Prediction of Intracerebral Haemorrhage Expansion with Clinical, Laboratory, Pharmacologic, and Noncontrast Radiographic Variables

Combining 2.5D deep learning and conventional features in a joint model for the early detection of sICH expansion

The study aims to investigate the potential of training efficient deep learning models by using 2.5D (2.5-Dimension) masks of sICH. Furthermore, it intends to evaluate and compare the predictive performance of a joint model incorporating four types of features with standalone 2.5D deep learning, radiomics, radiology, and clinical models for early expansion in sICH. A total of 254 sICH patients were enrolled retrospectively and divided into two groups according to whether the hematoma was enlarged or not. The 2.5D mask of sICH is constructed with the maximum axial, coronal and sagittal planes of the hematoma, which is used to train the deep learning model and extract deep learning features. Predictive models were built on clinic, radiology, radiomics and deep learning features separately and four type features jointly. The diagnostic performance of each model was measured using the receiver operating characteristic curve (AUC), Accuracy, Recall, F1 and decision curve analysis (DCA). The AUCs of the clinic model, radiology model, radiomics model, deep learning model, joint model, and nomogram model on the train set (training and Cross-validation) were 0.639, 0.682, 0.859, 0.807, 0.939, and 0.942, respectively, while the AUCs on the test set (external validation) were 0.680, 0.758, 0.802, 0.857, 0.929, and 0.926. Decision curve analysis showed that the joint model was superior to the other models and demonstrated good consistency between the predicted probability of early hematoma expansion and the actual occurrence probability. Our study demonstrates that the joint model is a more efficient and robust prediction model, as verified by multicenter data. This finding highlights the potential clinical utility of a multifactorial prediction model that integrates various data sources for prognostication in patients with intracerebral hemorrhage. The Critical Relevance Statement: Combining 2.5D deep learning features with clinic features, radiology markers, and radiomics signatures to establish a joint model enabling physicians to conduct better-individualized assessments the risk of early expansion of sICH.

Hybrid clinical-radiomics model based on fully automatic segmentation for predicting the early expansion of spontaneous intracerebral hemorrhage: A multi-center study

BACKGROUND

Early prediction of hematoma expansion (HE) is important for the development of therapeutic strategies for spontaneous intracerebral hemorrhage (sICH). Radiomics can help to predict early hematoma expansion in intracerebral hemorrhage. However, complex image processing procedures, especially hematoma segmentation, are time-consuming and dependent on assessor experience. We provide a fully automated hematoma segmentation method, and construct a hybrid predictive model for risk stratification of hematoma expansion.

PURPOSE

To propose an automatic approach for predicting early hemorrhage expansion after spontaneous intracerebral hemorrhage using deep-learning and radiomics methods.

METHODS

A total of 258 patients with sICH were retrospectively enrolled for model construction and internal validation, while another two cohorts (n=87, 149) were employed for independent validation. For hemorrhage segmentation, an iterative segmentation procedure was performed to delineate the area using an nnU-Net framework. Radiomics models of intra-hemorrhage and multiscale peri-hemorrhage were established and evaluated, and the best discriminative-scale peri-hemorrhage radiomics model was selected for further analysis. Combining clinical factors and intra- and peri-hemorrhage radiomics signatures, a hybrid nomogram was constructed for the early HE prediction using multivariate logistic regression. For model validation, the receiver operating characteristic (ROC) curve analyses and DeLong test were used to evaluate the performances of the constructed models, and the calibration curve and decision curve analysis were performed for clinical application.

RESULTS

Our iterative auto-segmentation model showed satisfactory results for hematoma segmentation in all four cohorts. The Dice similarity coefficient of this hematoma segmentation model reached 0.90, showing an expert-level accuracy in hematoma segmentation. The consumed time of the efficient delineation was significantly decreased, from 18 min to less than 2 min, with the assistance of the auto-segmentation model. The radiomics model of 2-mm peri-hemorrhage had a preferable area under ROC curve (AUC) of 0.840 (95 % confidence interval [CI]: 0.768, 0.912) compared with the original (0-mm dilatation) model with an AUC of 0.796 (95 % CI: 0.717, 0.875). The clinical-radiomics hybrid model showed better performances for HE prediction, with AUC of 0.853, 0.852, 0.772, and 0.818 in the training, internal validation, and independent validation cohorts 1 and 2, respectively.

