Aneurysmal Subarachnoid Hemorrhage (aSAH) Outcome Prediction

• Serum peroxiredoxin 2 emerges as a predictive biomarker of poor neurological outcome and delayed cerebral ischemia after aneurysmal subarachnoid hemorrhage: a prospective cohort study
• The relationship between the triglyceride-glucose index and functional outcomes in patients with aneurysmal subarachnoid hemorrhage: a retrospective cohort study
• Invisible Until It Burst: Unexpected Subarachnoid Hemorrhage From a Rapid-Onset Infectious Aneurysm in a Patient With Endocarditis
• Evaluating extreme temperature values and patient outcomes in aneurysmal subarachnoid hemorrhage: Post-hoc insights from the Earlydrain trial
• Evaluation of cerebral blood flow after subarachnoid hemorrhage using near-field coupling and machine learning
• Prognostic Factors and Imaging Strategies in Unknown Subarachnoid Hemorrhage: A Retrospective Study
• Risk factors for surgery-related cerebral infarction after aneurysmal subarachnoid hemorrhage
• Different cerebrospinal fluid drainage methods and chronic hydrocephalus in patients with aneurysmal subarachnoid hemorrhage

Outcome prediction for patients with aneurysmal subarachnoid hemorrhage (aSAH) is critical for guiding clinical decision-making and providing prognostic information. Prediction models for aSAH typically consider clinical, radiological, and laboratory parameters.

• Initial neurological status:
  1. Measured by scales like the Glasgow Coma Scale (GCS) or the World Federation of Neurosurgical Societies (WFNS) score.
  2. Poor initial neurological status correlates with worse outcomes.
• Age:
  1. Older patients tend to have worse outcomes.
• Comorbidities:
  1. Conditions like hypertension, diabetes, or cardiovascular disease may affect recovery.

• Fisher Grade:
  1. Indicates the amount and distribution of blood on initial CT. Higher grades are associated with a higher risk of vasospasm and worse outcomes.
• Aneurysm characteristics:
  1. Location, size, and rupture status of the aneurysm are crucial for prognosis.
• Presence of hydrocephalus:
  1. May indicate increased intracranial pressure and worse outcomes.
• Cerebral edema or infarction:
  1. Suggests secondary brain injury, often linked to poor outcomes.

• Serum sodium levels:
  1. Hyponatremia is common in aSAH and may indicate complications like cerebral salt-wasting syndrome.
• Inflammatory markers:
  1. Elevated C-reactive protein (CRP) or white blood cell count correlates with worse outcomes.
• Markers of brain injury:
  1. Levels of S100B or neuron-specific enolase (NSE) may reflect neuronal damage.

Several models and scoring systems have been developed to predict the outcomes of aSAH patients:

• Grades I to V based on clinical severity.
• Higher grades are associated with poor outcomes.

• Combines GCS and motor deficits.
• Simple and widely used in clinical practice.

• Assesses the amount of subarachnoid blood on CT. Higher grades predict vasospasm and poor outcomes.

• Used to assess functional outcomes, typically at follow-up.

• Incorporates clinical and imaging data to predict survival and functional outcome.

• Uses factors like age, WFNS grade, and imaging features to predict long-term outcomes.

Machine learning (ML) advances are enhancing the ability to predict aSAH outcomes. ML models integrate large datasets with complex interactions to predict outcomes with improved accuracy. Examples include:

• Random forests, support vector machines, or neural networks trained on clinical and imaging data.
• Use of AI to analyze CT angiography or digital subtraction angiography for subtle vasospasm or aneurysm rupture indicators.

Outcome prediction models are used to:

1. Guide treatment decisions (e.g., timing of surgical clipping or coiling).
2. Stratify risk for complications, such as vasospasm or delayed cerebral ischemia (DCI).
3. Inform patients and families about prognosis.
4. Facilitate research by standardizing outcome assessments.

1. Many models require external validation for different populations.
2. Predictions are probabilistic, not deterministic, and should complement clinical judgment.
3. Outcome depends not only on the initial severity but also on the quality of care and complications.

A retrospective cohort study included 82 patients suffering from aSAH. Basaran et al. evaluated the predictive efficacy of AtlasGPT and ChatGPT 4.0 by examining the area under the curve (AUC), sensitivity, specificity, and Youden's Index, in comparison to established clinical grading scales such as the World Federation of Neurological Surgeons (WFNS) scale, Simplified Endovascular Brain Edema Score (SEBES), and Fisher scale. This assessment focused on four endpoints: in-hospital mortality, the need for decompressive hemicraniectomy, and functional outcomes at discharge and after 6-month follow-up. In-hospital mortality occurred in 22% of the cohort, and 34.1% required decompressive hemicraniectomy during treatment. At hospital discharge, 28% of patients exhibited a favorable outcome (mRS ≤ 2), which improved to 46.9% at the 6-month follow-up. Prognostication utilizing the WFNS grading scale for 30-day in-hospital survival revealed an AUC of 0.72 with 59.4% sensitivity and 83.3% specificity. AtlasGPT provided the highest diagnostic accuracy (AUC 0.80, 95% CI: 0.70-0.91) for predicting the need for decompressive hemicraniectomy, with 82.1% sensitivity and 77.8% specificity. Similarly, the WFNS score and AtlasGPT demonstrated high prognostic values for discharge outcomes with AUCs of 0.74 and 0.75, respectively. Long-term functional outcome predictions were best indicated by the WFNS scale, with an AUC of 0.76. The study demonstrates the potential of integrating AI models such as AtlasGPT with clinical scales to enhance outcome prediction in aSAH patients. While established scales like WFNS remain reliable, AI language models show promise, particularly in predicting the necessity for surgical intervention and short-term functional outcomes. The study explored the use of advanced AI language models, AtlasGPT and ChatGPT 4.0, to predict outcomes for patients with aneurysmal subarachnoid hemorrhage (aSAH). It found that AtlasGPT provided the highest diagnostic accuracy for predicting the need for decompressive hemicraniectomy, outperforming traditional clinical scales, while both AI models showed promise in enhancing outcome predictions when integrated with established clinical assessment tools ¹⁾

This study demonstrates the potential of AI models like AtlasGPT and ChatGPT 4.0 to enhance outcome prediction in aSAH, particularly for short-term functional outcomes and surgical interventions. However, its limitations, including a small sample size and lack of external validation, necessitate further research before these models can be widely adopted in clinical practice. The integration of AI with established tools appears promising, but future studies must address gaps in generalizability, clinical applicability, and interpretability to realize the full potential of AI in aSAH management.