Aneurysmal Subarachnoid Hemorrhage Research Articles

Predictive Model for Estimating the Risk of Epilepsy After Aneurysmal Subarachnoid Hemorrhage: The RISE Score.

The occurrence of seizures after aneurysmal subarachnoid hemorrhage (aSAH) is associated with a poorer functional and cognitive prognosis and less favorable quality of life. It would be of value to promptly identify patients at risk of epilepsy to optimize follow-up protocols and design preventive strategies. Our aim was to develop a predictive score to help stratify epilepsy risk in patients with aSAH. This is a retrospective, longitudinal study of all adults with aSAH admitted to our center (2012-2021). We collected demographic data, clinical and radiologic variables, data on early-onset seizures (EOSs), and data on development of epilepsy. Exclusion criteria were previous structural brain lesion, epilepsy, and ≤7 days' follow-up. Multiple Cox regression was used to evaluate factors independently associated with unprovoked remote seizures (i.e., epilepsy). The best fitting regression model was used to develop a predictive score. Performance was evaluated in an external validation cohort of 308 patients using receiver-operating characteristic curve analysis. From an initial database of 743 patients, 419 met the inclusion criteria and were included in the analysis. The mean age was 60 ± 14 years, 269 patients (64%) were women, and 50 (11.9%) developed epilepsy within a median follow-up of 4.2 years. Premorbid modified Rankin Score (mRS) (hazard ratio [HR] 4.74 [1.8-12.4], p = 0.001), VASOGRADE score (HR 2.45 [1.4-4.2], p = 0.001), surgical treatment (HR 2.77 [1.6-4.9], p = 0.001), and presence of EOSs (HR 1.84 [1.0-3.4], p = 0.05) were independently associated with epilepsy. The proposed scale, designated RISE, scores 1 point for premorbid mRS ≥ 2 (R), VASOGRADE-Yellow (I, Ischemia), surgical intervention (S), and history of EOSs (E) and 2 points for VASOGRADE-Red. RISE stratifies patients into 3 groups: low (0-1), moderate (2-3), and high (4-5) risk (2.9%, 20.8%, and 75.7% developed epilepsy, respectively). On validation in a cohort from a different tertiary care center (N = 308), the new scale yielded a similar risk distribution and good predictive power for epilepsy within 5 years after aSAH (area under the curve [AUC] 0.82; 95% CI 0.74-0.90). The RISE scale is a robust predictor of post-SAH epilepsy with immediate clinical applicability. In addition to facilitating personalized diagnosis and treatment, RISE may be of value for exploring future antiepileptogenesis strategies.

Development and performance assessment of novel machine learning models for predicting postoperative pneumonia in aneurysmal subarachnoid hemorrhage patients: external validation in MIMIC-IV

BackgroundPostoperative pneumonia (POP) is one of the primary complications after aneurysmal subarachnoid hemorrhage (aSAH) and is associated with postoperative mortality, extended hospital stay, and increased medical fee. Early identification of pneumonia and more aggressive treatment can improve patient outcomes. We aimed to develop a model to predict POP in aSAH patients using machine learning (ML) methods.MethodsThis internal cohort study included 706 patients with aSAH undergoing intracranial aneurysm embolization or aneurysm clipping. The cohort was randomly split into a train set (80%) and a testing set (20%). Perioperative information was collected from participants to establish 6 machine learning models for predicting POP after surgical treatment. The area under the receiver operating characteristic curve (AUC), precision-recall curve were used to assess the accuracy, discriminative power, and clinical validity of the predictions. The final model was validated using an external validation set of 97 samples from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database.ResultsIn this study, 15.01% of patients in the training set and 12.06% in the testing set with POP after underwent surgery. Multivariate logistic regression analysis showed that mechanical ventilation time (MVT), Glasgow Coma Scale (GCS), Smoking history, albumin level, neutrophil-to-albumin Ratio (NAR), c-reactive protein (CRP)-to-albumin ratio (CAR) were independent predictors of POP. The logistic regression (LR) model presented significantly better predictive performance (AUC: 0.91) than other models and also performed well in the external validation set (AUC: 0.89).ConclusionA machine learning model for predicting POP in aSAH patients was successfully developed using a machine learning algorithm based on six perioperative variables, which could guide high-risk POP patients to take appropriate preventive measures.