CONCLUSIONS

The fully automatic clinical-radiomics model based on deep learning and radiomics exhibits a good ability for hematoma segmentation and a favorable performance in stratifying HE risks.

Research on predicting hematoma expansion in spontaneous intracerebral hemorrhage based on deep features of the VGG-19 network

PURPOSE

To construct a clinical noncontrastive computed tomography (NCCT) deep learning joint model for predicting early hematoma expansion (HE) after cerebral hemorrhage (sICH) and evaluate its predictive performance.

METHODS

All 254 patients with primary cerebral hemorrhage from January 2017 to December 2022 in the General Hospital of the Western Theater Command were included. According to the criteria of hematoma enlargement exceeding 33% or the volume exceeding 6 ml, the patients were divided into the HE group and the hematoma non-enlargement (NHE) group. Multiple models and the 10-fold cross-validation method were used to screen the most valuable features and model the probability of predicting HE. The area under the curve (AUC) was used to analyze the prediction efficiency of each model for HE.

RESULTS

They were randomly divided into a training set of 204 cases in an 8:2 ratio and 50 cases of the test set. The clinical imaging deep feature joint model (22 features) predicted the area under the curve of HE as follows: clinical Navie Bayes model AUC 0.779, traditional radiology logistic regression (LR) model AUC 0.818, deep learning LR model AUC 0.873, and clinical NCCT deep learning multilayer perceptron model AUC 0.921.

CONCLUSION

The combined clinical imaging deep learning model has a high predictive effect for early HE in sICH patients, which is helpful for clinical individualized assessment of the risk of early HE in sICH patients.

Hematoma expansion prediction based on SMOTE and XGBoost algorithm

Hematoma expansion (HE) is a high risky symptom with high rate of occurrence for patients who have undergone spontaneous intracerebral hemorrhage (ICH) after a major accident or illness. Correct prediction of the occurrence of HE in advance is critical to help the doctors to determine the next step medical treatment. Most existing studies focus only on the occurrence of HE within 6 h after the occurrence of ICH, while in reality a considerable number of patients have HE after the first 6 h but within 24 h. In this study, based on the medical doctors recommendation, we focus on prediction of the occurrence of HE within 24 h, as well as the occurrence of HE every 6 h within 24 h. Based on the demographics and computer tomography (CT) image extraction information, we used the XGBoost method to predict the occurrence of HE within 24 h. In this study, to solve the issue of highly imbalanced data set, which is a frequent case in medical data analysis, we used the SMOTE algorithm for data augmentation. To evaluate our method, we used a data set consisting of 582 patients records, and compared the results of proposed method as well as few machine learning methods. Our experiments show that XGBoost achieved the best prediction performance on the balanced dataset processed by the SMOTE algorithm with an accuracy of 0.82 and F1-score of 0.82. Moreover, our proposed method predicts the occurrence of HE within 6, 12, 18 and 24 h at the accuracy of 0.89, 0.82, 0.87 and 0.94, indicating that the HE occurrence within 24 h can be predicted accurately by the proposed method.

The Clinical Value of Neutrophil-to-Lymphocyte Ratio and Platelet-to-Lymphocyte Ratio for Predicting Hematoma Expansion and Poor Outcomes in Patients with Acute Intracerebral Hemorrhage

There is little knowledge of the effect of inflammatory markers on the prognoses of hematoma expansion (HE) in patients with intracranial hemorrhage (ICH). We evaluated the impact of neutrophil-lymphocyte ratio (NLR) and platelet-lymphocyte ratio (PLR) on HE and worse outcomes after acute ICH. This study included 520 consecutive patients with ICH from the registry database enrolled over 80 months. Patients' whole blood samples were collected upon arrival in the emergency department. Brain computed tomography scans were performed during hospitalization and repeated at 24 h and 72 h. The primary outcome measure was HE, defined as relative growth >33% or absolute growth <6 mL. A total of 520 patients were enrolled in this study. Multivariate analysis showed that NLR and PLR were associated with HE (NLR: odds ratio [OR], [95% CI] = 1.19 [1.12-1.27], p < 0.001; PLR: OR, [95% CI] = 1.01 [1.00-1.02], p = 0.04). Receiver operating characteristic curve analysis revealed that NLR and PLR could predict HE (AUC of NLR: 0.84, 95% CI [0.80-0.88], p < 0.001; AUC of PLR: 0.75 95% CI [0.70-0.80], p < 0.001). The cut-off value of NLR for predicting HE was 5.63, and that of PLR was 23.4. Higher NLR and PLR values increase HE risk in patients with ICH. NLR and PLR were reliable for predicting HE after ICH.

