Neurocritical care · 2026
Cerebral Vasospasm Detection and Delayed Cerebral Ischemia Prediction after Aneurysmal Subarachnoid Hemorrhage: A Scoping Review.
СтатьяScoping ReviewНазвание и аннотация: Перевод пока не подготовлен
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
de Andrade João Brainer Clares, de Lima Froio Nayara, Nascimento Paula Sanchez, Querobin Sophia Oliveira, de Oliveira Nathalia Souza, Carneiro Thiago S
Аннотация
BACKGROUND/OBJECTIVE: Aneurysmal subarachnoid hemorrhage (aSAH) is complicated by angiographic cerebral vasospasm and delayed cerebral ischemia (DCI), distinct entities with different reference standards and surveillance targets. We mapped the evidence on (1) diagnostic accuracy of vasospasm monitoring tools using digital subtraction angiography (DSA) as the reference standard and (2) detection and prediction of DCI using the 2010 consensus definition.
METHODS: We performed a Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR)-guided scoping review of PubMed/MEDLINE, Google Scholar, and PubMed Central (1 January 2005-31 January 2025). Vasospasm studies required DSA as comparator; DCI studies were categorized by outcome definition (2010 consensus vs. alternative definitions). Methodological quality was assessed with QUADAS-2 (diagnostic accuracy) and Prediction Model Risk of Bias Assessment Tool (PROBAST; prediction models).
RESULTS: From 762 records, 136 studies met inclusion criteria. For vasospasm detection (DSA reference), transcranial Doppler (18 studies; n = 3256) demonstrated sensitivity 66% (60-72) and specificity 97% (95-99), while computed tomography angiography (CTA) (15 studies; n = 2145) showed sensitivity 76% (72-80) and specificity 93% (91-95). For DCI detection (2010 consensus), CT perfusion (30 studies; n = 1786) yielded sensitivity 86% (78-86) and specificity 80% (75-85) with substantial protocol heterogeneity; continuous electroencephalography (cEEG) (8 studies; n = 523) achieved sensitivity 88% (80-94) and specificity 89% (83-93). Across 84 DCI predictor studies, machine-learning models (23 models) outperformed traditional risk factors (area under the curve [AUC] 0.68-0.88 vs. ~0.63), but only 44% underwent independent validation, and 67% were at high risk of bias.
CONCLUSIONS: The evidence supports a two-track framework: DSA-referenced tools for vasospasm detection and consensus-defined markers for DCI detection/prediction. Multimodal strategies integrating physiologic and imaging data are most promising, whereas machine-learning approaches require prospective, externally validated evaluations before routine clinical deployment.
METHODS: We performed a Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR)-guided scoping review of PubMed/MEDLINE, Google Scholar, and PubMed Central (1 January 2005-31 January 2025). Vasospasm studies required DSA as comparator; DCI studies were categorized by outcome definition (2010 consensus vs. alternative definitions). Methodological quality was assessed with QUADAS-2 (diagnostic accuracy) and Prediction Model Risk of Bias Assessment Tool (PROBAST; prediction models).
RESULTS: From 762 records, 136 studies met inclusion criteria. For vasospasm detection (DSA reference), transcranial Doppler (18 studies; n = 3256) demonstrated sensitivity 66% (60-72) and specificity 97% (95-99), while computed tomography angiography (CTA) (15 studies; n = 2145) showed sensitivity 76% (72-80) and specificity 93% (91-95). For DCI detection (2010 consensus), CT perfusion (30 studies; n = 1786) yielded sensitivity 86% (78-86) and specificity 80% (75-85) with substantial protocol heterogeneity; continuous electroencephalography (cEEG) (8 studies; n = 523) achieved sensitivity 88% (80-94) and specificity 89% (83-93). Across 84 DCI predictor studies, machine-learning models (23 models) outperformed traditional risk factors (area under the curve [AUC] 0.68-0.88 vs. ~0.63), but only 44% underwent independent validation, and 67% were at high risk of bias.
CONCLUSIONS: The evidence supports a two-track framework: DSA-referenced tools for vasospasm detection and consensus-defined markers for DCI detection/prediction. Multimodal strategies integrating physiologic and imaging data are most promising, whereas machine-learning approaches require prospective, externally validated evaluations before routine clinical deployment.