Abstract
Background and purpose
non-contrast CT hypodensities (HD) represent an alternative to the CT angiography spot sign (SS) to predict intracerebral hemorrhage (ICH) expansion. However, previous studies suggested that these markers predicted hematoma expansion independently from each other. We investigated whether the integration of SS and HD improved the stratification of ICH expansion risk.
Methods
a single center cohort of consecutive ICH patients was retrospectively analyzed. Patients with available CT angiography, baseline and follow up non-contrast CT images available were included. Trained readers reviewed all the images for SS and HD presence and the study population was classified into four groups: SS and HD negative (SS−HD−), SS positive only (SS+HD−), HD positive only (SS−HD+) and SS and HD positive (SS+HD+). ICH expansion was defined as hematoma growth >33% or>6mL. The association between SS and HD presence and ICH expansion was investigated with multivariable logistic regression.
Results
745 subjects qualified for the analysis (median age 73, 54.1% males). The rates of ICH expansion were: 9.3% in SS−HD−, 25.8% in SS+HD−, 27.4% in SS−HD+, and 55.6% in SS+HD+ patients (p<0.001). After adjustment for potential confounders and keeping SS−HD− subjects as reference, the risk of ICH expansion was increased in SS+HD− and SS−HD+ patients (odds ratio, OR 2.93, p=0.002 and OR 3.02, p<0.001 respectively). SS+HD+ subjects had the highest risk of hematoma growth (OR 9.50, p<0.001).
Conclusions
integration of SS and HD improves the stratification of hematoma growth risk and may help the selection of ICH patients for anti-expansion treatment in clinical trials.
Keywords: stroke, intracerebral hemorrhage, hematoma expansion, hypodensities, spot sign
INTRODUCTION
Intracerebral hemorrhage (ICH) is the deadliest type of stroke and accounts for 10-20% of all cerebrovascular events (1). Up to one third of ICH patients experience hematoma growth in the first hours after stroke onset. Being potentially preventable and independently associated with unfavorable prognosis, ICH expansion represents an appealing target for acute ICH treatment (2). Accurate stratification of ICH expansion risk is crucial to identify those patients at highest risk of clinical deterioration because of active bleeding and therefore most likely to benefit from anti-expansion therapies. The CT angiography (CTA) spot sign (SS) is a robust imaging predictor of hematoma growth (3,4) but CTA is not routinely performed in many settings. Non-contrast CT (NCCT) hypodensities (HD) have been described and validated to be a reliable alternative to SS for the prediction of hematoma expansion (5). It is notable, however, that these two markers predict hematoma growth independently (6). Therefore, a combined analysis of SS and HD may provide additional yield in the identification of ICH patients at high risk of active bleeding. We investigated whether the integration of SS and HD improved the stratification of ICH expansion risk. In particular, we tested the hypothesis that patients showing evidence of both SS and HD are at highest risk of hematoma expansion.
METHODS
Because of the sensitive nature of the data collected for this study, requests to access the dataset from qualified researchers trained in human subject confidentiality protocols may be sent to the corresponding author.
Patient selection
All aspects of this study received Institutional Review Board approval. Informed consent was either obtained by patients, family members, or waived by the Institutional Review Board. Patients were retrospectively selected from a cohort of consecutive ICH patients admitted at a single academic urban medical center between 1994 and 2015 (6,7). We included patients with: 1) primary ICH diagnosed on NCCT, 2) availability of baseline and follow-up NCCT images performed within 48h from onset 3) availability of CTA images. We applied the following exclusion criteria: 1) post-traumatic intracranial hemorrhage, 2) secondary cause of the hemorrhage such as vascular or neoplastic intracranial lesions 3) evidence of ischemic stroke with hemorrhagic transformation; 4) primary intraventricular hemorrhage (IVH) 5) surgical hematoma evacuation before follow-up NCCT, 6) missing clinical and demographic data.
Clinical Variables
Clinical and demographic data was acquired through patient or family members’ interviews or with retrospective review of hospital charts. The following variables were collected: medical history of hypertension, diabetes mellitus, antithrombotic treatment (antiplatelet therapy, oral anticoagulant treatment), statin treatment, admission systolic and diastolic blood pressure (SBP and DBP respectively), Glasgow Coma Scale score (GCS), time from symptom onset to baseline NCCT. BP was managed according to the American Heart Association/American Stroke Association guidelines (8).
Image acquisition
Baseline and follow-up NCCT scans were obtained with an axial technique (5-mm slice thickness reconstruction), with the following acquisition parameters: tube potential range 120-140 kVp, tube current range 100–500 mA. Baseline and follow-up ICH volumes on NCCT images were calculated with a semi-automated computer assisted technique (Analyze Direct 11.0 software). All patients underwent a follow-up NCCT scan at 24 hours from onset or earlier in case of clinical deterioration.
First pass CTA images were obtained scanning from the base of the skull to the vertex using an axial technique with the following acquisition parameters: 0.5 pitch, 1.25-mm collimation, 100 to 140 kV(p), with a tube current ranging from 80 to 630 mA. Intravenous iodinated contrast material was administered by a power injector at 4 to 5 mL/s with semiautomatic contrast bolus triggering technique (9).
