Association between temporal trajectories of inflammatory markers and prognosis in ischemic stroke

The post-stroke inflammatory response is a dynamic and heterogeneous process that cannot be fully captured by single time-point measurements. This study aimed to characterize the temporal trajectories of four inflammatory markers-neutrophil-to-lymphocyte ratio (NLR), monocyte-to-lymphocyte ratio (MLR), systemic immune-inflammation index (SII), and systemic inflammation response index (SIRI)-during the acute phase of ischemic stroke and to evaluate their associations with 3-month functional outcomes.

We retrospectively analyzed patients with acute ischemic stroke admitted within 24 h of onset at Dongyang People's Hospital from January to December 2024. Complete blood counts collected within the first 7 days were used to construct latent class growth models. Model selection was based on Akaike Information Criterion, the Bayesian Information Criterion, entropy, average posterior probability, and minimum class size. Logistic regression was performed to examine associations between inflammatory trajectories and 3-month modified Rankin Scale outcomes.

A total of 735 patients were included. Distinct inflammatory trajectories were identified for each marker: three classes for NLR (stable-low, rapid-increase, gradual-decrease), three for MLR (persistent-low, persistent-moderate, persistent-high), three for SII (stable-low, persistent-moderate, marked-decline with rapid-rebound), and four for SIRI (stable-low, gradual-increase, gradual-decrease, rapid-increase). Stable-low patterns constituted the majority across all markers. For NLR, the rapid-increase trajectory consistently predicted poor 3-month outcomes (Model 3: OR 3.492, 95% CI 1.514-10.265). For SIRI, both gradual-increase and rapid-increase trajectories were independently associated with poor prognosis (Model 3: OR 3.894, 95% CI 1.395-10.872; OR 4.264, 95% CI 1.201-15.136). In contrast, MLR and SII trajectories were not associated with functional outcomes after full adjustment.

Distinct temporal inflammatory trajectories are observed during the 1 week after ischemic stroke. Increasing trajectories of NLR and SIRI strongly predict poor 3-month outcomes. These findings underscore the clinical importance of monitoring early inflammatory evolution rather than relying on single measurements and provide a novel framework for prognostic assessment in ischemic stroke.