Search

Background: The objective of this study was to define clinically meaningful phenotypes of intracerebral hemorrhage (ICH) using machine learning., Methods: We used patient data from two US medical centers and the Antihypertensive Treatment of Acute Cerebral Hemorrhage-II clinical trial. We used k-prototypes to partition patient admission data. We then used silhouette method calculations and elbow method heuristics to optimize the clusters. Associations between phenotypes, complications (e.g., seizures), and functional outcomes were assessed using the Kruskal-Wallis H-test or χ 2 test., Results: There were 916 patients; the mean age was 63.8 ± 14.1 years, and 426 patients were female (46.5%). Three distinct clinical phenotypes emerged: patients with small hematomas, elevated blood pressure, and Glasgow Coma Scale scores > 12 (n = 141, 26.6%); patients with hematoma expansion and elevated international normalized ratio (n = 204, 38.4%); and patients with median hematoma volumes of 24 (interquartile range 8.2-59.5) mL, who were more frequently Black or African American, and who were likely to have intraventricular hemorrhage (n = 186, 35.0%). There were associations between clinical phenotype and seizure (P = 0.024), length of stay (P = 0.001), discharge disposition (P < 0.001), and death or disability (modified Rankin Scale scores 4-6) at 3-months' follow-up (P < 0.001). We reproduced these three clinical phenotypes of ICH in an independent cohort (n = 385) for external validation., Conclusions: Machine learning identified three phenotypes of ICH that are clinically significant, associated with patient complications, and associated with functional outcomes. Cerebellar hematomas are an additional phenotype underrepresented in our data sources., Competing Interests: Conflict of interest: The authors declare no conflicts of interest. Ethical Approval/Informed Consent: A waiver of informed consent was obtained from the institutional review boards (Northwestern University STU00213824 and Johns Hopkins University) for this research because patients were not contacted, nor was their care changed., (© 2024. Springer Science+Business Media, LLC, part of Springer Nature and Neurocritical Care Society.)

Background: Seizures are a harmful complication of acute intracerebral hemorrhage (ICH). "Early" seizures in the first week after ICH are a risk factor for deterioration, later seizures, and herniation. Ideally, seizure medications after ICH would only be administered to patients with a high likelihood to have seizures. We developed and validated machine learning (ML) models to predict early seizures after ICH., Methods: We used two large datasets to train and then validate our models in an entirely independent test set. The first model ("CAV") predicted early seizures from a subset of variables of the CAVE score (a prediction rule for later seizures)-cortical hematoma location, age less than 65 years, and hematoma volume greater than 10 mL-whereas early seizure was the dependent variable. We attempted to improve on the "CAV" model by adding anticoagulant use, antiplatelet use, Glasgow Coma Scale, international normalized ratio, and systolic blood pressure ("CAV + "). For each model we used logistic regression, lasso regression, support vector machines, boosted trees (Xgboost), and random forest models. Final model performance was reported as the area under the receiver operating characteristic curve (AUC) using receiver operating characteristic models for the test data. The setting of the study was two large academic institutions: institution 1, 634 patients; institution 2, 230 patients. There were no interventions., Results: Early seizures were predicted across the ML models by the CAV score in test data, (AUC 0.72, 95% confidence interval 0.62-0.82). The ML model that predicted early seizure better in the test data was Xgboost (AUC 0.79, 95% confidence interval 0.71-0.87, p = 0.04) compared with the CAV model AUC., Conclusions: Early seizures after ICH are predictable. Models using cortical hematoma location, age less than 65 years, and hematoma volume greater than 10 mL had a good accuracy rate, and performance improved with more independent variables. Additional methods to predict seizures could improve patient selection for monitoring and prophylactic seizure medications., (© 2022. Springer Science+Business Media, LLC, part of Springer Nature and Neurocritical Care Society.)

Background: To test the hypothesis that appearances of intracranial hematomas on diagnostic computed tomography (CT) are not idiosyncratic and reflect a biologically plausible mechanism, we evaluated the association between hematoma appearance on CT, biomarkers of platelet activity, and antiplatelet or anticoagulant medication use prior to admission., Methods: We studied 330 consecutively identified patients from 2006 to 2019. Biomarkers of platelet activity (platelet aspirin assay) and medication history (aspirin, clopidogrel) were prospectively recorded on admission. A blinded interpreter recorded the presence of hematoma appearances from the diagnostic scan. Associations were tested with parametric or nonparametric statistics, as appropriate., Results: The black hole sign (101, 30%) was most prevalent, followed by the island sign (57, 17%) and blend sign (32, 10%). There was reduced platelet activity in patients with a black hole sign (511 [430-610] vs. 562 [472-628] aspirin reaction units, P = 0.01) or island sign (505 [434-574] vs. 559 [462-629] aspirin reaction units, P = 0.004). Clopidogrel use prior to admission was associated with the black hole sign (odds ratio 2.25, 95% confidence interval 1.02-4.98, P = 0.04)., Conclusions: In patients with acute intracerebral hemorrhage, hematoma appearances on CT are associated with biomarkers of platelet activity and clopidogrel use prior to admission. Appearances of intracranial hematomas on CT may reflect reduced hemostasis from antiplatelet medication use., (© 2021. Springer Science+Business Media, LLC, part of Springer Nature and Neurocritical Care Society.)

