Intracranial and Blood Pressure Variability and In-Hospital Outcomes in Intracranial Device-Monitored Patients with Spontaneous Intracerebral Hemorrhage

Predicting hematoma expansion using machine learning: An exploratory analysis of the ATACH 2 trial

INTRODUCTION

Hematoma expansion (HE) in patients with intracerebral hemorrhage (ICH) is a key predictor of poor prognosis and potentially amenable to treatment. This study aimed to build a classification model to predict HE in patients with ICH using deep learning algorithms without using advanced radiological features.

METHODS

Data from the ATACH-2 trial (Antihypertensive Treatment of Acute Cerebral Hemorrhage) was utilized. Variables included in the models were chosen as per literature consensus on salient variables associated with HE. HE was defined as increase in either >33% or 6 mL in hematoma volume in the first 24 h. Multiple machine learning algorithms were employed using iterative feature selection and outcome balancing methods. 70% of patients were used for training and 30% for internal validation. We compared the ML models to a logistic regression model and calculated AUC, accuracy, sensitivity and specificity for the internal validation models respective models.

RESULTS

Among 1000 patients included in the ATACH-2 trial, 924 had the complete parameters which were included in the analytical cohort. The median [interquartile range (IQR)] initial hematoma volume was 9.93.mm3 [5.03-18.17] and 25.2% had HE. The best performing model across all feature selection groups and sampling cohorts was using an artificial neural network (ANN) for HE in the testing cohort with AUC 0.702 [95% CI, 0.631-0.774] with 8 hidden layer nodes The traditional logistic regression yielded AUC 0.658 [95% CI, 0.641-0.675]. All other models performed with less accuracy and lower AUC. Initial hematoma volume, time to initial CT head, and initial SBP emerged as most relevant variables across all best performing models.

CONCLUSION

We developed multiple ML algorithms to predict HE with the ANN classifying the best without advanced radiographic features, although the AUC was only modestly better than other models. A larger, more heterogenous dataset is needed to further build and better generalize the models.

Comparison of two minimally invasive surgical approaches for hypertensive intracerebral hemorrhage: a study based on postoperative intracranial pressure parameters

BACKGROUND

Increased intracranial pressure (ICP) in patients with hypertensive intracerebral hemorrhage (HICH) has been associated with poor prognosis. The transsylvian insular approach (TIA) and the transcortical (TCA) approach are applied for patients with HICH. We aimed to compare the postoperative ICP parameters of TIA and TCA to identify which procedure yields better short-term outcomes in patients with basal ganglia hematoma volumes ranging from 30 to 50 mL.

METHODS

Eighty patients with basal ganglia hematomas 30-50 mL were enrolled in this study. Patients were implanted with ICP probes and divided into TIA and TCA groups according to the procedure. The ICP values were continuously recorded for five days at four-hour intervals. Short-term outcomes were evaluated using the length of hospitalization and postoperative consciousness recovery time.

RESULTS

No statistically significant differences were found in age, sex, GCS score at admission, hematoma volume, and hematoma clearance rate (p > 0.05). The results showed that postoperative initial ICP, ICP on the first postoperative day, mean ICP, DICP20 mmHg × 4 h, postoperative consciousness recovery time, the length of hospitalization, mannitol utilization rate and the mannitol dosage were lower in the TIA group than in the TCA group (p < 0.05). Postoperative consciousness recovery time was positively correlated with ICP on the first postoperative day, and the length of hospitalization was positively correlated with mean ICP.

CONCLUSIONS

TIA is more effective than TCA in improving the short-term outcomes of patients with basal ganglia hematoma volumes ranging from 30 to 50 mL according to comparisons of postoperative ICP parameters.

Galectin-3 promotes brain injury by modulating the phenotype of microglia via binding TLR-4 after intracerebral hemorrhage

BACKGROUND

Intracerebral hemorrhage (ICH) is a stroke subtype with high mortality and disability rate, and neuroinflammation is involved in secondary brain injury. Galectin-3 (Gal-3) is one of the scaffold proteins of Galectins. Studies have indicated that Gal-3 plays an important role in the physiological and pathological state of the nervous system. Here we focus on the role of Gal-3 in ICH, especially in neuroinflammation.

METHODS

Injection of autologous blood into the right basal ganglia was used to simulate ICH injury, and the level of Gal-3 in brain was regulated by related means. The changes of Gal-3 were detected by western blot and immunofluorescence, the level of neuroinflammation by immunofluorescence staining and ELISA. Apoptosis and neuron loss were detected by TUNEL staining FJB staining and Nissl staining, and neurological deficits were judged by neurobehavioral tests.

RESULTS

The protein level of Gal-3 increased at 24 h after ICH. Downregulation of Gal-3 level can reduce the infiltration of M1-type microglia and peripheral inflammatory cells, thus alleviating post-ICH neuroinflammation, and reducing cell apoptosis and neuron loss in brain tissue. ICH-induced neurological damage was rescued. Meanwhile, the promotion in the expression level of Gal-3 increased neuroinflammatory activation and nerve cell death, aggravating ICH-induced brain injury.

CONCLUSIONS

This study proves that Gal-3 is involved in neuroinflammation and nerve damage after ICH. Gal-3 expression should not be encouraged early on to prevent neuroinflammation. which provides a new possibility for clinical treatment for ICH patients.

The L-shaped correlation between systolic blood pressure and short-term and long-term mortality in patients with cerebral hemorrhage

BACKGROUND

A large amount of evidence has shown the necessity of lowering blood pressure (BP) in patients with acute cerebral hemorrhage, but whether reducing BP contributes to lower short-term and long-term mortality in these patients remains uncertain.

AIMS

We aimed to explore the association between BP, including systolic and diastolic BP, during intensive care unit (ICU) admission and 1-month and 1-year mortality after discharge of patients with cerebral hemorrhage.

METHODS

A total of 1085 patients with cerebral hemorrhage were obtained from the Medical Information Mart for Intensive Care III (MIMIC-III) database. Maximum and minimum values of systolic and diastolic BP in these patients during their ICU stay were recorded, and endpoint events were defined as the 1-month mortality and 1-year mortality after the first admission. Multivariable adjusted models were performed for the association of BP with the endpoint events.

RESULTS

We observed that patients with hypertension were likely to be older, Asian or Black and had worse health insurance and higher systolic BP than those without hypertension. The logistic regression analysis showed inverse relationships between systolic BP-min (odds ratio (OR) = 0.986, 95% CI 0.983-0.989, P < 0.001) and diastolic BP-min (OR = 0.975, 95% CI 0.968-0.981, P < 0.001) and risks of 1-month, as well as 1-year mortality when controlling for confounders including age, sex, race, insurance, heart failure, myocardial infarct, malignancy, cerebral infarction, diabetes and chronic kidney disease. Furthermore, smooth curve analysis suggested an approximate L-shaped association of systolic BP with the risk of 1-month mortality and 1-year mortality. Reducing systolic BP in the range of 100-150 mmHg has a lower death risk in these patients with cerebral hemorrhage.

CONCLUSION

We observed an L-shaped association between systolic BP levels and the risks of 1-month and 1-year mortality in patients with cerebral hemorrhage, which supported that lowering BP when treating an acute hypertensive response could reduce short-term and long-term mortality.