Pathophysiology of Early Brain Injury and Its Association with Delayed Cerebral Ischemia in Aneurysmal Subarachnoid Hemorrhage: A Review of Current Literature

Aneurysmal subarachnoid hemorrhage (aSAH) accounts for ~5% of strokes and carries high morbidity and mortality (overall mortality ~35%; ~20% of survivors with significant morbidity). Delayed cerebral ischemia (DCI), affecting ~20–40% of aSAH patients, is a major determinant of poor outcomes and typically manifests 3–4 days post-ictus, peaking at 7–8 days. Earlier onset (<7 days) correlates with higher mortality and infarct load. Historically, large-vessel vasospasm was considered central to DCI; however, agents like clazosentan improved angiographic vasospasm without improving outcomes, and infarcts often occur outside vasoconstricted territories. Standardized definitions now incorporate clinical deterioration and cerebral infarction (Vergouwen 2010). This review examines the pathophysiology of EBI in the first 72 hours after ictus—autoregulation failure, neuroinflammation, microthrombosis, BBB breakdown, spreading depolarizations—and its association with DCI, as well as risk factors, biomarkers, monitoring, imaging, and predictive models to inform future management and research.

The review synthesizes evidence redefining DCI as a multifactorial process initiated by early brain injury (EBI). It outlines evolution from a vasospasm-centric model to one emphasizing microvascular dysfunction, autoregulatory failure, inflammation, and cortical spreading depolarizations. Standardization of DCI definitions (Vergouwen et al., 2010) improved outcome correlations. Risk factors include smoking, female sex, diabetes, and poor clinical grade; genetic contributions are suggested by family and twin studies. Genetic association results for vasospasm/DCI are heterogeneous; some studies highlight NRG1 expression (AUROC ~0.96) and DNA methylation markers (INSR, CDHR5, STEAP3). Pathophysiologic mechanisms include transient global cerebral ischemia at ictus due to ICP surges reducing CPP; sympathetic catecholamine surge (with systemic effects); glutamate-mediated excitotoxicity; nitric oxide dysregulation leading to vasoconstriction and BBB breakdown; calcium channel activation; platelet aggregation with microthrombi; and spreading depolarizations causing cytotoxic edema and propagating injury. Biomarkers such as macrophage migration inhibitory factor (MIF) correlate with severity, outcomes, and DCI risk. Imaging markers of EBI/DCI risk include early MRI ADC changes, CT-based SEBES score, and automated volumetric measures (selective sulcal volume). Multimodal monitoring (PbtO2, microdialysis LPR, EEG including quantitative measures, pupillometry) and CT perfusion parameters (e.g., prolonged MTT, reduced CBF) contribute to earlier DCI detection. The review also discusses combined clinical-radiologic scores (VASOGRADE, HAIR), machine learning-based prediction models, and the need for multicenter standardized research.