Background: Intracranial hemorrhage (ICH) is a devastating complication for patients with ventricular assist devices (VADs). The safety of emergent anticoagulation reversal with four-factor prothrombin complex concentrate (PCC) and optimal timing of anticoagulation resumption are not clear. In addition, lactate dehydrogenase (LDH) is used as a biomarker for thromboembolic risk, but its utility in guiding anticoagulation management after reversal with PCC has not be described., Methods: We retrospectively reviewed a consecutive series of patients with VADs presenting with ICH between 2014 and 2020 who received four-factor PCC for rapid anticoagulation reversal. We collected the timing of PCC administration, timing of resumption of anticoagulation, survival, occurrence of thromboembolic events, and LDH levels throughout hospitalization., Results: We identified 16 ICH events in 14 patients with VADs treated with rapid anticoagulation reversal using four-factor PCC (11 intraparenchymal, 4 subdural, 1 subarachnoid hemorrhage). PCC was administered at a mean of 3.3 ± 0.3 h after imaging diagnosis of ICH. Overall mortality was 63%. Survivors had higher presenting Glasgow Coma Scale (median 15, interquartile range [IQR] 15-15 versus 14, IQR 8-14.7, P = 0.041). In all six instances where the patient survived, anticoagulation was resumed on average 9.16 ± 1.62 days after reversal. There were no thromboembolic events prior to resumption of anticoagulation. Three events occurred after anticoagulation resumption and within 3 months of reversal: VAD thrombosis in a patient with thrombosis at the time of reversal, ischemic stroke, and readmission for elevated LDH in the setting of subtherapeutic international normalized ratio., Conclusions: Our limited series found no thromboembolic complications immediately following anticoagulation reversal with PCC prior to resumption of anticoagulation. LDH trends may be useful to monitor thromboembolic risk after reversal., (© 2021. Springer Science+Business Media, LLC, part of Springer Nature and Neurocritical Care Society.)

Background/objective: Demonstrating a benefit of acute treatment to patients with intracerebral hemorrhage (ICH) requires identifying which patients have a potentially modifiable outcome, where treatment could favorably shift a patient's expected outcome. A decision rule for which patients have a modifiable outcome could improve the targeting of treatments. We sought to determine which patients with ICH have a modifiable outcome., Methods: Patients with ICH were prospectively identified at two institutions. Data on hematoma volumes, medication histories, and other variables of interest were collected. ICH outcomes were evaluated using the modified Rankin Scale (mRS), assessed at 14 days and 3 months after ICH, with "good outcome" defined as 0-3 (independence or better) and "poor outcome" defined as 4-6 (dependence or worse). Supervised machine learning models identified the best predictors of good versus poor outcomes at Institution 1. Models were validated using repeated fivefold cross-validation as well as testing on the entirely independent sample at Institution 2. Model fit was assessed with area under the ROC curve (AUC)., Results: Model performance at Institution 1 was strong for both 14-day (AUC of 0.79 [0.77, 0.81] for decision tree, 0.85 [0.84, 0.87] for random forest) and 3 month (AUC of 0.75 [0.73, 0.77] for decision tree, 0.82 [0.80, 0.84] for random forest) outcomes. Independent predictors of functional outcome selected by the algorithms as important included hematoma volume at hospital admission, hematoma expansion, intraventricular hemorrhage, overall ICH Score, and Glasgow Coma Scale. Hematoma expansion was the only potentially modifiable independent predictor of outcome and was compatible with "good" or "poor" outcome in a subset of patients with low hematoma volumes, good Glasgow Coma scale and premorbid modified Rankin Scale scores. Models trained on harmonized data also predicted patient outcomes well at Institution 2 using decision tree (AUC 0.69 [0.63, 0.75]) and random forests (AUC 0.78 [0.72, 0.84])., Conclusions: Patient outcomes are predictable to a high level in patients with ICH, and hematoma expansion is the sole-modifiable predictor of these outcomes across two outcome types and modeling approaches. According to decision tree analyses predicting outcome at 3 months, patients with a high Glasgow Coma Scale score, less than 44.5 mL hematoma volume at admission, and relatively low premorbid modified Rankin Score in particular have a modifiable outcome and appear to be candidates for future interventions to improve outcomes after ICH.