External validation study on the value of deep learning algorithm for the prediction of hematoma expansion from noncontrast CT scans

Background

Hematoma expansion is an independent predictor of patient outcome and mortality. The early diagnosis of hematoma expansion is crucial for selecting clinical treatment options. This study aims to explore the value of a deep learning algorithm for the prediction of hematoma expansion from non-contrast computed tomography (NCCT) scan through external validation.

Methods

102 NCCT images of hypertensive intracerebral hemorrhage (HICH) patients diagnosed in our hospital were retrospectively reviewed. The initial computed tomography (CT) scan images were evaluated by a commercial Artificial Intelligence (AI) software using deep learning algorithm and radiologists respectively to predict hematoma expansion and the corresponding sensitivity, specificity and accuracy of the two groups were calculated and compared. Comparisons were also conducted among gold standard hematoma expansion diagnosis time, AI software diagnosis time and doctors’ reading time.

Results

Among 102 HICH patients, the sensitivity, specificity, and accuracy of hematoma expansion prediction in the AI group were higher than those in the doctor group(80.0% vs 66.7%, 73.6% vs 58.3%, 75.5% vs 60.8%), with statistically significant difference (p < 0.05). The AI diagnosis time (2.8 ± 0.3 s) and the doctors’ diagnosis time (11.7 ± 0.3 s) were both significantly shorter than the gold standard diagnosis time (14.5 ± 8.8 h) (p < 0.05), AI diagnosis time was significantly shorter than that of doctors (p < 0.05).

Conclusions

Deep learning algorithm could effectively predict hematoma expansion at an early stage from the initial CT scan images of HICH patients after onset with high sensitivity and specificity and greatly shortened diagnosis time, which provides a new, accurate, easy-to-use and fast method for the early prediction of hematoma expansion.

Association between Serum Lipid and Hematoma Expansion after Spontaneous Intracerebral Hemorrhage in Chinese Patients

OBJECTIVES

Although several studies have shown that interventions to lower blood lipid concentration may reduce the risk of coronary arterial disease and ischemic stroke, the correlation between serum lipid levels and hemorrhagic stroke remains controversial. To clarify any possible association between serum lipid and hematoma expansion, we examined various serum lipid indices in patients with and without early hematoma expansion.

METHODS

Data of 572 intracerebral hemorrhage (ICH) patients from the cerebral small vessel disease cohort of Peking Union Medical College Hospital were retrospectively analyzed. Patients who finished the baseline brain computed tomography (CT) examination within 6 h post-ictus and the follow-up CT within 48 h after initial CT were included in the study. Hematoma expansion was delimited as an enlargement of hemorrhage volume over 33% or 12.5 mL between baseline and subsequent CT. Both uni- and multivariate logistic regression analyses were conducted to explore the association between early hematoma growth and various serum lipid indices, including triglycerides, total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), non-HDL-C, ratios of LDL-C/HDL-C and LDL-C/TC, as well as other demographic and clinical features.

RESULTS

Out of 157 patients included in the analysis, hematoma growth occurred in 45 (28.7%). Only higher baseline systolic blood pressure was found to be correlated with an increased risk of hematoma growth based on both univariate (odds ratio [OR] 1.014, 95% confidence interval [CI]: 1.002-1.026, P = .024) and multivariate logistic regression analyses (OR 1.022, 95%CI: 1.008-1.037, P = .003). No associations were detected between the various serum lipid indices examined and other clinical features with a likelihood of early hematoma growth between groups or within various subgroups defined by different characteristics including age, gender, baseline Glasgow Coma Scale score, systolic blood pressure, intraventricular extension, and hematoma location.

CONCLUSIONS

No association between various indices of serum lipid and hematoma growth was identified among patients and subgroups with spontaneous ICH in the Chinese population; these findings may help to guide lipid management after ICH. However, further multi-centered, larger scale studies are expected to verify our results.