Analysis of Cerebral Spinal Fluid Drainage and Intracranial Pressure Peaks in Patients with Subarachnoid Hemorrhage.

After aneurysmal subarachnoid hemorrhage (aSAH), elevated intracranial pressure (ICP) due to disrupted cerebrospinal fluid (CSF) dynamics is a critical concern. An external ventricular drainage (EVD) is commonly employed for management; however, optimal strategies remain debated. The randomized controlled Earlydrain trial showed that an additional prophylactic lumbar drainage (LD) after aneurysm treatment improves neurological outcome. We performed a post hoc investigation on the impact of drainage volumes and critical ICP values on patient outcomes after aSAH. Using raw patient data from Earlydrain, we analyzed CSF drainage amounts and ICP measurements in the first 8days after aSAH. Outcomes were the occurrence of secondary infarctions and the score on the modified Rankin scale after 6months, dichotomized in values of 0-2 as favorable and 3-6 as unfavorable. Repeated measurements were considered with generalized estimation equations. Earlydrain recruited 287 patients, of whom 221 received an EVD and 140 received an LD. Higher EVD volumes showed a trend to more secondary infarctions (p = 0.09), whereas higher LD volumes were associated with less secondary infarctions (p = 0.009). The mean total CSF drainage was 1052 ± 659mL and did not differ concerning infarction and neurological outcome. Maximum ICP values were higher in patients with poor outcomes but not related to drainage volumes via EVD. After adjustment for aSAH severity and total CSF drainage, higher LD volume was linked to favorable outcome (per 100mL: odds ratio 0.61 (95% confidence interval 0.39-0.95), p = 0.03), whereas higher EVD amounts were associated with unfavorable outcome (per 100mL: odds ratio 1.63 (95% confidence interval 1.05-2.54), p = 0.03). Findings indicate that effects of CSF drainage via EVD and LD differ. Higher amounts and higher proportions of LD volumes were associated with better outcomes, suggesting a potential quantity-dependent protective effect. Optimizing LD volume and mitigating ICP spikes may be a strategy to improve patient outcomes after aSAH. ClinicalTrials.gov identifier: NCT01258257.

Impact of time to start of tranexamic acid treatment on rebleed risk and outcome in aneurysmal subarachnoid hemorrhage.

The ULTRA-trial investigated effectiveness of ultra-early administration of tranexamic acid (TXA) in subarachnoid hemorrhage (SAH) and showed that TXA reduces the risk of rebleeding without concurrent improvement in clinical outcome. Previous trials in bleeding conditions, distinct from SAH, have shown that time to start of antifibrinolytic treatment influences outcome. This post-hoc analysis of the ULTRA-trial investigates whether the interval between hemorrhage and start of TXA impacts the effect of TXA on rebleeding and functional outcome following aneurysmal SAH. A post-hoc comparative analysis was conducted between aneurysmal SAH patients of the ULTRA-trial, receiving TXA and usual care to those receiving usual care only. We assessed confounders, hazard ratio (HR) of rebleeding and odds ratio (OR) of good outcome (modified Rankin Scale 0-3) at 6 months, and investigated the impact of time between hemorrhage and start of TXA on the treatment effect, stratified into time categories (0-3, 3-6 and >6 h). Sixty-four of 394 patients (16.2%) in the TXA group experienced a rebleeding, compared to 83 of 413 patients (19.9%) with usual care only (HR 0.86, 95% confidence interval (CI): 0.62-1.19). Time to start of TXA modifies the effect of TXA on rebleeding rate (p < 0.001), with a clinically non-relevant reduction observed only when TXA was initiated after 6 h (absolute rate reduction 1.4%). Tranexamic acid treatment showed no effect on good outcome (OR 0.96, 95% CI: 0.72-1.27) with no evidence of effect modification on the time to start of TXA (p = 0.53). This study suggests that the effect of TXA on rebleeding is modified by time to treatment, providing a protective, albeit clinically non-relevant, effect only when started after 6 h. No difference in functional outcome was seen. Routine TXA treatment in the aneurysmal SAH population, even within a specified time frame, is not recommended to improve functional outcome.