Background: Fever is associated with worse outcome after intracerebral hemorrhage (ICH). Autonomic dysfunction, commonly seen after brain injury, results in reduced heart rate variability (HRV). We sought to investigate whether HRV was associated with the development of fever in patients with ICH., Methods: We prospectively enrolled consecutive patients with spontaneous ICH in a single-center observational study. We included patients who presented directly to our emergency department after symptom onset, had a 10-second electrocardiogram (EKG) performed within 24 h of admission, and were in sinus rhythm. Patient temperature was recorded every 1-4 h. We defined being febrile as having a temperature of ≥ 38 °C within the first 14 days, and fever burden as the number of febrile days. HRV was defined by the standard deviation of the R-R interval (SDNN) measured on the admission EKG. Univariate associations were determined by Fisher's exact, Mann-Whitney U, or Spearman's rho correlation tests. Variables associated with fever at p ≤ 0.2 were entered in a logistic regression model of being febrile within 14 days., Results: There were 248 patients (median age 63 [54-74] years, 125 [50.4%] female, median ICH Score 1 [0-2]) who met the inclusion criteria. Febrile patients had lower HRV (median SDNN: 1.72 [1.08-3.60] vs. 2.55 [1.58-5.72] msec, p = 0.001). Lower HRV was associated with more febrile days (R = - 0.22, p < 0.001). After adjustment, lower HRV was independently associated with greater odds of fever occurrence (OR 0.92 [95% CI 0.87-0.97] with each msec increase in SDNN, p = 0.002)., Conclusions: HRV measured on 10-second EKGs is a potential early marker of parasympathetic nervous system dysfunction and is associated with subsequent fever occurrence after ICH. Detecting early parasympathetic dysfunction may afford opportunities to improve ICH outcome by targeting therapies at fever prevention.

Background: Fever is associated with worse functional outcomes after intracerebral hemorrhage (ICH); however, there are few prospective data to quantify the relationship with health-related quality of life (HRQoL). We tested the hypothesis that increased burden of fever is independently associated with decreased HRQoL at follow-up., Methods: In this prospective observational cohort study of 106 ICH patients admitted to a tertiary care hospital between 2011 and 2015, we recorded the highest core temperature each calendar day for 14 days after ICH onset. Fever burden was defined as the number of days with a fever ≥ 100.4 °F (38 °C). HRQoL outcomes were measured with Neuro-QoL domains of Cognitive Function and Mobility at 28 days, 3 months, and 1 year. Results were analyzed using mixed effects regression analysis., Results: Each additional day with a fever was independently associated with lower Mobility HRQoL (T-score - 0.9, [- 1.6 to - 0.2]; p = 0.01) and Cognitive Function HRQoL (T-score - 1.3 [- 2.0 to - 0.6]; p = 0.001) after correction for National Institutes of Health Stroke Scale score on admission, age, and time to follow-up., Conclusions: Each additional day with a fever was predictive of worse HRQoL domains of Cognitive Function and Mobility after ICH up to 1 year. These data extend previous evidence on the negative association of fever and functional outcomes to the domains of Cognitive Function and Mobility HRQoL. HRQoL outcomes may be a sensitive and powerful way to measure the efficacy of fever control in future research.

Background: Delirium symptoms are associated with later worse functional outcomes and long-term cognitive impairments, but the neuroanatomical basis for delirium symptoms in patients with acute brain injury is currently uncertain. We tested the hypothesis that hematoma location is predictive of delirium symptoms in patients with intracerebral hemorrhage, a model disease where patients are typically not sedated or bacteremic., Methods: We prospectively identified 90 patients with intracerebral hemorrhage who underwent routine twice-daily screening for delirium symptoms with a validated examination. Voxel-based lesion-symptom mapping with acute computed tomography was used to identify hematoma locations associated with delirium symptoms (N = 89)., Results: Acute delirium symptoms were predicted by hematoma of right-hemisphere subcortical white matter (superior longitudinal fasciculus) and parahippocampal gyrus. Hematoma including these locations had an odds ratio for delirium of 13 (95 % CI 3.9-43.3, P < 0.001). Disruption of large-scale brain networks that normally support attention and conscious awareness was thus associated with acute delirium symptoms., Conclusions: Higher odds ratio for delirium was increased due to hematoma location. The location of neurological injury could be of high prognostic value for predicting delirium symptoms.

Background: Clinical outcomes are typically assessed by trained staff. We tested the hypothesis that outcomes reported by the patient or a caregiver on the web would be correlated with a validated interview., Methods: We assessed surviving patients with intracerebral and subarachnoid hemorrhage at 1- , 3- , and 12-month follow-up with a validated interview for the modified Rankin Scale (mRS, a validated ordinal scale from 0, no symptoms to 5, severe disability). Health-related quality of life (HRQoL) was assessed on the web with NIH Patient-Reported Outcomes Measurement Information System (PROMIS) and Neuro-QOL using computer adaptive testing by the patient, proxy reporting by a caregiver, or proxy entry by study staff., Results: A coincident mRS and HRQoL assessment was available for 149 (71 %) of 209 patients at one, three, or 12 months. There were 89 assessments with proxy entry by study staff, 89 by the patient on the web, and 58 with proxy report by a caregiver on the web. PROMIS physical function assessments were completed in median of 4 questions, and T scores were associated with the mRS (P < 0.001), regardless of respondent. Mean T scores in every category of the mRS were different from every other category (P ≤ 0.003 for all). Results were similar for Neuro-QOL mobility., Conclusions: Web-based HRQoL assessment with NIH PROMIS and Neuro-QOL is feasible and correlated with a validated interview for the mRS. T scores distinguished between individual categories of the mRS, detecting modest differences in physical function and mobility HRQoL that are difficult to detect with the mRS. PROMIS and Neuro-QOL provide powerful and sensitive outcomes for potentially large cohorts.