Association between serum glucose level and spot sign in intracerebral hemorrhage

Hyperglycemia was proved to cause neuron death in both animal experiments and poor outcome of hemorrhage patients, but the predictive ability of admission blood glucose level for early hematoma growth in patients with intracranial hemorrhage (ICH) is still controversial. Spot sign is a well-established imaging predictor for early hematoma growth, implying active microvascular bleeding. Here, we aim to assess associations between admission serum glucose and early hematoma expansion in ICH patients, as well as spot sign.We retrospectively reviewed all the patients with ICH from January 2017 to March 2018 in West China Hospital, Sichuan University. Admission blood glucose, clinical variables, radiological characteristics, and laboratorial parameters were obtained from medical record. According to computed tomography (CT) and computed tomography angiography (CTA) scan results, hematoma expansion and spot sign were identified by 2 experienced neuroradiologists. Multivariate logistic regression analyses were employed to adjust the associations of hematoma expansion and spot sign with other clinical parameters.Around 42 patients exhibited early hematoma expansions and 26 exhibited spot signs over 138 enrolled patients. The average level of admission blood glucose was 7.55 mmol/L. Multivariate logistic regression analyses revealed that Glasgow Coma Scale (GCS) score on admission, hematoma volume, spot sign, and hyperglycemia were associated with hematoma expansion, whereas admission serum glucose and hematoma size were only associated with spot sign, respectively.Admission blood glucose level is correlated with hematoma growth and incidence of spot sign. These results indicated that hyperglycemia probably plays a critical role in the pathological process of the active bleeding. Further studies should be drawn urgently to understand the potential molecular mechanism of systemic hyperglycemia in affecting prognosis of patients with ICH.

Neutrophil to lymphocyte ratio predicts island sign in patients with intracranial hemorrhage

Our previously studies indicated that inflammatory responses are involved in the hematoma expansion (HE) after intracranial hemorrhage (ICH) ictus. Here, we aim to evaluate the correlations among the ratio of neutrophil to lymphocyte ratio (NLR), HE, and island sign in patients with ICH.Patients with spontaneous ICH were retrospectively included. Clinical characteristics, imaging features, and laboratory parameters were obtained. Multivariable analysis was performed to evaluate the association of NLR with HE or island sign. Receiver-operator analysis was also used to estimate their predictive abilities for HE and its imaging features.A total of 279 patients were enrolled in present study, and 78 patients had early hematoma growth, while 43 of them exhibited island sign. Elevation of both leukocyte (odds ratio [OR] 1.136, 95% confidence interval [CI] 1.037-1.245, P < .01) and neutrophil absolute numbers (OR 1.169, 95% CI 1.065-1.284, P < .01), as well as reduction of lymphocyte counts (OR 0.052, 95% CI 0.016-0.167, P < .01) were strongly associated with the existence of island sign. Moreover, despite the predictive ability of NLR on the existence of island sign (OR 1.063, 95% CI 1.036-1.090, P < .01), it also showed the best predictive accuracy (sensitivity 76.74%, specificity 79.66%, positive predictive value 40.70%, negative predictive value 94.90%, area under the curve 0.817) by comparing with peripheral leukocyte counts.The NLR could be used as an independently marker for reflecting the island sign in patients with ICH. Our findings indicated that systemic inflammatory responses might be involved in the pathologic process of active bleeding in cerebral.

Hyperglycemia Predicts Blend Sign in Patients with Intracerebral Hemorrhage

Background

Predictive values of admission blood glucose for early hematoma expansion in patients with intracranial hemorrhage (ICH) remain controversial. Blend sign is a novel image predictor for early hematoma growth that suggests presence of active bleeding. We investigated the association between hyperglycemia and blend sign in predicting early hematoma growth in ICH patients.

Material/Methods

All patients with intracranial hemorrhage were retrospectively reviewed. Clinical characteristics and radiological parameters were collected. Blood glucose was measured within 24 h after onset. CT scan results for hematoma expansion and blend sign were evaluated by 2 readers. Multivariate logistic regression analyses were applied to reveal the associations between hematoma growth and blend sign, as well as other variables.