One-year survival of aneurysmal subarachnoid hemorrhage after airplane transatlantic transfer – a monocenter retrospective study

BackgroundAneurysmal subarachnoid hemorrhage (aSAH) is preferentially treated by prompt endovascular coiling, which is not available in Guadeloupe. Subsequently, patients are transferred to Paris, France mainland, by commercial airplane (6751 km flight) after being managed according to guidelines. This study describes the characteristics, management and outcomes related to these patients.MethodsRetrospective observational cohort study of 148 patients admitted in intensive care unit for a suspected aSAH and transferred by airplane over a 10-year period (2010–2019).ResultsThe median [interquartile range] age was 53 [45–64] years and 61% were female. On admission, Glasgow coma scale was 15 [13–15], World Federation of Neurological Surgeons (WFNS) grading scale was 1 [1–3] and Fisher scale was 4 [2–4]. External ventricular drainage and mechanical ventilation were performed prior to the flight respectively in 42% and 47% of patients. One-year mortality was 16% over the study period. By COX logistic regression analysis, acute hydrocephalus (hazard ratio [HR] 2.34, 95% confidence interval [CI] 0.98–5.58) prior to airplane transfer, WFNS grading scale on admission (HR 1.53, 95% CI 1.16–2.02) and age (OR 1.03, 95% 1.00–1.07) were associated with one-year mortality.ConclusionWhen necessary, transatlantic air transfer of patients with suspected aSAH after management according to local guidelines seems feasible and safe.

Three Pillars of Recovery After Aneurysmal Subarachnoid Hemorrhage: A Narrative Review.

Aneurysmal subarachnoid hemorrhage (aSAH) is a devastating neurologic disease with high mortality and disability. There have been global improvements in survival, which has contributed to the prevalence of patients living with long-term sequelae related to this disease. The focus of active research has traditionally centered on acute treatment to reduce mortality, but now there is a great need to study the course of short- and long-term recovery in these patients. In this narrative review, we aim to describe the core pillars in the preservation of cerebral function, prevention of complications, the recent literature studying neuroplasticity, and future directions for research to enhance recovery outcomes following aSAH.

OxyHbMeter—a novel bedside medical device for monitoring cell-free hemoglobin in the cerebrospinal fluid—proof of principle

Delayed cerebral ischemia (DCI) occurs in up to one third of patients suffering from aneurysmal subarachnoid hemorrhage (aSAH). Untreated, it leads to secondary cerebral infarctions and is frequently associated with death or severe disability. After aneurysm rupture, erythrocytes in the subarachnoid space lyse and liberate free hemoglobin (Hb), a key driver for the development of DCI. Hemoglobin in the cerebrospinal fluid (CSF-Hb) can be analyzed through a two-step procedure of centrifugation to exclude intact erythrocytes and subsequent spectrophotometric quantification. This analysis can only be done in specialized laboratories but not at the bedside in the intensive care unit. This limits the number of tests done, increases the variability of the results and restricts accuracy. Bedside measurements of CSF-Hb as a biomarker with a point of care diagnostic test system would allow for a continuous monitoring for the risk of DCI in the individual patient. In this study, a microfluidic chip was explored that allows to continuously separate blood particles from CSF or plasma based on acoustophoresis. An in vitro test bench was developed to test in-line measurements with the developed microfluidic chip and a spectrometer. The proof of principle for a continuous particle separation device has been established with diluted blood and CSF samples from animals and aSAH patients, respectively. Processing 1 mL of blood in our microfluidic device was achieved within around 70 min demonstrating only minor deviations from the gold standard centrifugation (7% average error of patient samples), while saving several hours of processing time and additionally the reduction of deviations in the results due to manual labor.