Anemia and bleeding are paramount concerns in neurocritical care and often relate to the severity of intracranial hemorrhage. Anemia is generally associated with worse outcomes, and efforts to minimize anemia through reduced volume of blood sampled are encouraged. Point-of-care-testing reliably detects the use of non-steroidal anti-inflammatory drugs that may worsen bleeding and reduce platelet activity, particularly in patients with intracerebral hemorrhage. How best to monitor the effect of platelet transfusion or platelet-activating therapy is not well studied. For patients known to take novel oral anticoagulants, drug-specific coagulation tests before neurosurgical intervention are prudent.

Background: Extension of hemorrhage into the subarachnoid space in primary intracerebral hemorrhage (ICH) has recently been associated with poor outcomes, although the mechanisms underlying that association are uncertain. The objectives of this study are to confirm the association between fever and poor outcomes after ICH, and to determine whether subarachnoid hemorrhage extension (SAHE) is associated with fevers., Methods: Patients with primary ICH were enrolled into a prospective registry between December 2006 and July 2012. SAHE was identified on imaging by blinded expert reviewers. Patient temperature was recorded hourly, and we defined febrile as any recorded temperature >38 °C within the first 14 days. Regression models were developed to test whether fever was associated with poor outcome and whether the occurrence of SAHE was a predictor of fever., Results: Of the 235 patients studied, 39.7 % had SAHE and 58 % had fever. Fever was associated with higher modified Rankin scores at 3 months (odds ratio, OR 1.8 [1.04-3.12], p = 0.04) after adjustment for ICH score. SAHE was a predictor of fevers (OR 1.82 [95 % confidence interval 1.02-3.24], p = 0.04) after adjustment for ICH score, and remained significant after adjustment for other confounders like pneumonia identified in the univariate analysis., Conclusions: Our data confirm the deleterious effect of fever on the outcome of patients with ICH and show that SAHE is an independent predictor of fever after ICH. SAHE may provoke dysfunctional thermoregulation similar to what is observed after aneurysmal subarachnoid hemorrhage, creating mechanistic pathway between SAHE and poor functional outcomes.

Background: Recovery is common after subarachnoid hemorrhage (SAH), even in patients who are severely disabled at hospital discharge. Little is known about predictors of late recovery in such patients, even though such knowledge may influence treatment decisions. We hypothesized that cerebral infarction volume would be associated with 3 months outcomes in patients who are severely disabled at 14 days., Methods: We prospectively identified consecutive aneurysmal SAH patients, documented the development of cerebral infarction, and ascertained the modified Rankin Scale (mRS) at 14 days and 3 months. We included patients with mRS 4 or 5 and NIH Stroke Scale (NIHSS) at least 8 on hospital day 14 (i.e., severe neurologic impairment) and calculated infarct volume in a semi-automated fashion using CT imaging. We explored outcome determinants with ordinal regression., Results: At 14 days, 66 patients were severely disabled, 65 (98.5 %) of whom had mRS of 5; the median NIHSS was 21 [14-24]. At 3 months, 20 (32.8 %) of the 61 patients with known outcomes were independent. Larger infarction volumes were associated with death (20.4 vs. 0.85 mL, P = 0.02). In ordinal regression, increased infarct volume was associated with the worse mRS after correction for WFNS grade, age, and withdrawal of life support (OR 1.01 per mL of infarct, 95 % CI 1.01-1.03, P = 0.01)., Conclusions: After SAH, even with severe neurological injury at 14 days, good recovery is frequent and is associated with lower infarction volume. These data may help clinicians inform surrogate decision makers as they plan the future care of such severely disabled patients.

Background: Readmission within 30 days is increasingly evaluated as a measure of quality of care. There are few data on the rates of readmission after subarachnoid hemorrhage (SAH)., Objective: We sought to determine the predictors of 30-day readmission in patients with SAH., Methods: We prospectively identified 283 patients with SAH admitted between 2006 and 2012. Readmission was determined by means of an automated query with confirmation in the electronic medical record., Results: Overall, 21 (8 %) patients were readmitted for infection (n = 8), headache (n = 5), hydrocephalus (n = 4), cardiovascular causes (n = 2), medication-related complications (n = 1), and cerebral ischemia (n = 1). Readmission was associated with longer intensive care unit (ICU) length of stay (LOS) (15.4 [13.4-19.3] vs. 12.2 [8.2-18.5] days, P = 0.02), hospital LOS (22.2 [17.4-23.0] vs. 16.8 [12.0-24.1] days, P = 0.01), and placement of an external ventricular drain (EVD, OR 3.9, 95 % CI 1.3-12.0, P = 0.01). Readmission was not associated with admission neurologic grade, NIH Stroke scale at 14 days, modified Rankin scale at 3 months, history of cardiovascular disease, or radiographic cerebral infarction (P > 0.1)., Conclusions: Demographics, severity of neurologic injury, radiographic cerebral infarction, and outcomes were not associated with readmission after SAH. Markers of a more complicated hospital course (ICU and hospital LOS, EVD placement) were associated with 30-day readmission. Most readmissions were for infections acquired after discharge. Readmission within 30 days is difficult to predict, and, since the most common reason was infection acquired after discharge, it may be difficult to prevent without an integrated health system and coordinated care.

