconsistent, thus reducing its diagnostic sensitivity and intergenerational validity. Most of the incorporated studies were observational, hence exposing them to the risks of bias and confounding that interfere with causal inference. The studies were also inclined to be single-centered, hence limiting their applicability to heterogeneous patients and hospital settings. In turn, the use of other clinical and laboratory parameters as they may modulate the meaning of PLR and be relied upon as independent markers. In some comparisons, pooled SMD values exceeded 2, which is unusually large for SMDs. Such extreme values may challenge the assumption of approximate normality underlying SMDs and could reflect skewed distributions, outliers, or variations in PLR measurement across studies. The included studies exhibited heterogeneity in terms of patient populations, study design, and severity classification methods. Although meta-regression did not identify significant moderators, unmeasured confounding cannot be excluded. Most studies were observational, limiting causal inference. Future research needs to focus on prospective, multicentric validation, with the aim to determine optimal cut-off levels of PLR as well as on ascertaining whether dynamic monitoring of this parameter throughout the course of disease will provide further improvement of its prognostic ability. Finally, excluding non-English studies could limit the breadth of conclusions. Conclusion In summary, this systematic review and meta-analysis demonstrate that elevated PLR is significantly associated with increased severity and adverse outcomes in acute pancreatitis. Given its simplicity, affordability, and accessibility, PLR may serve as a supplementary biomarker alongside established severity scores. Incorporating PLR into clinical practice may facilitate timely identification of high-risk patients and improve patient outcomes. Supplementary Information Additional file 1: Completed PRISMA checklist. Additional file 2: Search strategy. Additional file 3: Risk of bias assessment. Additional file 4: Additional analyses. Abbreviations Acute pancreatitis Platelet-to-lymphocyte ratio Severe acute pancreatitis Mild acute pancreatitis Moderately severe acute pancreatitis Preferred Reporting Items for Systematic Reviews and Meta-Analyses World Health Organization Virtual Health Library Standardized mean difference Confidence interval Intensive care unit Receiver operating characteristic Area under the curve Modified Marshall system score C-reactive protein Bedside Index of Severity in Acute Pancreatitis Computed tomography Neutrophil-to-lymphocyte ratio Acute Physiology and Chronic Health Evaluation II Joanna Briggs Institute persistent organ failure International Prospective Register of Systematic Reviews Authors’ contributions SM conceptualized the research idea, MI and SM managed and planned the research process. OA, AA, SM, AK, MA and MM undertook database searches and articles screening. MR, MA, MMR, undertook risk of bias assessment. MA, MHM, and AA extracted and summarized data. SM analyzed data. MI, MA and AA drafted the manuscript. All authors read and approved of the final manuscript. Funding No fund. Data availability No datasets were generated or analysed during the current study. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Additional file 1: Completed PRISMA checklist. Additional file 2: Search strategy. Additional file 3: Risk of bias assessment. Additional file 4: Additional analyses. Data Availability Statement No datasets were generated or analysed during the current study.