Combined Transcranial Doppler and Melatonin Levels to Predict Delayed Cerebral Ischemia After Subarachnoid Hemorrhage.

To investigate the early prediction value of transcranial Doppler ultrasound (TCD) combined with serum melatonin level for delayed cerebral ischemia (DCI) caused by subarachnoid hemorrhage (SAH). This paper is a prospective study. A total of 120 patients with SAH treated were included. The patients were divided into the DCI group (40 cases) and non-DCI group (80 cases) according to whether DCI occurred 14 days after SAH (DCI usually occurs 4 to 14d after bleeding). Baseline data, serum melatonin level, and TCD test results within 24hours after admission were compared between the 2 groups. Multivariate logistic analysis was used to analyze the factors affecting the occurrence of DCI after SAH. The value of serum melatonin level, middle cerebral artery mean blood flow velocity (MBFV) and their combination in predicting DCI in SAH patients was evaluated. Univariate analysis showed that there were statistically significant differences in the proportion of Fisher grade, Hunt-Hess grade, serum melatonin level, middle cerebral artery systolic blood flow velocity (Vs), MBFV and pulse index (PI) between the 2 groups (P<0.05). Serum melatonin levels, middle cerebral artery Vs, MBFV, and PI in the DCI group were higher than those in non-DCI group. Logistic regression (LR) analysis showed that serum melatonin level (OR=1.796, 95% CI: 1.575-4.123) and middle cerebral artery MBFV (OR=3.279, 95% CI: 2.112-4.720] were the influencing factors for DCI in SAH patients (P<0.05). Middle cerebral artery MBFV and serum melatonin levels were higher in patients with SAH complicated with DCI, and the combination of the 2 could provide a reference for early clinical prediction of DCI in patients with aneurysmal subarachnoid hemorrhage (aSAH).

Contemporary management of aneurysmal subarachnoid haemorrhage. An update for the intensivist.

Aneurysmal subarachnoid haemorrhage (aSAH) is a rare yet profoundly debilitating condition associated with high global case fatality and morbidity rates. The key determinants of functional outcome include early brain injury, rebleeding of the ruptured aneurysm and delayed cerebral ischaemia. The only effective way to reduce the risk of rebleeding is to secure the ruptured aneurysm quickly. Prompt diagnosis, transfer to specialized centers, and meticulous management in the intensive care unit (ICU) significantly improved the prognosis of aSAH. Recently, multimodality monitoring with specific interventions to correct pathophysiological imbalances has been proposed. Vigilance extends beyond intracranial concerns to encompass systemic respiratory and haemodynamic monitoring, as derangements in these systems can precipitate secondary brain damage. Challenges persist in treating aSAH patients, exacerbated by a paucity of robust clinical evidence, with many interventions showing no benefit when tested in rigorous clinical trials. Given the growing body of literature in this field and the issuance of contemporary guidelines, our objective is to furnish an updated review of essential principles of ICU management for this patient population. Our review will discuss the epidemiology, initial stabilization, treatment strategies, long-term prognostic factors, the identification and management of post-aSAH complications. We aim to offer practical clinical guidance to intensivists, grounded in current evidence and expert clinical experience, while adhering to a concise format.

Temperature Control in Acute Brain Injury: An Update.