Background: Anticoagulation increases the risk of intracerebral hemorrhage (ICH), yet whether different underlying disease processes are equally affected is unknown. We tested the hypothesis that coagulopathy, measured by admission international normalized ratio (INR), disproportionately increases the risk for lobar hemorrhages., Methods: Patients with primary ICH were enrolled into a registry between December 2006 and February 2012 with prospective data acquisition and systematic follow up. Logistic regression was used to test whether lobar versus deep ICH location was independently associated with INR, and then whether INR had an influence on mortality. Spearman's correlation coefficient was used to test for an association between INR and hematoma volume separately in the lobar and deep ICH groups., Results: 221 patients were studied. Patients with lobar ICH were older (71 vs. 62 years old, p < 0.001) and more likely to have prior ICH (10 vs. 0 %, p < 0.001). INR >1.4 was observed on admission more frequently in lobar versus deep ICH (19 vs. 8 %, p = 0.02). Lobar ICH location was independently associated with INR >1.4 (OR: 2.51, 95 % CI: 1.03-6.14, p = 0.043). ICH volume correlated with INR in lobar ICH (p = 0.009), but not deep ICH (p = 0.8). Death at 1 month was independently associated with INR >1.4 (OR: 7.6, 95 % CI: 2.4-24.1, p = 0.001) after correction for the ICH Score., Conclusions: Abnormal coagulation occurs disproportionally in lobar versus deep ICH, and is associated with larger ICH volumes and higher mortality. These findings suggest a unique risk interaction between coagulopathy and underlying brain pathology due to cerebral amyloid angiopathy.

Background: In patients with acute intracerebral hemorrhage (ICH), reduced platelet activity on admission predicts hemorrhage growth and poor outcomes. We tested the hypotheses that platelet transfusion improves measured platelet activity. Further, we hypothesized that earlier treatment in patients at high risk for hemorrhage growth and poor outcome would reduce follow-up hemorrhage size and poor clinical outcomes., Methods: We prospectively identified consecutive patients with ICH who had reduced platelet activity on admission and received a platelet transfusion. We defined high-risk patients as per a previous publication, reduced platelet activity, or known anti-platelet therapy (APT) and the diagnostic CT within 12 h of symptom onset. Platelet activity was measured with the VerifyNow-ASA (Accumetrics, CA), ICH volumes on CT with computerized quantitative techniques, and functional outcomes with the modified Rankin Scale (mRS) at 3 months., Results: Forty-five patients received a platelet transfusion with an increase in platelet activity from 472 ± 50 (consistent with an aspirin effect) to 561 ± 92 aspirin reaction units (consistent with no aspirin effect, P < 0.001). For high-risk patients, platelet transfusion within 12 h of symptom onset, as opposed to >12 h, was associated with smaller follow-up hemorrhage size (8.4 [3-17.4] vs. 13.8 [12.3-62.5] ml, P = 0.04) and increased odds of independence (mRS < 4) at 3 months (11 of 20 vs. 0 of 7, P = 0.01). There were similar results for patients with known APT., Conclusions: In patients at high risk for hemorrhage growth and poor outcome, early platelet transfusion improved platelet activity assay results and was associated with smaller final hemorrhage size and more independence at 3 months.

Background: Craniotomy is potentially life-saving in selected patients with intracerebral hemorrhage (ICH). Aside from specific scenarios (cerebellar hemorrhage with hydrocephalus, midline shift from an accessible lesion, etc.) the indications for surgical decompression are controversial. Based on the earlier work that aspirin and reduced platelet activity are associated with larger hemorrhage size and hemorrhage growth, we tested the hypothesis that aspirin or reduced platelet activity would be associated with increased odds of craniotomy, likely through hemorrhage growth., Methods: We prospectively identified patients with spontaneous ICH and routinely measured platelet activity on admission. Patients were prospectively tracked, and outcomes were obtained with the modified Rankin Scale (mRS)., Results: There were 187 patients in the sample. Craniotomy (N = 32, 17%) was associated with a higher initial ICH volume (37.9 [20-63] vs. 12 [5-24] ml, P < 0.001) and location (P = 0.005). In multivariate logistic regression, after controlling for ICH volume and location, any known aspirin use (OR 3.4, 95% CI 1.1-10.4, P = 0.03), platelet activity ≤550 aspirin reaction units (OR 3.1, 95% CI 1.05-9.3, P = 0.04), or an elevated PFA-EPI closure time (OR 3.2, 95% CI 1.02-10.3, P = 0.04) were associated with increased odds of craniotomy. Craniotomy was not associated with mRS at 14 days, 28 days, or 3 months., Conclusions: After correction for ICH volume and location, aspirin use or reduced platelet activity was associated with similar increased odds for craniotomy.

