AUTHOR=Dong Jinjiang , Ye Xiaofeng , Wang Popo , Ye Qijun , Peng Deqing 

TITLE=HMOX1 mRNA combined with early imaging parameters for predicting the prognosis of hypertensive cerebral hemorrhage patients undergoing conservative treatment

JOURNAL=Frontiers in Neurology

VOLUME=Volume 16 - 2025

YEAR=2025

URL=https://www.frontiersin.org/journals/neurology/articles/10.3389/fneur.2025.1657188

DOI=10.3389/fneur.2025.1657188

ISSN=1664-2295

ABSTRACT=ObjectiveTo address the limitations of current clinical tools in predicting the prognosis of hypertensive cerebral hemorrhage (HICH) patients undergoing conservative treatment, this study aimed to evaluate the predictive value of combining HMOX1 mRNA levels with early imaging parameters.MethodsThis retrospective diagnostic accuracy study involved 208 HICH patients (mean age 50 ± 9.5 years, 149 men) treated from December 2022 to December 2024. Patients were divided into good prognosis (mRS score ≤ 2, n = 112) and poor prognosis (mRS score > 2, n = 96) groups based on 90-day mRS scores. The index tests were HMOX1 mRNA levels and early imaging parameters (hemorrhage location, hematoma volume, and shape), with the reference standard being the 90-day mRS score. Statistical analyses included hierarchical regression, multivariate logistic regression, and ROC curve evaluations. Bootstrap resampling (1,000 samples) confirmed that the OR values of key risk factors were stable.ResultsHMOX1 mRNA expression was higher in the poor prognosis group (1.36 ± 0.41 vs. 1.00 ± 0.22, p < 0.001). Delong test showed that the AUC of the combination of HMOX1 mRNA and imaging parameters (0.871, 95% CI: 0.818–0.924) was significantly higher than that of single indicators (all p < 0.05). A nomogram model incorporating these factors plus clinical and laboratory indicators showed good discrimination (C-index: 0.814) and calibration (AUC: 0.94), and its AUC was significantly higher than that of the above combined indicator (p = 0.003).ConclusionThe ‘HMOX1 mRNA + imaging parameters' combined indicator and the nomogram model both significantly improve prognostic prediction accuracy in HICH patients undergoing conservative treatment, with the nomogram model showing the best efficiency.