- DCI incidence and impact: DCI occurs in ~20–40% of aSAH and is strongly associated with poor outcomes; standardized DCI definition correlates with outcomes (e.g., OR for poor outcomes at discharge 2.65 in patients with clinical DCI deterioration). - Vasospasm vs outcomes: Despite frequent vasospasm (50–70%), only 20–40% develop DCI; clazosentan reversed vasospasm without outcome benefit, and infarcts often occur outside spastic vessels. - EBI as driver: EBI within first 72 h is the main contributor to mechanisms leading to DCI, including autoregulatory failure, catecholamine surge, excitotoxicity, NO depletion, BBB disruption, microthrombosis, and spreading depolarizations. - Biomarkers: Serum MIF levels are elevated in aSAH vs controls and independently predict 6-month unfavorable outcomes; higher MIF associates with DCI and outperforms APACHE-II, CRP, IL-6 for DCI discrimination (AUC 0.780, 95% CI 0.710–0.849). - Imaging indices of EBI/DCI risk: - Early MRI shows cytotoxic/vasogenic edema (ADC changes) not seen on CT; in high-grade cohorts, 71% mortality and only 12% satisfactory outcomes when early diffusion restriction present. - SEBES (CT-based 0–4): High scores (3–4) independently predict DCI (OR 2.24, 95% CI 1.58–3.17) and unfavorable outcomes (OR 3.45, 95% CI 1.95–6.07); associated with ICP therapies, decompressive surgery, and infarction. - Volumetric global cerebral edema: Selective sulcal volume (SSV) <5 mL identifies GCE; early low SSV within 72 h independently predicts poor outcomes, especially in patients <70 years. - Spreading depolarizations: DISCHARGE-1 showed peak total depression time predicts delayed infarction/DCI; 60-min cutoff sensitivity 0.76, specificity 0.59; 180-min cutoff sensitivity 0.62, specificity 0.83. SDs likely serve as independent markers of brain injury; ketamine can suppress SDs; memantine shows promise preclinically. - Multimodal monitoring: Drops in PbtO2 correlate with vasospasm/DCI; higher cerebral lactate blunts PbtO2 response to FiO2 challenge (a surrogate of ischemia risk); non-linear PbtO2–CPP relationship shows suboptimal CPP associates with hypoxia. Microdialysis LPR elevations associate with ischemia and poor outcomes. NIRS lacked correlation with PbtO2 in SAH. Abnormal decreases in Neurological Pupil index and sonographic vasospasm individually associate with DCI. - EEG predictors: Decreasing alpha-delta ratio and other EEG alarms can predict DCI days in advance; quantitative EEG aids early detection. - CT perfusion: Perfusion deficits confer ~23-fold higher likelihood of DCI; thresholds such as MTT >6.4 s or regional CBF <25 mL/100 g/min correlate with DCI; perfusion deficits during deterioration predict subsequent infarction (88% vs 38%). Variability in processing and thresholds limits universal adoption. - Risk factors: Smoking is a modifiable risk for DCI; independent risks include female sex, diabetes, and poor SAH grade; older age with good grade may confer lower DCI risk. Genetic predisposition is substantial (heritability ~41%); results for specific variants are inconsistent; NRG1 expression predicts DCI (AUROC 0.96); methylation markers (INSR, CDHR5; STEAP3 site cg25713625) associate with DCI/unfavorable outcomes. Weighted polygenic risk scores show high AUROC for intracranial aneurysm (0.95) and AIS (0.842). - Grading and prediction: Combined scores (VASOGRADE, HAIR) stratify DCI and mortality risk; dynamic and ML models improve prediction of DCI and outcomes.

The review reframes DCI as a consequence of early, multifactorial brain injury rather than solely large-vessel vasospasm. Recognizing EBI’s central role explains why vasospasm-focused therapies alone failed to improve outcomes and underscores early detection and treatment of autoregulatory failure, inflammation, microthrombosis, BBB disruption, and spreading depolarizations. Incorporating biomarkers (e.g., MIF), imaging (SEBES, volumetric sulcal metrics, CT perfusion), and multimodal monitoring (PbtO2, microdialysis, EEG, pupillometry) can identify patients at risk before irreversible injury. Predictive models and combined grading systems can guide personalized interventions. These insights inform clinical practice by promoting early, targeted management strategies and point to research directions focusing on microvascular and microcellular mechanisms to ultimately improve functional outcomes post-aSAH.

DCI after aSAH is a complex, multifactorial process initiated by early brain injury. Mechanisms include sympathetic surge, cytokine release, NO dysregulation with vasoconstriction, microthrombosis, BBB breakdown, and spreading depolarizations. Shifting focus from macrovessel vasospasm to EBI, global cerebral edema, and microcellular pathophysiology can improve risk stratification and management. Radiographic tools (SEBES, volumetric sulcal measures), advanced monitoring (PbtO2, microdialysis, EEG), and perfusion imaging enable earlier detection of evolving ischemia. Future work should validate biomarkers and predictive models, standardize monitoring and imaging thresholds, and pursue multicenter studies to translate mechanistic insights into therapies that improve outcomes.

As a narrative review, no systematic methodology or predefined search strategy is provided. Many cited studies are observational, small, or single-center, with heterogeneous populations and variable definitions, limiting generalizability. Invasive multimodal monitoring lacks standardization in placement, thresholds, and interpretation, contributing to mixed results. CT perfusion analyses vary by post-processing and quantitative thresholds, affecting sensitivity/specificity. Genetic association studies show inconsistent and population-specific findings due to sample sizes, heterogeneity, and potential confounding (e.g., stratification, gene–environment interactions). Overall, variability in methods and measures across studies constrains firm clinical recommendations and underscores the need for large, multicenter, standardized prospective trials.