Related Experiment Video
Updated: Jun 11, 2026
Intracranial Pressure Monitoring In Nontraumatic Intraventricular Hemorrhage Rodent Model
Published on: February 8, 2022
Do-not-resuscitate orders and predictive models after intracerebral hemorrhage
D B Zahuranec1, L B Morgenstern, B N Sánchez
1University of Michigan Cardiovascular Center, 1500 East Medical Center Drive, SPC#5855, Ann Arbor, MI 48109-5855, USA. zdarin@umich.edu
Insights
Intracerebral hemorrhage (ICH) predictive models show reduced accuracy when early do-not-resuscitate (DNR) orders are present. Clinicians must use caution, as DNR status significantly impacts model performance and perceived accuracy.
Area of Science:
- Neurology
- Clinical Epidemiology
- Medical Informatics
Background:
- Intracerebral hemorrhage (ICH) is a critical condition with significant mortality.
- Predictive models are used to estimate outcomes in ICH patients.
- The impact of early do-not-resuscitate (DNR) orders on model accuracy is not well understood.
Purpose of the Study:
- To evaluate the accuracy of common ICH predictive models.
- To compare model performance in ICH patients with and without early DNR orders.
Main Methods:
- Retrospective analysis of 487 spontaneous ICH cases.
- Comparison of the ICH Score, Cincinnati model, and ICH grading scale (ICH-GS) against observed 30-day mortality.
- Stratification of analysis based on the presence of early DNR orders.
Main Results:
- Overall observed 30-day mortality was 42.7%.
- Models underestimated mortality in patients with early DNR orders (observed 83.5% vs. predicted 57.2-77.8%).
- Models showed varied performance in patients without DNR orders (observed 20.8% vs. predicted 26.6-41.1%).
Conclusions:
- Early DNR status significantly impacts the performance of ICH predictive models.
- Failure to account for DNR status may lead to a false sense of model accuracy.
- Clinical application of these models requires careful consideration of patient DNR status.
Objective:
To quantify the accuracy of commonly used intracerebral hemorrhage (ICH) predictive models in ICH patients with and without early do-not-resuscitate orders (DNR).
Methods:
Spontaneous ICH cases (n = 487) from the Brain Attack Surveillance in Corpus Christi study (2000-2003) and the University of California, San Francisco (June 2001-May 2004) were included. Three models (the ICH Score, the Cincinnati model, and the ICH grading scale [ICH-GS]) were compared to observed 30-day mortality with a chi(2) goodness-of-fit test first overall and then stratified by early DNR orders.
Results:
Median age was 71 years, 49% were female, median Glasgow Coma Scale score was 12, median ICH volume was 13 cm(3), and 35% had early DNR orders. Overall observed 30-day mortality was 42.7% (95% confidence interval [CI] 38.3-47.1), with the average model-predicted 30-day mortality for the ICH Score, Cincinnati model, and ICH-GS at 39.9% (p = 0.005), 40.4% (p = 0.007), and 53.9% (p < 0.001). However, for patients with early DNR orders, the observed 30-day mortality was 83.5% (95% CI 78.0-89.1), with the models predicting mortality of 64.8% (p < 0.001), 57.2% (p < 0.001), and 77.8% (p = 0.02). For patients without early DNR orders, the observed 30-day mortality was 20.8% (95% CI 16.5-25.7), with the models predicting mortality of 26.6% (p = 0.05), 31.4% (p < 0.001), and 41.1% (p < 0.001).
Conclusions:
ICH prognostic model performance is substantially impacted when stratifying by early DNR status, possibly giving a false sense of model accuracy when DNR status is not considered. Clinicians should be cautious when applying these predictive models to individual patients.