The C.O.S.T.A.S. project (Clinically Oriented Spasmolysis Treatment- A Scorescale): a pilot study.

INTRODUCTION

Symptomatic cerebral vasospasm (CVS) following aneurysmal subarachnoid hemorrhage (SAH) presents a significant therapeutic dilemma. Intra-arterial spasmolysis is a potentially life-saving intervention, associated with procedural risks.

RESEARCH QUESTION: The aim of C.O.S.T.A.S. (Clinically Oriented Spasmolysis Treatment - A Score Scale) project is to develop a clinically based, patient-oriented scoring system, to identify patients with SAH who may benefit from intra-arterial spasmolysis for CVS.

MATERIAL AND

METHODS

Parameters and clinical data from SAH-patients were categorized and weighted according to their clinical relevance creating the scoring-system. Parameters included are age, smoking, intracranial arterial atherosclerosis, body-mass index, renal function, mechanical ventilation metrics, noradrenaline requirements, and an artificial intelligence (AI)-based vasospasm-risk prediction model. Each parameter was assigned a score ranging from 0 to 3 points, with a maximum cumulative score of 18.

RESULTS

The scale provides a comprehensive overview of each patient's clinical status. It integrates data from artificial intelligence-based model, along with other relevant factors, identifying patients at higher risk of developing CVS who may benefit from intra-arterial spasmolysis. In addition, it serves as a triage tool by efficiently assessing the patient's condition and determining their suitability for invasive intervention, balancing the potential risks and benefits of the procedure.

DISCUSSION AND

CONCLUSION

The C.O.S.T.A.S. project is the first scoring-scale specifically designed to guide intra-arterial spasmolysis in SAH patients with CVS. While currently in prototype form, it offers a practical, transparent, and adaptable framework for individualized treatment decisions. A multicenter application, testing, and indexing are crucial for achieving maximal classification accuracy and parameter weighting.