Temperature control in severe acute brain injury (SABI) is a key component of acute management. This manuscript delves into the complex role of temperature management in SABI, encompassing conditions like traumatic brain injury (TBI), acute ischemic stroke (AIS), intracerebral hemorrhage (ICH), aneurysmal subarachnoid hemorrhage (aSAH), and hypoxemic/ischemic brain injury following cardiac arrest. Fever is a common complication in SABI and is linked to worse neurological outcomes due to increased inflammatory responses and intracranial pressure (ICP). Temperature management, particularly hypothermic temperature control (HTC), appears to mitigate these adverse effects primarily by reducing cerebral metabolic demand and dampening inflammatory pathways. However, the effectiveness of HTC varies across different SABI conditions. In the context of post-cardiac arrest, the impact of HTC on neurological outcomes has shown inconsistent results. In cases of TBI, HTC seems promising for reducing ICP, but its influence on long-term outcomes remains uncertain. For AIS, clinical trials have yet to conclusively demonstrate the benefits of HTC, despite encouraging preclinical evidence. This variability in efficacy is also observed in ICH, aSAH, bacterial meningitis, and status epilepticus. In pediatric and neonatal populations, while HTC shows significant benefits in hypoxic-ischemic encephalopathy, its effectiveness in other brain injuries is mixed. Although the theoretical basis for employing temperature control, especially HTC, is strong, the clinical outcomes differ among various SABI subtypes. The current consensus indicates that fever prevention is beneficial across the board, but the application and effectiveness of HTC are more nuanced, underscoring the need for further research to establish optimal temperature management strategies. Here we provide an overview of the clinical evidence surrounding the use of temperature control in various types of SABI.

Impact of Socioeconomic Status on Outcomes in Patients with Aneurysmal Subarachnoid Hemorrhage (P10-5.009)

Impact of Socioeconomic Status on Outcomes in Patients with Aneurysmal Subarachnoid Hemorrhage (P10-5.009)

Assessing Effects of Intrathecal Nicardipine on Cerebral Autoregulation and Cortical Activity in Aneurysmal Subarachnoid Hemorrhage (P10-5.019)

Assessing Effects of Intrathecal Nicardipine on Cerebral Autoregulation and Cortical Activity in Aneurysmal Subarachnoid Hemorrhage (P10-5.019)

Predicting Delayed Cerebral Ischemia in Patients with Aneurysmal Subarachnoid Hemorrhage: A Machine Learning Analysis (P3-2.004)

Predicting Delayed Cerebral Ischemia in Patients with Aneurysmal Subarachnoid Hemorrhage: A Machine Learning Analysis (P3-2.004)

Cisternal Score Accuracy for Cerebral Vasospasm Risk Prediction in Aneurysmal Subarachnoid Hemorrhage: A Post Hoc Analysis of a Single Center Study (P3-2.005)

Cisternal Score Accuracy for Cerebral Vasospasm Risk Prediction in Aneurysmal Subarachnoid Hemorrhage: A Post Hoc Analysis of a Single Center Study (P3-2.005)

Meningothelial Cells (MTC) of the Arachnoid Membrane Respond to Neutrophils: A Model for Delayed Cerebral Injury (DCI) After Aneurysmal Subarachnoid Hemorrhage (aSAH) (P3-2.002)

Meningothelial Cells (MTC) of the Arachnoid Membrane Respond to Neutrophils: A Model for Delayed Cerebral Injury (DCI) After Aneurysmal Subarachnoid Hemorrhage (aSAH) (P3-2.002)

Rates of operative intervention and in-hospital mortality following subarachnoid hemorrhage in patients with COVID-19.

