Can a five-variable model predict my 90-day outcome after aneurysmal subarachnoid hemorrhage?
After an aneurysmal subarachnoid hemorrhage (aSAH), doctors often want to estimate your recovery at 90 days. A recent study developed a simple model using five factors measured when you first arrive at the hospital: your age, your pre-stroke disability level (modified Rankin Scale, or mRS), the severity of your neurological symptoms (World Federation of Neurosurgical Societies grade), the amount of bleeding seen on CT scan (modified Fisher grade), and whether there is bleeding within the brain tissue (intracerebral hemorrhage). This model was tested in 245 patients and showed good ability to predict who would have a favorable outcome (mRS 0-3) at 90 days 2. The model's accuracy was high, with an area under the curve (AUC) of 0.917 in the development group and 0.868 in a separate validation group 2. So, yes, this five-variable model can give you and your doctor a reliable estimate of your 90-day outcome.
What the research says
The five-variable model was developed and validated in a single-center study of 245 patients with aSAH 2. The model uses admission data: age, pre-morbid mRS (a measure of how independent you were before the bleed), World Federation of Neurosurgical Societies (WFNS) grade (a measure of your consciousness level), modified Fisher grade (how much blood is on the initial CT scan), and presence of intracerebral hemorrhage (blood inside the brain tissue) 2. For predicting a favorable 90-day outcome (mRS 0-3, meaning you can walk and care for yourself), the model's performance was excellent: AUC of 0.917 in the development cohort and 0.868 in the temporal validation cohort 2. Even when the definition of favorable outcome was stricter (mRS 0-2), the AUCs were 0.920 and 0.840, respectively 2. The model also worked well in patients aged 70 and older, though with slightly lower accuracy (AUC 0.842 for mRS 0-3) 2. These results suggest the model is a reliable tool for early risk stratification.
Other research has explored different prediction approaches for aSAH. One study used machine learning (CatBoost model) to predict functional status at discharge, with key factors including Glasgow Coma Scale score, age, and liver function markers 3. Another study focused on predicting pneumonia after aSAH embolization, using variables like lactate dehydrogenase and body mass index 4. Additionally, a meta-analysis found that aneurysm location (front vs. back of the brain) does not significantly affect mortality or functional outcomes 5. While these studies address different aspects of aSAH prognosis, the five-variable model is specifically designed for 90-day outcome prediction using only admission data, making it practical for early use 2.
It's important to note that the five-variable model was developed in a single center and needs further validation in larger, more diverse populations 2. Also, the model predicts outcomes at a group level; individual results can vary. Your doctor can use this model as one piece of information, along with your overall health and treatment response, to guide your care.
What to ask your doctor
- Can you calculate my 90-day outcome using the five-variable model (age, pre-morbid mRS, WFNS grade, modified Fisher grade, and ICH on CT)?
- What do the model's results mean for my treatment plan and recovery expectations?
- Are there other factors, like my liver function or risk of delayed cerebral ischemia, that might affect my prognosis?
- How often will my outcome be reassessed, and are there any interventions that could improve my chances?
- Is this model used routinely at your center, and how does it compare to other prediction tools?
This question is drawn from common patient questions about this topic and answered using cited medical research. We do not provide individualized advice.