Predictive Value of Combined CRP and INR for Intracranial Hypertension in Cerebral Venous Thrombosis

TL;DR

This study shows that combining CRP and INR can predict intracranial hypertension in cerebral venous thrombosis patients without invasive testing.

Contribution

A noninvasive predictive model using CRP and INR for intracranial hypertension in CVT patients is proposed and validated.

Findings

Patients with CRP > 5.5 g/L or INR < 0.99 had higher risk of intracranial hypertension.

High CRP and low INR conferred 9.778 times higher risk of intracranial hypertension.

Adding CRP and INR improved risk discrimination and reclassification for intracranial hypertension.

Abstract

Intracranial hypertension (IH) is a frequently observed clinical manifestation of cerebral venous thrombosis (CVT), which reflects the severity of the disease. The gold standard of intracranial pressure (ICP) is through invasive lumbar puncture.

We aimed to develop a noninvasive model combining biomarkers and clinical features to predict IH in CVT patients, facilitating early risk stratification.

The patients with CVT were consecutively enrolled in the Second Affiliated Hospital of Soochow University and the First Affiliated Hospital of Soochow University between January 2011 and June 2024, which were divided into two groups: CVT‐IH group and CVT + IH group based on ICP levels. Additionally, participants were further categorized into four groups according to the cut‐off of C‐reactive protein (CRP) and international normalized ratio (INR) by the receiver operating characteristic (ROC) curves. Logistic regression models were employed to calculate odds ratios (ORs) and 95% confidence intervals (CIs) of IH across the four subgroups.

157 individuals were finally included, 61 of whom had IH. Participants with CRP > 5.5 g/L or INR < 0.99 were more likely to experience IH. Those with high CRP and low INR conferred 9.778 times higher risk of IH compared with that of those with low CRP and high INR. Simultaneously adding CRP and INR to the basic model with established risk factors significantly improved risk discrimination and reclassification for IH of CVT patients.

Combination of CRP and INR better predicted the occurrence of IH for CVT patients, which might provide a noninvasive way of assessing ICP of CVT patients.

Figures