Results

Out of 164 patients with ICH, 52 exhibited early hematoma growth and 18 of these were diagnosed with blend sign. Average blood glucose was 7.53 mmol/L among all patients. By using multivariate analyses, the time of CT scan baseline, GCS score, hematoma size, blend sign, and blood glucose were associated with hematoma expansion, whereas only hyperglycemia was associated with blend sign.

Conclusions

Admission hyperglycemia is associated with hematoma expansion in the presence of blend sign. These findings suggest that elevated blood glucose is a possible factor predicting continuous bleeding. Strategies to control blood glucose and ameliorate hematoma growth are urgently needed and will be investigated in our future studies.

Absolute risk and predictors of the growth of acute spontaneous intracerebral haemorrhage: a systematic review and meta-analysis of individual patient data

Background

Intracerebral haemorrhage growth is associated with poor clinical outcome and is a therapeutic target for improving outcome. We aimed to determine the absolute risk and predictors of intracerebral haemorrhage growth, develop and validate prediction models, and evaluate the added value of CT angiography.

Methods

In a systematic review of OVID MEDLINE—with additional hand-searching of relevant studies' bibliographies— from Jan 1, 1970, to Dec 31, 2015, we identified observational cohorts and randomised trials with repeat scanning protocols that included at least ten patients with acute intracerebral haemorrhage. We sought individual patient-level data from corresponding authors for patients aged 18 years or older with data available from brain imaging initially done 0·5–24 h and repeated fewer than 6 days after symptom onset, who had baseline intracerebral haemorrhage volume of less than 150 mL, and did not undergo acute treatment that might reduce intracerebral haemorrhage volume. We estimated the absolute risk and predictors of the primary outcome of intracerebral haemorrhage growth (defined as >6 mL increase in intracerebral haemorrhage volume on repeat imaging) using multivariable logistic regression models in development and validation cohorts in four subgroups of patients, using a hierarchical approach: patients not taking anticoagulant therapy at intracerebral haemorrhage onset (who constituted the largest subgroup), patients taking anticoagulant therapy at intracerebral haemorrhage onset, patients from cohorts that included at least some patients taking anticoagulant therapy at intracerebral haemorrhage onset, and patients for whom both information about anticoagulant therapy at intracerebral haemorrhage onset and spot sign on acute CT angiography were known.

Findings

Of 4191 studies identified, 77 were eligible for inclusion. Overall, 36 (47%) cohorts provided data on 5435 eligible patients. 5076 of these patients were not taking anticoagulant therapy at symptom onset (median age 67 years, IQR 56–76), of whom 1009 (20%) had intracerebral haemorrhage growth. Multivariable models of patients with data on antiplatelet therapy use, data on anticoagulant therapy use, and assessment of CT angiography spot sign at symptom onset showed that time from symptom onset to baseline imaging (odds ratio 0·50, 95% CI 0·36–0·70; p<0·0001), intracerebral haemorrhage volume on baseline imaging (7·18, 4·46–11·60; p<0·0001), antiplatelet use (1·68, 1·06–2·66; p=0·026), and anticoagulant use (3·48, 1·96–6·16; p<0·0001) were independent predictors of intracerebral haemorrhage growth (C-index 0·78, 95% CI 0·75–0·82). Addition of CT angiography spot sign (odds ratio 4·46, 95% CI 2·95–6·75; p<0·0001) to the model increased the C-index by 0·05 (95% CI 0·03–0·07).

Interpretation

In this large patient-level meta-analysis, models using four or five predictors had acceptable to good discrimination. These models could inform the location and frequency of observations on patients in clinical practice, explain treatment effects in prior randomised trials, and guide the design of future trials.

Funding

UK Medical Research Council and British Heart Foundation.

Imaging-based outcome prediction in patients with intracerebral hemorrhage

Besides the established spot sign in computed tomography angiography (CTA), recently investigated imaging predictors of hematoma growth in noncontrast computed tomography (NCCT) suggest great potential for outcome prediction in patients with intracerebral hemorrhage (ICH). Secondary hematoma growth is an appealing target for therapeutic interventions because in contrast to other determined factors, it is potentially modifiable. Even more initial therapy studies failed to demonstrate clear therapeutic benefits, there is a need for an effective patient selection using imaging markers to identify patients at risk for poor outcome and thereby tailor individual treatments for every patient. Hence, this review gives an overview about the current literature on NCCT imaging markers for neurological outcome prediction and aims to clarify the association with the established spot sign. Moreover, it demonstrates the clinical impact of these parameters and gives a roadmap for future imaging research in patients with intracerebral hemorrhage.