Background and Purpose: In patients with subarachnoid hemorrhage (SAH), higher hemoglobin (HGB) has been associated with better outcomes, but packed red blood cell (PRBC) transfusions with worse outcomes. We performed a prospective pilot trial of goal HGB after SAH., Methods: Forty-four patients with SAH and high risk for vasospasm were randomized to goal HGB concentration of at least 10 or 11.5 g/dl. We obtained blinded clinical outcomes at 14 days (NIH Stroke Scale and modified Rankin Scale, mRS), 28 days (mRS), and 3 months (mRS), and blinded interpretation of brain MRI for cerebral infarction at 14 days. This trial is registered at www.stroketrials.org., Results: Forty-four patients were randomized. Patients with goal HGB 11.5 g/dl received more PRBC units per transfusion [1 (1-2) vs. 1 (1-1), P < 0.001] and more total PRBC units [3 (2-4) vs. 2 (1-3), P = 0.045]. Prospectively defined safety endpoints were not different between groups. HGB concentration was different between study groups from day 4 onwards. The number of cerebral infarctions on MRI (6 of 20 vs. 9 of 22), NIH Stroke Scale scores at 14 days [1 (0-9.75) vs. 2 (0-16)], and rates of independence on the mRS at 14 days (65% vs. 44%) and 28 days (80% vs. 67%) were similar, but favored higher goal HGB (P > 0.1 for all)., Conclusions: Higher goal hemoglobin in patients with SAH seems to be safe and feasible. A phase III trial of goal HGB after SAH is warranted.

Background: Although neurogenic stunned myocardium (NSM) after aneurysmal subarachnoid hemorrhage (SAH) is well described, its clinical significance remains poorly defined. We investigated the influence of left ventricular (LV) dysfunction and cerebral vasospasm on cerebral infarction, serious cardiovascular events, and functional outcome after SAH., Methods: Of the 481 patients enrolled in the University Columbia SAH Outcomes Project between 10/96 and 05/02, we analyzed a subset of 119 patients with at least one echocardiogram, serial transcranial Doppler (TCD) data, and with no prior history of cardiac disease. LV dysfunction was defined as an ejection fraction <40% on echocardiography. Infarction from vasospasm was adjudicated by the study team after comprehensive review of all clinical and imaging data. Functional outcome was assessed at 15 and 90 days with the modified Rankin Scale (mRS)., Results: Eleven percent of patients had LV dysfunction (N = 13). Younger age, hydrocephalus, and complete filling of the quadrigeminal and fourth ventricles were associated with LV dysfunction (all P < 0.05). Despite a similar frequency of pre-existing hypertension, 0% of patients with LV dysfunction reported taking antihypertensive medication, compared to 35% of those without (P = 0.009). There was a significant association between LV dysfunction and infarction from vasospasm after adjusting for clinical grade, age, and peak TCD flow velocity (P = 0.03). Patients with LV dysfunction also had higher rates of hypotension requiring vasopressors (P = 0.001) and pulmonary edema (P = 0.002). However, there was no association between LV dysfunction and outcome at 14 days after adjustment for established prognostic variables., Conclusions: LV dysfunction after SAH increases the risk of cerebral infarction from vasospasm, hypotension, and pulmonary edema, but with aggressive ICU support does not affect short-term survival or functional outcome. Antihypertensive medication may confer cardioprotection and reduce the risk of catecholamine-mediated injury after SAH.

Background: Hyponatremia is common in neurocritical care and is associated with poor outcome, but the optimal treatment is not known. We wished to test the hypothesis that for neurocritical care patients with severe hyponatremia (Na < 130 mmol/l) or hyponatremia (Na < 135 mmol/l) with depressed Glasgow Coma Scale (GCS) that conivaptan use would lead to increased serum sodium compared to usual care., Methods: We prospectively screened 249 neurocritical care patients with hyponatremia for a prospective, randomized pilot (goal N = 20) trial. Study interventions were usual care, or usual care plus conivaptan 20 mg IV as a bolus followed by 20 mg IV over 24 h, the lower FDA-approved dose. Patients were prospectively followed for changes in serum and urine electrolytes and clinical examinations with a blinded examiner. This study is registered at www.clinicaltrials.gov (NCT00727090)., Results: Despite the prevalence of hyponatremia, recruitment was difficult, and the study was terminated after six patients were enrolled, three in each group. Most hyponatremia in screened but non-randomized patients was transient or not associated with depressed GCS. Conivaptan led to higher serum sodium compared to usual care. The change in serum sodium from baseline, the pre-specified endpoint, was significantly different between groups at six (7.0 +/- 1.7 vs. -0.6 +/- 2.1 mmol/l, P = 0.008), 24 (9.7 +/- 3.2 vs. 0 +/- 1.0 mmol/l), and 36 h (8.0 +/- 5.6 vs. -1.7 +/- 2.1 mmol/l, P = 0.05). There were no apparent differences in clinical examination as a result of treatment. Adverse events were similar, and all randomized patients completed the protocol., Conclusions: Despite an inclusive protocol, most patients were not candidates for conivaptan therapy for hyponatremia. The role of conivaptan in the Neuro-ICU remains to be defined.

