Triage of 5 Noncontrast Computed Tomography Markers and Spot Sign for Outcome Prediction After Intracerebral Hemorrhage

Background and Purpose— Besides the established spot sign (SS) in computed tomography angiography (CTA), there is growing evidence that different imaging markers in noncontrast CT offer great value for outcome prediction in patients with intracerebral hemorrhage (ICH). However, it is unclear how the concurrent presence of each sign independently contributes to the predictive power of poor outcome. We, therefore, aimed to clarify the predictive value of 5 recently published noncontrast CT parameters (blend sign, black hole sign, island sign, hematoma heterogeneity, and hypodensities) and the established SS in 1 consecutive series of patients with ICH. Methods— Retrospective study of patients with ICH at 2 German tertiary stroke centers; inclusion criteria were (1) spontaneous ICH and (2) noncontrast CT and CTA performed on admission within 6 hours after onset of symptoms. We defined a binary outcome (good outcome [modified Rankin Scale score of ⩽3] versus poor outcome [modified Rankin Scale score of >3]) at discharge. The predictive value of each sign was assessed in univariate and multivariable logistic regression models. Results— Of 201 patients with spontaneous ICH, 28 (13.9%) presented with black hole sign, 38 (18.9%) with blend sign, 120 (59.7%) with hypodensities, 97 (48.3%) with heterogeneous densities, 53 with island sign (26.4%), and 45 (22.4%) with SS. In univariable logistic regression, higher hematoma volume (P<0.001), intraventricular hemorrhage (P=0.002), and the presence of black hole sign/blend sign/hypodensities/island sign/SS/heterogeneous density (all P<0.001) on admission CT were associated with poor outcome. Multivariable analysis confirmed intraventricular hemorrhage (odds ratio, 2.20; P=0.025), higher hematoma volume (odds ratio, 1.02 per mL; P<0.019), the presence of hypodensities (odds ratio, 2.47; P=0.018), and SS (odds ratio, 12.22; P<0.001) as independent predictors of poor outcome. Conclusions— This study demonstrates the degree of interaction between 5 recent noncontrast CT imaging markers and SS and their individual contribution for outcome prediction in patients with ICH. Of the CT variables indicating poor outcome SS on CTA and hypodensities were the most reliable outcome predictors.

[1]  Qi Li,et al.  Black Hole Sign: Novel Imaging Marker That Predicts Hematoma Growth in Patients With Intracerebral Hemorrhage , 2016, Stroke.

[2]  Y. Fujii,et al.  Multivariate analysis of predictors of hematoma enlargement in spontaneous intracerebral hemorrhage. , 1998, Stroke.

[3]  B. Efron Estimating the Error Rate of a Prediction Rule: Improvement on Cross-Validation , 1983 .

[4]  W. Heindel,et al.  Computed Tomographic Blend Sign Is Associated With Computed Tomographic Angiography Spot Sign and Predicts Secondary Neurological Deterioration After Intracerebral Hemorrhage , 2017, Stroke.

[5]  Qi Li,et al.  Island Sign: An Imaging Predictor for Early Hematoma Expansion and Poor Outcome in Patients With Intracerebral Hemorrhage , 2017, Stroke.

[6]  D. Krieger,et al.  European Stroke Organisation (ESO) Guidelines for the Management of Spontaneous Intracerebral Hemorrhage , 2014, International journal of stroke : official journal of the International Stroke Society.

[7]  T. Steiner,et al.  Options to Restrict Hematoma Expansion After Spontaneous Intracerebral Hemorrhage , 2010, Stroke.

[8]  Andreas Charidimou,et al.  Noncontrast Computed Tomography Hypodensities Predict Poor Outcome in Intracerebral Hemorrhage Patients , 2016, Stroke.

[9]  W. Schwindt,et al.  Comparison of Spot Sign, Blend Sign and Black Hole Sign for Outcome Prediction in Patients with Intracerebral Hemorrhage , 2017, Journal of stroke.

[10]  S. Greenberg,et al.  Association Between Hypodensities Detected by Computed Tomography and Hematoma Expansion in Patients With Intracerebral Hemorrhage. , 2016, JAMA neurology.

[11]  A. Demchuk,et al.  Intracerebral Hematoma Morphologic Appearance on Noncontrast Computed Tomography Predicts Significant Hematoma Expansion , 2015, Stroke.

[12]  Qi Li,et al.  Blend Sign on Computed Tomography: Novel and Reliable Predictor for Early Hematoma Growth in Patients With Intracerebral Hemorrhage , 2015, Stroke.

[13]  Fa-jin Lv,et al.  Blend sign predicts poor outcome in patients with intracerebral hemorrhage , 2017, PloS one.

[14]  Mark W Parsons,et al.  Hematoma growth and outcomes in intracerebral hemorrhage , 2012, Neurology.

[15]  W. Heindel,et al.  Imaging-based outcome prediction in patients with intracerebral hemorrhage , 2018, Acta Neurochirurgica.

[16]  Richard Simon,et al.  Overfitting in prediction models - is it a problem only in high dimensions? , 2013, Contemporary clinical trials.

[17]  J. Broderick,et al.  Early hemorrhage growth in patients with intracerebral hemorrhage. , 1997, Stroke.

[18]  S. Mayer,et al.  Density and Shape as CT Predictors of Intracerebral Hemorrhage Growth , 2009, Stroke.

[19]  Hematoma growth is a determinant of mortality and poor outcome after intracerebral hemorrhage , 2006, Neurology.

[20]  J. Broderick,et al.  Volume of Intracerebral Hemorrhage: A Powerful and Easy‐to‐Use Predictor of 30‐Day Mortality , 1993, Stroke.