Prognostic model for aneurysmal subarachnoid hemorrhage patients requiring mechanical ventilation

Area of Science:

• Neurosurgery
• Critical Care Medicine
• Biostatistics

Background:

• Aneurysmal subarachnoid hemorrhage (aSAH) is a leading cause of mortality and disability.
• Predicting long-term prognosis for aSAH patients requiring mechanical ventilation presents significant challenges.

Purpose of the Study:

• To develop a predictive model for long-term outcomes in aSAH patients on mechanical ventilation.
• To utilize easily accessible clinical variables for prognostic estimation.

Main Methods:

• Least Absolute Shrinkage and Selection Operator (LASSO)-penalized Cox regression was employed.
• Prognostic factors were identified and validated using training and validation sets.
• Model performance was assessed using ROC curves, calibration curves, and decision curve analysis (DCA).

Main Results:

• Key independent prognostic factors identified include Simplified Acute Physiology Score 2, early brain injury, rebleeding, and ICU length of stay.
• The model demonstrated high predictive accuracy with AUC values of 0.82, 0.81, and 0.80 for 1-, 2-, and 4-year survival, respectively.
• A web-based nomogram was developed for clinical application.

Conclusions:

• The developed model serves as a valuable tool for predicting long-term outcomes in mechanically ventilated aSAH patients.
• This tool can aid in personalized interventions and improved patient care strategies.