Higher low-density lipoprotein cholesterol levels are associated with decreased mortality in patients with intracerebral hemorrhage

BACKGROUND AND AIMS

The relationship between lipoprotein levels, low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C) and clinical outcome after intracerebral hemorrhage (ICH) remains controversial. We sought to evaluate the association of lipoprotein cholesterol levels and statin dosage with clinical and neuroimaging outcomes in patients with ICH.

METHODS

Data on consecutive patients hospitalized with spontaneous acute ICH was prospectively collected over a 5-year period and retrospectively analyzed. Demographic characteristics, clinical severity documented by NIHSS-score and ICH-score, neuroimaging parameters, pre-hospital statin use and doses, and LDL-C and HDL-C levels were recorded. Outcome events characterized were hematoma volume, hematoma expansion, in-hospital functional outcome, and in-hospital mortality.

RESULTS

A total of 672 patients with acute ICH [(mean age 61.6 ± 14.0 years, 43.6% women, median ICH score 1 (IQR: 0-2)] were evaluated. Statin pretreatment was not associated with neuroimaging or clinical outcomes. Higher LDL-C levels were associated with several markers of poor clinical outcome and in-hospital mortality. LDL-C levels were independently and negatively associated with the cubed root of hematoma volume (linear regression coefficient -0.021, 95% CI: -0.042--0.001; p = 0.049) on multiple linear regression models. Higher admission LDL-C (OR 0.88, 95% CI 0.77-0.99; p = 0.048) was also an independent predictor for decreased hematoma expansion. Higher admission LDL-C levels were independently (p < 0.001) associated with lower likelihood of in-hospital mortality (OR per 10 mg/dL increase 0.68, 95% CI: 0.57-0.80) in multivariable logistic regression models.

CONCLUSIONS

Higher LDL-C levels at hospital admission were an independent predictor for lower likelihood of hematoma expansion and decreased in-hospital mortality in patients with acute spontaneous ICH. This association requires independent confirmation.

Accuracy of Shape Irregularity and Density Heterogeneity on Noncontrast Computed Tomography for Predicting Hematoma Expansion in Spontaneous Intracerebral Hemorrhage: A Systematic Review and Meta-Analysis

OBJECTIVE

This systematic review and meta-analysis was aimed to evaluate the predictive values of shape irregularity and density heterogeneity of hematoma on noncontrast computed tomography (NCCT) for hematoma expansion (HE).

METHODS

A literature search was performed in PubMed, Embase, Scopus, Web of Science, and Cochrane Library. Studies about predictive values of shape regularity or density heterogeneity of hematoma on NCCT for HE in spontaneous intracerebral hemorrhage were included. Meta-analysis was performed to pool the data. Publication bias assessment, subgroup analysis, and univariate meta-regression were conducted.

RESULTS

A total of 7 studies with 2294 patients were included. The pooled sensitivity, specificity, positive likelihood ratio, and negative likelihood ratio of shape irregularity were 67%, 47%, 1.30, and 0.71, respectively. In contrast, the pooled sensitivity, specificity, positive likelihood ratio, and negative likelihood ratio of density irregularity were 52%, 69%, 1.70, and 0.69, respectively.

CONCLUSIONS

Considering the relatively low sensitivity and specificity, the predictive values of shape irregularity and density heterogeneity of hematoma for HE are limited. Further studies are still needed to find optimal NCCT predictors for HE in spontaneous intracerebral hemorrhage patients.

Predictors of hematoma expansion predictors after intracerebral hemorrhage

Despite years of effort, intracerebral hemorrhage (ICH) remains the most devastating form of stroke with more than 40% 30-day mortality worldwide. Hematoma expansion (HE), which occurs in one third of ICH patients, is strongly predictive of worse prognosis and potentially preventable if high-risk patients were identified in the early phase of ICH. In this review, we summarize data from recent studies on HE prediction and classify those potential indicators into four categories: clinical (severity of consciousness disturbance; blood pressure; blood glucose at and after admission); laboratory (hematologic parameters of coagulation, inflammation and microvascular integrity status), radiographic (interval time from ICH onset; baseline volume, shape and density of hematoma; intraventricular hemorrhage; especially the spot sign and modified spot sign) and integrated predictors (9-point or 24-point clinical prediction algorithm and PREDICT A/B). We discuss those predictors’ underlying pathophysiology in HE and present opportunities to develop future therapeutic strategies.

