Previously submitted to: Interactive Journal of Medical Research (no longer under consideration since Jun 08, 2026)
Date Submitted: Oct 6, 2025
Warning: This is an author submission that is not peer-reviewed or edited. Preprints - unless they show as "accepted" - should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.
Utility of Dynamic Hemoglobin-to-Red Cell Distribution Width Ratio (HRR) Trajectories for In-Hospital Mortality Risk Stratification in Patients with Intracerebral Haemorrhage: A Dual-database Retrospective Analysis
ABSTRACT
Background:
Intracerebral haemorrhage (ICH) causes high acute mortality and long-term disability. Dynamic biomarkers aid real-time risk stratification. The hemoglobin-to-red cell distribution width ratio (HRR) integrates oxygen-carrying capacity and inflammatory stress, but its temporal association with ICH mortality remains unclear.
Objective:
This study aimed to characterize the patterns of dynamic HRR trajectories in hospitalized ICH patients and explore the utility of these trajectories for in-hospital mortality risk stratification, with validation based on two independent databases.
Methods:
This retrospective study included 998 ICH patients from MIMIC-IV and 857 from eICU. K-means clustering grouped 7-day HRR trajectories; mixed linear models (MLMs) and confounder-adjusted Cox regression evaluated associations with in-hospital mortality. ROC curves using temporal HRR values at different time points validated mortality predictive value. Sensitivity analyses (Schoenfeld test and subgroup analysis) assessed robustness.
Results:
Three HRR clusters were observed, including Cluster 1 (rapid decline to low levels) with mortality of 25.4% in MIMIC-IV and 21.0% in eICU, Cluster 2 (gradual decline) with mortality of 14.8% and 15.4%, and Cluster 3 (high-stable/mild increase) with the lowest mortality of 8.6% and 5.7%. MLMs demonstrated higher HRR in Clusters 2 and 3 compared with Cluster 1 (all P<0.001), and non-survivors presented lower HRR and steeper declines. Cluster 3 independently reduced mortality risk, with HR=0.522 (95% CI: 0.319-0.854) in MIMIC-IV and HR=0.370 (95% CI: 0.197-0.696) in eICU. For mortality prediction, 7-day temporal HRR outperformed single-time-point HRR, and AUC improved over time; the peak AUC was 0.696 in MIMIC-IV and 0.730 in eICU, both observed in 7-day temporal models.
Conclusions:
Dynamic HRR trajectories independently predict in-hospital mortality in ICH. Low or declining HRR identifies high-risk patients while high-stable or mildly increasing HRR correlates with better outcomes, rendering HRR a cost-effective and readily measurable biomarker derived from routine complete blood count, a test available in most primary care settings.
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.