Predictive Model of Cerebral Vasospasm in Subarachnoid Hemorrhage Based on Regression Equation

Machine learning predicts cerebral vasospasm in patients with subarachnoid haemorrhage

BACKGROUND

Cerebral vasospasm (CV) is a feared complication which occurs after 20-40% of subarachnoid haemorrhage (SAH). It is standard practice to admit patients with SAH to intensive care for an extended period of resource-intensive monitoring. We used machine learning to predict CV requiring verapamil (CVRV) in the largest and only multi-center study to date.

METHODS

Patients with SAH admitted to UCLA from 2013 to 2022 and a validation cohort from VUMC from 2018 to 2023 were included. For each patient, 172 unique intensive care unit (ICU) variables were extracted through the primary endpoint, namely first verapamil administration or no verapamil. At each institution, a light gradient boosting machine (LightGBM) was trained using five-fold cross validation to predict the primary endpoint at various hospitalization timepoints.

FINDINGS

A total of 1750 patients were included from UCLA, 125 receiving verapamil. LightGBM achieved an area under the ROC (AUC) of 0.88 > 1 week in advance and ruled out 8% of non-verapamil patients with zero false negatives. Our models predicted "no CVRV" vs "CVRV within three days" vs "CVRV after three days" with AUCs = 0.88, 0.83, and 0.88, respectively. From VUMC, 1654 patients were included, 75 receiving verapamil. VUMC predictions averaged within 0.01 AUC points of UCLA predictions.

INTERPRETATION

We present an accurate and early predictor of CVRV using machine learning with multi-center validation. This represents a significant step towards optimized clinical management and resource allocation in patients with SAH.

FUNDING

Robert E. Freundlich is supported by National Center for Advancing Translational Sciences federal grant UL1TR002243 and National Heart, Lung, and Blood Institute federal grant K23HL148640; these funders did not play any role in this study. The National Institutes of Health supports Vanderbilt University Medical Center which indirectly supported these research efforts. Neither this study nor any other authors personally received financial support for the research presented in this manuscript. No support from pharmaceutical companies was received.

Neutrophil to Lymphocyte Ratio and Serum Biomarkers : A Potential Tool for Prediction of Clinically Relevant Cerebral Vasospasm after Aneurysmal Subarachnoid Hemorrhage

OBJECTIVE

Subarachnoid hemorrhage (SAH) is a condition characterized by bleeding in the subarachnoid space, often resulting from the rupture of a cerebral aneurysm. Delayed cerebral ischemia caused by vasospasm is a significant cause of mortality and morbidity in SAH patients, and inflammatory markers such as systemic inflammatory response index (SIRI), systemic inflammatory index (SII), neutrophil-to-lymphocyte ratio (NLR), and derived NLR (dNLR) have shown potential in predicting clinical vasospasm and outcomes in SAH patients. This article aims to investigate the relationship between inflammatory markers and cerebral vasospasm after aneurysmatic SAH (aSAH) and evaluate the predictive value of various indices, including SIRI, SII, NLR, and dNLR, in predicting clinical vasospasm.

METHODS

A retrospective analysis was performed on a cohort of 96 patients who met the inclusion criteria out of a total of 139 patients admitted Trakya University Hospital with a confirmed diagnosis of aSAH between January 2013 and December 2021. Diagnostic procedures, neurological examinations, and laboratory tests were performed to assess the patients' condition. The Student's t-test compared age variables, while the chi-square test compared categorical variables between the non-vasospasm (NVS) and vasospasm (VS) groups. Receiver operating characteristic (ROC) curve analyses were used to evaluate the diagnostic accuracy of laboratory parameters, calculating the area under the ROC curve, cut-off values, sensitivity, and specificity. A significance level of p<0.05 was considered statistically significant.

RESULTS

The study included 96 patients divided into two groups : NVS and VS. Various laboratory parameters, such as NLR, SII, and dNLR, were measured daily for 15 days, and statistically significant differences were found in NLR on 7 days, with specific cut-off values identified for each day. SII showed a significant difference on day 9, while dNLR had significant differences on days 2, 4, and 9. Graphs depicting the values of these markers for each day are provided.

CONCLUSION

Neuroinflammatory biomarkers, when used alongside radiology and scoring scales, can aid in predicting prognosis, determining severity and treatment decisions for aSAH, and further studies with larger patient groups are needed to gain more insights.

Retracted: Predictive Model of Cerebral Vasospasm in Subarachnoid Hemorrhage Based on Regression Equation

[This retracts the article DOI: 10.1155/2022/3397967.].

Systemic Inflammatory Response in Spontaneous Subarachnoid Hemorrhage from Aneurysmal Rupture versus Subarachnoid Hemorrhage of Unknown Origin

OBJECTIVE

It is well known that spontaneous non-aneurysmal subarachnoid hemorrhage (SAH), also known as sine materia SAH (smSAH), has usually a better course and prognosis than its aneurysmal counterpart (aSAH). This might depend on different inflammatory mechanisms initiated by bleeding events of different origins. The aim of the present study was to explore the systemic inflammatory response in spontaneous SAH, comparing aSAH and smSAH.

METHODS

We performed a prospective observational study over a consecutive series of patients with SAH. For these patients, we collected all clinical data and, furthermore, performed venous blood sampling over six time points to analyze blood cells. We further performed the analysis of lymphocytes and monocytes by means of flow cytometry to quantify common subtypes. Statistical analysis included a t-student test, Chi-square test, multivariate logistic regression, and ROC analysis.

RESULTS

48 patients were included: six (12.5%) with a diagnosis of spontaneous smSAH, and forty-two patients (87.5%) with aSAH. Significant differences on Day 0 were found for neutrophils and a systemic neuro-inflammatory index, namely, systemic inflammatory response index (SIRI). At the ROC analysis, neutrophil-to-lymphocyte ratio (NLR), lymphocyte-to-monocyte ratio (LMR), and SIRI exhibited satisfactory predictive power on day 0. At the multivariable logistic regression analysis, the combined index (NLR, LMR, SIRI at day 0) yielded an OR of 0.59 (95% CI 0.29-1.21]). LMR at day 0 yielded an OR of 1.25 ([95% CI 0.94-1.68]), NLR at day 0 exhibited an OR of 0.68 ([95% CI 0.42-1.09]), and SIRI at day 0 displayed an OR of 0.31 ([95% CI 0.06-1.49]).

CONCLUSION

This preliminary study indicated a possible role of some inflammatory indices that point out the importance of innate and adaptive immunity in the etiopathogenetic mechanisms. Drugs modulating these responses could eventually counteract or, at least, reduce secondary damage associated with SAH.