Background: Many ICUs have implemented protocols for tight glucose control, but there are few data on hypoglycemia and neurologic outcomes in patients with subarachnoid hemorrhage (SAH)., Methods: We prospectively ascertained 172 patients with SAH, who were treated according to a standard protocol for target glucose 80-110 mg/dl. Outcomes were assessed with the modified Rankin scale (mRS) at 14 days, 28 days, and 3 months., Results: Worse neurologic injury at admission (P < 0.001) and a history of diabetes (P = 0.002) were associated with increased glucose variance. There was lower nadir glucose in patients with radiographic cerebral infarction (81 +/- 15 vs. 87 +/- 16 mg/dl, P = 0.02), symptomatic vasospasm (78 +/- 12 vs. 84 +/- 16 mg/dl, P = 0.04) and angiographic vasospasm (79 +/- 14 vs. 86 +/- 16 mg/dl, P = 0.01), but maximum and mean glucose values were not different. Glucose < 80 mg/dl was earlier and more frequent in patients with worse functional outcome at 3 months (P < 0.001). Progressive reductions in nadir glucose were associated with increasing functional disability at 3 months (P = 0.001) after accounting for neurologic grade and mean glucose. Severe hypoglycemia (<40 mg/dl) occurred in one patient., Conclusions: In patients with SAH, nadir glucose < 80 mg/dl is associated with cerebral infarction, vasospasm, and worse functional outcomes in multivariate models. Protocols for target glucose 80-110 mg/dl effectively control hyperglycemia, but may place patients with SAH at risk for vasospasm, cerebral infarction, and poor outcome even when severe hypoglycemia does not occur.

Background: Anti-platelet medication and reduced platelet activity are associated with an increased risk of death after intracerebral hemorrhage (ICH). The optimal assay for assessing platelet activity is not defined. We hypothesized that reduced platelet activity would be common after ICH., Patients and Methods: We prospectively enrolled 72 consecutive patients with ICH and routinely measured platelet activity with both the PFA-100 (Siemens AG, Germany) and the VerifyNow-ASA (Accumetrics, CA, USA) systems on admission. We prospectively recorded anti-platelet medication use prior to ICH., Results: VerifyNow-ASA measurements were associated with aspirin (P = 0.001) and clopidogrel (P = 0.01) use prior to ICH. Combined clopidogrel and aspirin therapy was more potent than either alone. Of 33 patients with reduced platelet activity on the VerifyNow-ASA assay, 14 (42%) were not known to take anti-platelet agents. Of 27 patients with reduced platelet activity on the PFA-100, a related but different 14 (52%) were not known to take anti-platelet agents. There was a poor agreement between the assays (κ = 0.26, P = 0.07) on which patients had reduced platelet activity among the patients not known to take aspirin., Conclusions: A medication history does not reliably identify patients with reduced platelet activity after ICH, and this may explain studies that found no association between known aspirin use and outcomes. Future studies should screen for unknown use of anti-platelet medications after ICH. Neither assay perfectly identified patients who reportedly used anti-platelet medication before ICH.

Introduction: There are few predictors of acute lung injury (ALI) or the acute respiratory distress syndrome (ARDS) after subarachnoid hemorrhage (SAH). We hypothesized that cardiac troponin I, which is associated with cardiovascular morbidity, would also predict ALI., Methods: We prospectively enrolled 171 consecutive patients with SAH. Troponin was routinely measured on admission and the next day and subsequently if abnormal. We prospectively recorded the maximum troponin, in-hospital events, and clinical endpoints. ALI and ARDS were defined by standard criteria., Results: Acute lung injury was found in 10 patients (6%), ARDS in an additional 14 (8%), and pulmonary edema without lung injury in 9 (5%). Maximum troponin was different in patients without lung injury or pulmonary edema (0.03 [0.02-0.12] mcg/l), ALI (0.17 [0.04-1.4]), or ARDS (0.31 [0.9-1.8], P < 0.001). In ROC analysis, a cutoff of 0.04 mcg/l had 91% sensitivity and 42% specificity for ALI or ARDS (AUC = 0.75, P < 0.001). Troponin was associated with ALI or ARDS after accounting for neurologic grade in multivariate models without further contribution from pneumonia, packed red cell transfusion, gender, tobacco use, coronary artery disease, vasospasm, depressed ejection fraction on echocardiography, or CT grade. Lung injury was associated with worse functional outcome at 14 days, but not at 28 days or 3 months., Conclusion: Troponin I is associated with the development of ALI after SAH.