COVID-19 had massive effects on the healthcare system and multifactorial implications for the management of intensive care unit and cerebrovascular patients. This study aimed to assess the effect of COVID-19 on the outcomes of patients with aneurysmal subarachnoid hemorrhage (SAH). The National Inpatient Sample (NIS) was used to identify patients with nontraumatic SAH (ICD-10 code I60.x). Patients with nonaneurysmal cerebrovascular malformations or traumatic intracranial injuries were excluded. Only patients managed from April to December 2020 were included in the study given the availability of an ICD-10 code for COVID-19. Data on sociodemographic factors, hospital characteristics, comorbidities, NIS SAH Severity Score (NIS-SSS), surgical treatment, and death were acquired. Multivariable analysis was used to assess predictors of both surgical intervention and in-hospital mortality. In total, 6984 patients met the study criteria, 359 (5.1%) of whom had COVID-19. Those with COVID-19 were more likely to be younger and male and had a higher All Patient Refined Diagnosis-Related Groups illness severity subclass, and NIS-SSS. Moreover, patients with COVID-19 were less likely to undergo surgery (10.0% vs 23.6%, OR 0.35, p < 0.0001) and had significantly higher mortality rates (48.2% vs 22.7%, p < 0.0001). When controlling for other variables, COVID-19 was an independent predictor of death (OR 1.67, p = 0.0002). Aneurysm surgery was performed in 1597 patients (317 open and 1280 endovascular procedures). There was no difference between the cohorts positive and negative for COVID-19 in terms of time to surgery or type of surgery. COVID-19 had significant impacts on patients with nontraumatic SAH. Specifically, patients with COVID-19 were significantly less likely to undergo surgery and had higher in-hospital mortality rates; however, for patients who did undergo procedural intervention, there was no significant difference in the type of intervention. Multiple factors, from medical acuity to healthcare system limitations, may contribute to these findings. Further retrospective research is needed to identify both specific causes of lower intervention rates and other potential nonaneurysmal causes of SAH in patients with COVID-19.

Prediction of Intracranial Pressure in Patients with an Aneurysmal Subarachnoid Hemorrhage Using Optic Nerve Sheath Diameter via Explainable Predictive Modeling.

Background: The objective of this investigation was to formulate a model for predicting intracranial pressure (ICP) by utilizing optic nerve sheath diameter (ONSD) during endovascular treatment for an aneurysmal subarachnoid hemorrhage (aSAH), incorporating explainable predictive modeling. Methods: ONSD measurements were conducted using a handheld ultrasonography device during the course of endovascular treatment (n = 126, mean age 58.82 ± 14.86 years, and female ratio 67.46%). The optimal ONSD threshold associated with an increased ICP was determined. Additionally, the association between ONSD and ICP was validated through the application of a linear regression machine learning model. The correlation between ICP and various factors was explored through the modeling. Results: With an ICP threshold set at 20 cmH2O, 82 patients manifested an increased ICP, with a corresponding ONSD of 0.545 ± 0.08 cm. Similarly, with an ICP threshold set at 25 cmH2O, 44 patients demonstrated an increased ICP, with a cutoff ONSD of 0.553 cm. Conclusions: We revealed a robust correlation between ICP and ONSD. ONSD exhibited a significant association and demonstrated potential as a predictor of ICP in patients with an ICP ≥ 25 cmH2O. The findings suggest its potential as a valuable index in clinical practice, proposing a reference value of ONSD for increased ICP in the institution.

Effect of Statin Treatment in Patients with Aneurysmal Subarachnoid Hemorrhage: A Network Meta-Analysis.

There are knowledge gaps regarding the relative efficacy of statins for aneurysmal subarachnoid hemorrhage (aSAH). This study aims to examine the comparative effectiveness and determine the ranking of different statins with network meta‑analysis in patients with aSAH. MEDLINE, Embase, Pubmed, and Cochrane Central Register of Controlled Trials were searched from database inception until December 15, 2022. Outcomes included delayed cerebral ischemia (DCI), functional recovery, and mortality. Relative risk (RRs) ratios and associated 95% confidence intervals (CIs) were estimated. The values derived from surface under the cumulative ranking curve were obtained to rank the treatment hierarchy in the analysis. We identified 13 trials involving 1,885 patients. Atorvastatin 20 mg (RR 0.68, 95% CI 0.53-0.86), pravastatin 40 mg (RR 0.51, 95% CI 0.31-0.77), and simvastatin 80 mg (RR 0.54, 95% CI 0.40-0.70) were superior to the placebo in preventing DCI. Additionally, simvastatin 80 mg (RR 0.60, 95% CI 0.42-0.84) and pravastatin 40 mg (RR 0.56, 95% CI 0.32-0.93) were associated with a decreased risk of DCI than simvastatin 40 mg. Comparisons across treatment durations suggested that short-term (RR 0.62, 95% CI 0.50-0.76) statin therapy reduced risk of DCI. Simvastatin 80 mg might be the most effective intervention in reducing DCI. Additionally, short-term therapy might provide more benefits. Further research with longer follow-up is warranted to validate the current findings in patients with aSAH who are at high risk of DCI.

