Role of Neutrophil-to-Lymphocyte Ratio and Platelet-to-Lymphocyte Ratio As Diagnostic and Prognostic Markers in Cerebral Venous Sinus Thrombosis: A Systematic Review and Meta-Analysis

Introduction: Cerebral venous sinus thrombosis (CVST) is an uncommon etiology of stroke, accounting for about 1.0% of cases. It predominantly presents with subacute headaches in young individuals, complicating the differential diagnoses from primary headaches. Although magnetic resonance venography and computerized tomography are the gold standards for diagnosis, their uses were resource-limited settings. Inflammation plays a pivotal role in CVST pathogenesis. Systemic inflammatory markers such as neutrophil-to-lymphocyte ratio (NLR) and platelet-to-lymphocyte ratio (PLR), derived from a complete blood count, have shown promise in diagnosis and prognosis. This meta-analysis aims to clarify the clinical application of NLR and PLR in diagnosing CVST.

Objective: The primary objective was to evaluate the role of NLR and PLR in diagnosing CVST. The secondary objective was to explore the relationship between these markers and the prognosis of CVST.

Methods: We conducted a systematic search through PubMed, Embase, and Scopus until April 6, 2024. The search strategy included terms such as “cerebral venous sinus thrombosis,” “neutrophil to lymphocyte ratio,” and “platelet to lymphocyte ratio.” Independent reviewers assessed the quality of papers and extracted data on NLR and PLR obtained from blood samples collected upon hospital admission among patients with confirmed CVST diagnosis and those without CVST. We also extracted and analyzed the data from CVST patients with good and poor prognoses. This was defined by the modified Rankin score (mRS), with mRS 0-2 indicating a good prognosis and mRS 3-6 indicating a poor prognosis. Weighted mean differences (WMD), along with their 95% confidence intervals (CI), were calculated using the DerSimonian-Laird random-effects model by the “meta esize” command in Stata 17.0. Additionally, pooled sensitivity and specificity were determined by the “metadta” command.

Results: From 1,635 papers identified, 7 case-control studies, including 555 patients with CVST diagnosis and 955 without CVST, met the eligibility criteria. Three out of seven studies reported CVST prognosis from 326 patients with good prognosis and 89 with poor prognosis. NLR was higher in CVST patients than in controls (WMD 1.53, 95% CI 0.96-2.10, I² 88%). Subgroup analyses also showed increased NLR in CVST patients compared to healthy controls and CVST-mimic diseases (WMD 1.90, 95% CI 0.36-3.44, I² 93%; and 1.67, 95% CI 0.65-2.69, I² 90%, respectively). PLR was higher in CVST compared to controls, healthy individuals, and mimic diseases (WMD 39.06, 95% CI 26.47-51.66, I² 52%; 45.38, 95% CI 14.66-76.10, I² 58%; 31.19, 95% CI 12.26-50.12, I² 56%, respectively). Three studies with an NLR cut-off at 2.1 had pooled sensitivity and specificity in predicting CVST of 75% (95% CI 67-81, I² 54%) and 71% (95% CI 63-77, I² 40%), respectively. Patients with poor prognosis exhibited higher NLR and PLR compared to those with good prognosis (WMD 10.26, 95% CI 7.85-12.68, I² 77%; and 186.21, 95% CI 95.76-276.66, I² 94%, respectively).

Conclusions: Elevated NLR and PLR are associated with CVST diagnosis, especially in those with poor prognosis. Our meta-analysis is the first to explore an NLR cut-off at 2.1 as an adjunctive diagnostic threshold for CVST. Future research should focus on validating the clinical utility of NLR and PLR as an initial screening and prognostic tool, especially in resource-limited settings.

No relevant conflicts of interest to declare.