Introduction: Severity of illness scores predict all-cause mortality after intracerebral hemorrhage (ICH), but do not differentiate between proximate mechanisms or predict the timing. We hypothesized that death by neurologic criteria [brain death (BD)], withdrawal of life support, and cardiovascular death would be distinct after ICH., Methods: We prospectively enrolled patients with spontaneous ICH without underlying vascular malformation or neoplasm. We recorded clinical data and the proximate mechanism of death (BD, withdrawal of life support, cardiovascular death, or other cause). Time to death was compared with Kaplan-Meier methods (log-rank test). Data are median (IQR)., Results: Among 89 patients, 15 had withdrawal of life support, 5 had BD, 2 died from cardiac arrest, and 3 died from other causes. Among patients who died, ICH score, age, Glasgow Coma Scale, NIH Stroke Scale, and proximate cause were not associated with the proximate mechanism of death. The time to death was different (P < 0.001) depending on the proximate mechanism. Patients with BD died 1 [0-1] 1 day after ICH, withdrawal of life support led to death 5 [1-13] days after ICH, cardiac death occurred 35 [35-85] days after ICH, and other causes led to death 33 [26-33] days after ICH. Among patients where life support was withdrawn, a higher ICH score on admission was related to earlier death (P = 0.002)., Conclusions: Proximate mechanisms of death after ICH occur at distinct times. Withdrawal of life support leads to earlier death in patients with a higher severity of injury. Medical causes of death can be effectively prevented after ICH.

Introduction: Longer length of stay (LOS) is associated with higher complications and costs in ICU patients, while hospital protocols may decrease complications and LOS. We hypothesized that medical complications would increase LOS after spontaneous subarachnoid (SAH) and intracerebral (ICH) hemorrhage after accounting for severity of neurologic injury in a cohort of consecutively admitted patients., Methods: We prospectively recorded admission characteristics, hospital complications, and LOS for 122 patients with SAH and 56 patients with ICH from February 2006 through March 2008. A multidisciplinary Neuro-ICU team included a dedicated pharmacist and intensivist on daily rounds. Hospital protocols set glucose control with intravenous insulin, ventilator bundles, pharmacist involvement, and hand hygiene. Associations were explored with univariate statistics (t-tests, ANOVA, or non-parametric statistics as appropriate) and linear regression (repeated after log transformation of ICU and hospital LOS)., Results: Factors associated with longer LOS after SAH and ICH were similar. In both SAH and ICH the strongest drivers of LOS were infection, fever, and acute lung injury. For SAH, vasospasm and Glasgow Coma Scale were also significant in some models, while in patients with ICH the volume of the initial bleed was significant in some models., Conclusion: LOS after spontaneous brain hemorrhage is driven by medical complications even after the adoption of dedicated intensive care medical staff, pharmacist involvement, and evidence-based protocols for ICU care. Further alterations in care will be necessary to eliminate "preventable" complications and minimize LOS after brain hemorrhage.

Introduction: Each unit of packed red blood cells (PRBCs) is expected to raise circulating hemoglobin (HGB) by approximately 1 g/dL. There are few data on modifiers of this relationship other than gender and body mass index (BMI)., Methods: We recorded HGB before and after PRBC transfusion in a retrospective cohort of 103 patients and a prospective cohort of 93 patients with subarachnoid hemorrhage (SAH)., Results: In the retrospective cohort, 48 of 103 patients were transfused, and in the prospective cohort, 56 of 93 patients were transfused. In both groups, lower pre-transfusion HGB was associated with a larger increase in HGB (P < 0.001) after correction for the number of units of PRBCs given. In the prospective cohort, lower pre-transfusion HGB was associated with a greater rise in HGB (P < 0.001) after correction for number of units of PRBCs given, gender, and BMI in repeated measures analysis. Pre-transfusion HGB explained an additional 12% of variance in the data (P < 0.001). In both cohorts, the magnitude of the effect was similar., Conclusion: In patients with SAH, transfusion at lower HGB leads to a greater increase in HGB. Transfusion at lower HGB may be relatively more cost-effective, and this should be balanced against any potential benefit from higher HGB in SAH. One rather than 2 units of PRBCs are likely to be sufficient for most HGB targets after SAH, especially in patients with more severe anemia.

Introduction: Vasospasm is a major cause of morbidity after subarachnoid hemorrhage (SAH), and current screening techniques (angiography, transcranial Doppler [TCD], and clinical examination) have serious limitations. Brain oximetry is a promising noninvasive tool to detect reduced brain oxygenation from vasospasm., Methods: Consecutive SAH patients at high risk for vasospasm were monitored with the INVOS (Somanetics, IL, USA) 5100C cerebral oximeter. We prospectively collected oximeter readings (rO2) with concurrent values for vital signs, intracerebral pressure (ICP), arterial blood gas measurement, and hemoglobin (HGB). Data were prospectively collected every 12 h and at clinical events (angiography, transfusion, etc.). We prospectively recorded clinical history, clinical events, radiology results, and outcomes., Results: Six patients were measured 123 times. rO2 values were correlated with the contralateral side, HGB, blood pressure, and PaO2, but not with ICP or perfusion pressure. There were no measured effects of angiography or transfusion. Patterns relating rO2 readings to clinical, angiographic, or TCD evidence of vasospasm were unclear, and there were no associations with the outcome (cerebral infarction, NIH Stroke Scale, or modified Rankin Scale)., Conclusion: INVOS rO2 readings are associated with other factors that relate to cerebral oxygen delivery but seem to be of limited use as a screening tool for vasospasm or cerebral infarction after SAH.