Elevated cortical blood Flow Insufficiency volume as a predictor of adverse outcomes in aneurysmal subarachnoid hemorrhage: A large prospective quantitative computed tomography perfusion study

PurposeEarly hypoperfusion changes exist in patients with aneurysmal subarachnoid hemorrhage (aSAH). We aimed to investigate a readily obtainable quantitative computed tomography perfusion (CTP) parameter that could assist in quickly identifying patients at risk of delayed cerebral ischemia (DCI) and poor 90-day functional outcomes on admission. MethodsWe prospectively collected data between 2021.04 and 2022.12. Preoperative CTP data were post-processed using RAPID software. The cortical blood flow insufficiency (CBFI) was defined as Time-to-maximum > 4.0 s. Patients were categorized into four groups according to CBFI volume distribution. To minimize differences among the groups, we employed stabilized inverse probability of treatment weighting (sIPTW). The primary outcome was DCI and poor 90-day functional outcomes (modified Rankin Scale, 3–6) was the secondary outcome. Multivariable Cox or Logistic analysis were performed to estimate the association between CBFI volume and the study outcomes, both before and after sIPTW. ResultsAt baseline, the mean (SD) age of the 493 participants was 55.0 (11.8) years, and 299 (60.6%) were female. One hundred and seven participants with DCI and eighty-six participants with poor 90-day functional outcomes were identified. After sIPTW, CBFI volume demonstrated a significant association with DCI (Cox regression: Group 4 versus Group 1, HR 3.69, 95% CI 1.84–7.01) and poor 90-day functional outcomes (Logistic regression: Group 4 versus Group 1, OR 4.61, 95% CI 2.01–12.50). ConclusionIn this study, an elevated preoperative CBFI volume was associated with adverse outcomes in aSAH patients. More well-designed studies are needed to confirm this association.

396 Creating a High-Precision, Dynamic Machine Learning Model Using Time Series Data to Predict Cerebral Vasospsm in Aneurysmal Subarachnoid Hemorrhage

INTRODUCTION: Cerebral vasospasm (CV) occurs in up to one-third of patients who suffer from aneurysmal subarachnoid hemorrhage (aSAH) and can result in significant morbidity. Prior attempts to predict CV after aSAH have failed because they use static data collected on the day of presentation. We hypothesize that trending clinical and radiographic data including transcranial dopplers (TCDs) throughout 14 intensive care unit (ICU) days can be used to create a highly predictive, dynamic machine learning model for CV. METHODS: We conducted an IRB-approved retrospective study including 428 patients with aSAH hemorrhage treated at our institution between June 2014 and December 2018. For each patient, 74 features were trended including vital signs, lab values and ventriculostomy output. TCD measurements included anterior and posterior circulation velocities, Lindegaard Ratios and pulsatility indices. We employed a generative transformer-based model that uses patient representations from aggregate prior time points to predict the likelihood of CV. RESULTS: Of the 428 aSAH patients, 85 had angiographic-proven CV. There was no difference in demographic features including age, gender distribution and BMI between the CV and no-CV groups. Throughout 14 ICU days, the CV patients consistently displayed elevated mean arterial pressure and cerebral perfusion pressure. All TCD measurements were higher for the CV group. Initial logistic regression models based on static data from post-ictus days 1 through 4 showed steadily increasing F1 scores from 0.67 to 0.76. Feature analysis identified Lindegaard Ratios as the most heavily weighted features. CONCLUSIONS: Daily TCDs paired with additional longitudinal clinical data can be used to create a high-precision dynamic machine learning model for predicting CV after aSAH. These methods could also permit identification of patients at low risk of CV for reduced ICU length of stay.