Elevated Pulse Pressure Levels Are Associated With Increased In-Hospital Mortality in Acute Spontaneous Intracerebral Hemorrhage

OBJECTIVES

Clinical outcome after intracerebral hemorrhage (ICH) remains poor. Definitive phase-3 trials in ICH have failed to demonstrate improved outcomes with intensive systolic blood pressure (SBP) lowering. We sought to determine whether other BP parameters-diastolic BP (DBP), pulse pressure (PP), and mean arterial pressure (MAP)-showed an association with clinical outcome in ICH.

METHODS

We retrospectively analyzed a prospective cohort of 672 patients with spontaneous ICH and documented demographic characteristics, stroke severity, and neuroimaging parameters. Consecutive hourly BP recordings allowed for computation of SBP, DBP, PP, and MAP. Threshold BP values that transitioned patients from survival to death were determined from ROC curves. Using in-hospital mortality as outcome, BP parameters were evaluated with multivariable logistic regression analysis.

RESULTS

Patients who died during hospitalization had higher mean PP compared to survivors (68.5 ± 16.4 mm Hg vs. 65.4 ± 12.4 mm Hg; P = 0.032). The following admission variables were associated with significantly higher in-hospital mortality (P < 0.001): poorer admission clinical condition, intraventricular hemorrhage, and increased admission normalized hematoma volume. ROC analysis showed that mean PP dichotomized at 72.17 mm Hg, provided a transition point that maximized sensitivity and specific for mortality. The association of this increased dichotomized PP with higher in-hospital mortality was maintained in multivariable logistic regression analysis (odds ratio, 3.0; 95% confidence interval, 1.7-5.3; P < 0.001) adjusting for potential confounders.

CONCLUSION

Widened PP may be an independent predictor for higher mortality in ICH. This association requires further study.

Association Between Hypodensities Detected by Computed Tomography and Hematoma Expansion in Patients With Intracerebral Hemorrhage

IMPORTANCE

Hematoma expansion is a potentially modifiable predictor of poor outcome following an acute intracerebral hemorrhage (ICH). The ability to identify patients with ICH who are likeliest to experience hematoma expansion and therefore likeliest to benefit from expansion-targeted treatments remains an unmet need. Hypodensities within an ICH detected by noncontrast computed tomography (NCCT) have been suggested as a predictor of hematoma expansion.

OBJECTIVE

To determine whether hypodense regions, irrespective of their specific patterns, are associated with hematoma expansion in patients with ICH.

DESIGN, SETTING, AND PARTICIPANTS

We analyzed a large cohort of 784 patients with ICH (the development cohort; 55.6% female), examined NCCT findings for any hypodensity, and replicated our findings on a different cohort of patients (the replication cohort; 52.7% female). Baseline and follow-up NCCT data from consecutive patients with ICH presenting to a tertiary care hospital between 1994 and 2015 were retrospectively analyzed. Data analyses were performed between December 2015 and January 2016.

MAIN OUTCOMES AND MEASURES

Hypodensities were analyzed by 2 independent blinded raters. The association between hypodensities and hematoma expansion (>6 cm3 or 33% of baseline volume) was determined by multivariable logistic regression after controlling for other variables associated with hematoma expansion in univariate analyses with P ≤ .10.

RESULTS

A total of 1029 patients were included in the analysis. In the development and replication cohorts, 222 of 784 patients (28.3%) and 99 of 245 patients (40.4%; 321 of 1029 patients [31.2%]), respectively, had NCCT scans that demonstrated hypodensities at baseline (κ = 0.87 for interrater reliability). In univariate analyses, hypodensities were associated with hematoma expansion (86 of 163 patients with hematoma expansion had hypodensities [52.8%], whereas 136 of 621 patients without hematoma expansion had hypodensities [21.9%]; P < .001). The association between hypodensities and hematoma expansion remained significant (odds ratio, 3.42 [95% CI, 2.21-5.31]; P < .001) in a multivariable model; other independent predictors of hematoma expansion were a CT angiography spot sign, a shorter time to CT, warfarin use, and older age. The independent predictive value of hypodensities was again demonstrated in the replication cohort (odds ratio, 4.37 [95% CI, 2.05-9.62]; P < .001).

