# Cerebral vasospasm following arteriovenous malformation rupture: a population-based longitudinal study

## TL;DR

This study finds that cerebral vasospasm after ruptured brain malformations is rare but strongly linked to higher mortality and specific risk factors like subarachnoid hemorrhage and hydrocephalus.

## Contribution

The study is the first longitudinal analysis identifying risk factors for vasospasm and mortality after ruptured arteriovenous malformations using a large population-based dataset.

## Key findings

- Vasospasm occurred in 3% of ruptured AVM cases and was associated with more than double the mortality at 30 days and one year.
- Subarachnoid hemorrhage, hydrocephalus, and male sex were significant risk factors for vasospasm.
- Coma, hydrocephalus, and chronic kidney disease were the strongest predictors of 30-day mortality.

## Abstract

Vasospasm is a devastating sequelae of ruptured arteriovenous malformations (AVMs) in adults. Comorbidities and presenting factors have been suggested as risks, but only in cross-sectional studies. The objective of this study was to characterize risk factors associated with vasospasm and mortality in ruptured AVM.

Adult patients from the TriNetX Research Network were included, based on the ICD-10 codes, over a period of 20 years. Cox proportional hazard models assessed the hazards of vasospasm (I67.84) and mortality separately, adjusting for age, sex, comorbidities, substance use history, presenting factors (e.g., hypernatremia, hypokalemia), criteria of the National Inpatient Sample-Subarachnoid Hemorrhage Severity Score, and location of hemorrhage. Outcomes were assessed in the first 30 days following rupture.

Among 10,375 patients with ruptured AVMs, 523 (5.3%) died and 297 (3.0%) experienced vasospasm in the first 30 days. After matching for age and sex, vasospasm was associated with increased mortality at three months (11.1% vs. 4.8%, p = 0.003), six months (12.6% vs. 5.1%, p = 0.001), and one year (13.5% vs. 6.9%, p = 0.005). Female sex was protective against vasospasm within 30 days (HR = 0.714, p = 0.007) while the greatest risk factors present on admission included subarachnoid hemorrhage (6.086, p < 0.001), hydrocephalus (3.783, p < 0.001), and leukocytosis (2.0677, p < 0.001). The greatest risk factors for 30-day mortality were coma (HR = 3.700, p < 0.001), hydrocephalus (2.698, p < 0.001), and chronic kidney disease (1.596, p = 0.003).

In this retrospective cohort study of 10,375 adult patients with ruptured AVMs, vasospasm occurred in approximately 3%. Risk factors for vasospasm included subarachnoid hemorrhage, male sex, hydrocephalus, and leukocytosis. The presence of vasospasm was associated with a more than twofold increase in mortality at both 30 days and one year.