Abstract
BACKGROUND: Spontaneous intracerebral hemorrhage (ICH) is an acute cerebrovascular disease associated with high mortality and severe disability. Inflammation plays a critical role in the onset and progression of ICH. However, existing inflammatory markers have limited predictive capacity for the prognosis of ICH patients. This study aims to develop a novel Prognostic inflammation index (PII) based on leukocyte subset counts and evaluate its effectiveness in assessing the prognosis of ICH patients. METHODS: A total of 1021 consecutive ICH patients hospitalized between January 2021 and June 2023 were included as the derivation cohort. In addition, an internal temporal validation cohort of 366 patients hospitalized between January 2024 and December 2024 was assembled using identical inclusion/exclusion criteria. Using reduced-rank regression (RRR) based on leukocyte subsets (including neutrophils, monocytes, lymphocytes, eosinophils, and basophils), we constructed the PII. Multivariate logistic regression was employed to analyze the associations between PII, its dynamic trajectories, and patient outcomes, including poor prognosis, all-cause mortality, and stroke-associated infections. The predictive performance of PII was illustrated using a nomogram, and its efficacy was compared to conventional inflammatory markers through receiver operating characteristic (ROC) curve analysis. RESULTS: Patients with elevated PII were significantly associated with poor outcomes at 3, 6, and 12 months, stroke-associated infections, and all-cause mortality within 1 year (all p < 0.05). PII trajectory analysis revealed that patients with persistently high PII had a substantially increased risk of poor outcomes (p < 0.05). Moreover, compared to common systemic inflammatory markers such as the systemic immune-inflammation index (SII), systemic inflammation response index (SIRI), and aggregate inflammation systemic index (AISI), PII showed generally favorable discriminative performance across most endpoints; however, the pairwise AUC comparisons were exploratory and some comparisons yielded borderline p values that should be interpreted cautiously. CONCLUSION: PII, a composite inflammation index based on leukocyte subset counts, is an effective predictor of poor outcomes in ICH patients and shows favorable prognostic performance compared with traditional inflammatory markers.