CONCLUSION AND RELEVANCE

Hypodensities within an acute ICH detected on an NCCT scan may predict hematoma expansion, independent of other clinical and imaging predictors. This novel marker may help clarify the mechanism of hematoma expansion and serve as a useful addition to clinical algorithms for determining the risk of and treatment stratification for hematoma expansion.

The Accuracy of the Spot Sign and the Blend Sign for Predicting Hematoma Expansion in Patients with Spontaneous Intracerebral Hemorrhage

BACKGROUND Hematoma expansion is associated with poor outcome in intracerebral hemorrhage (ICH) patients. The spot sign and the blend sign are reliable tools for predicting hematoma expansion in ICH patients. The aim of this study was to compare the accuracy of the two signs in the prediction of hematoma expansion. MATERIAL AND METHODS Patients with spontaneous ICH were screened for the presence of the computed tomography angiography (CTA) spot sign and the non-contrast CT (NCCT) blend sign within 6 hours after onset of symptoms. The sensitivity, specificity, and positive and negative predictive values of the spot sign and the blend sign in predicting hematoma expansion were calculated. The accuracy of the spot sign and the blend sign in predicting hematoma expansion was analyzed by receiver-operator analysis. RESULTS A total of 115 patients were enrolled in this study. The spot sign was observed in 25 (21.74%) patients, whereas the blend sign was observed in 22 (19.13%) patients. Of the 28 patients with hematoma expansion, the CTA spot sign was found on admission CT scans in 16 (57.14%) and the NCCT blend sign in 12 (42.86%), respectively. The sensitivity, specificity, positive predictive value, and negative predictive value of the spot sign for predicting hematoma expansion were 57.14%, 89.66%, 64.00%, and 86.67%, respectively. In contrast, the sensitivity, specificity, positive predictive value, and negative predictive value of the blend sign were 42.86%, 88.51%, 54.55%, and 82.80%, respectively. The area under the curve (AUC) of the spot sign was 0.734, which was higher than that of the blend sign (0.657). CONCLUSIONS Both the spot sign and the blend sign seemed to be good predictors for hematoma expansion, and the spot sign appeared to have better predictive accuracy.

Noncontrast Computed Tomography Hypodensities Predict Poor Outcome in Intracerebral Hemorrhage Patients

BACKGROUND AND PURPOSE

Noncontrast computed tomographic (CT) hypodensities have been shown to be associated with hematoma expansion in intracerebral hemorrhage (ICH), but their impact on functional outcome is yet to be determined. We evaluated whether baseline noncontrast CT hypodensities are associated with poor clinical outcome.

METHODS

We performed a retrospective review of a prospectively collected cohort of consecutive patients with primary ICH presenting to a single academic medical center between 1994 and 2016. The presence of CT hypodensities was assessed by 2 independent raters on the baseline CT. Unfavorable outcome was defined as a modified Rankin score >3 at 90 days. The associations between CT hypodensities and unfavorable outcome were investigated using uni- and multivariable logistic regression models.

RESULTS

During the study period, 1342 patients presented with ICH and 800 met restrictive inclusion criteria (baseline CT available for review, and 90-day outcome available). Three hundred and four (38%) patients showed hypodensities on CT, and 520 (65%) patients experienced unfavorable outcome. In univariate analysis, patients with unfavorable outcome were more likely to demonstrate hypodensities (48% versus 20%; P<0.0001). After adjustment for age, admission Glasgow coma scale, warfarin use, intraventricular hemorrhage, baseline ICH volume, and location, CT hypodensities were found to be independently associated with an increase in the odds of unfavorable outcome (odds ratio 1.70, 95% confidence interval [1.10-2.65]; P=0.018).

CONCLUSIONS

The presence of noncontract CT hypodensities at baseline independently predicts poor outcome and comes as a useful and widely available addition to our ability to predict ICH patients' clinical evolution.