Images Analysis
Two raters (AM, Stroke Neurologist and GB, Neuroradiologist) blinded to clinical data and results of follow-up NCCT reviewed all the CTA and NCCT images for the presence of SS and HD, according to previously described criteria (5,6,9). All discrepancies in SS and HD detection were resolved with joint analysis of images to reach a consensus and make a final decision. Briefly, the SS was defined as presence of at least one focus of contrast pooling within the hemorrhage, with an attenuation ≥120 Hounsfield Units and lack of connection with normal or abnormal vessels surrounding the hemorrhage (9). HD were defined as presence of any hypodense region inside the hematoma having any morphology and size, disconnected from the surrounding brain parenchyma.
The study population was classified into four groups according to the presence of the two markers of hematoma expansion: SS and HD negative (SS−HD−); SS positive and HD negative (SS+HD−); SS negative and HD positive (SS−HD+) and SS positive and HD positive (SS+HD+). An illustrative example of SS+HD+ subject is provided in Figure I in the online-only Data Supplement.
The main outcome of our analysis was hematoma expansion defined as absolute hemorrhage growth > 6 mL or relative hemorrhage growth >33% from baseline ICH volume (10). In a secondary analysis a different definition of hematoma expansion was used (absolute hemorrhage growth > 12.5 mL or relative hemorrhage growth > 33%) (10).
Statistical Analyses
Continuous variables were expressed as median (interquartile range, IQR) or mean (standard deviation, SD) as appropriate based on their distribution evaluated with the Shapiro-Wilk test. Categorical variables were summarized as count (percentage). Continuous variables with normal and non-normal distribution were compared using the t test/anova and Mann-Whitney/Kruskal-Wallis test respectively. Differences in categorical variables were explored with χ2 test. The inter-rater reliability for SS and HD identification was assessed with Cohen’s κ statistic. A multivariable binary logistic regression model was used to investigate the association between the imaging markers of interest and ICH expansion. We adjusted all the regression models for predictors of hematoma expansion already known from the literature such as baseline ICH volume, anticoagulant treatment and time from onset to baseline NCCT (model 1) (11) and also for variables with a p value < 0.1 in univariate analysis (model 2). We calculated every single patients’ predicted probability of ICH growth using individual data combined with the binary logistic regression model estimates. The predicted probability of ICH expansion was expressed as a continuous variable ranging from 0 to 1. Finally we calculated the sensitivity, specificity, positive predictive value, negative predictive value and accuracy for ICH expansion for different SS and HD combinations. P values < 0.05 were considered statistically significant and the statistical package SPSS version 21, 2012 (www.spss.com) was used for all the analyses.
Results
A total of 745 subjects were included in the analysis (median age 73, 54.1 % males) of whom 148 (19.9%) had significant hematoma expansion. The main reasons for exclusion from the study were: CTA images not available (n = 315), missing baseline/follow up NCCT scan or poor quality images (n = 251) and NCCT scans obtained after 48 hours from symptom onset (n = 30). The clinical and demographic characteristics of the selected study population are summarized in table 1. Compared with the included subjects, excluded patients had smaller ICH volumes but a higher mortality at 90 days, likely because of the observed higher rate of warfarin treatment, hematoma expansion, IVH and infratentorial bleedings. Excluded patients were also less likely to be on antiplatelet treatment. All the remaining demographic, clinical and imaging characteristics were similar (all p > 0.20). The comparison between included vs excluded patients is summarized in table I in the online-only Data Supplement.
Table 1.
Population characteristics (n = 745)
| Variable | |
|---|---|
| Age, median (IQR), y | 73 (62-81) |
| Sex, male, n (%) | 403 (54.1) |
| History of hypertension, n (%) | 585 (78.5) |
| History of diabetes, n (%) | 157 (21.1) |
| Antiplatelet treatment, n (%) | 237 (31.8) |
| Anticoagulant treatment, n (%) | 105 (14.1) |
| SBP, mean (SD), mmHg | 176 (34) |
| DBP, mean (SD), mmHg | 92 (23) |
| GCS, median (IQR) | 14 (9-15) |
| Time from onset to NCCT | |
| Median (IQR), h | 4.7 (2.0-7.9) |
| ≤ 6h, n (%) | 379 (50.9) |
| > 6h, n (%) | 199 (26.7) |
| Unknown, n (%) | 167 (22.4) |
| Baseline ICH volume, median (IQR), mL | 17 (6-38) |
| Presence of IVH, n (%) | 305 (40.9) |
| ICH location | |
| Lobar, n (%) | 344 (46.2) |
| Deep, n (%) | 348 (46.7) |
| Infratentorial, n (%) | 53 (7.1) |
| SS+, n (%) | 161 (21.6) |
| HD+, n (%) | 223 (29.9) |
| SS and HD combined | |
| SS− HD-, n (%) | 460 (61.7) |
| SS+ HD-, n (%) | 62 (8.3) |
| SS− HD+, n (%) | 124 (16.6) |
| SS+ HD+, n (%) | 99 (13.3) |
| ICH expansion, n (%) | 148 (19.9) |
| Mortality at 90 days, n (%) | 214 (28.7) |
IQR indicates interquartile range; ICH, intracerebral hemorrhage; SD, standard deviation; SBP, systolic blood pressure; DBP diastolic blood pressure; SS, spot sign; HD, hypodensities; NCCT, non contrast CT scan; GCS, Glasgow Coma Scale; IVH, intraventricular hemorrhage.
ICH expansion was defined as hematoma growth>33% or >6 mL.
We have also compared the prevalence of HD and ICH expansion in patients with CTA images available vs patients who did not receive a CTA and observed no significant differences (HD frequency 29.9 vs 32.3%, p = 0.450 and ICH expansion frequency 20.0% vs 21.8%, p = 0.530 in patients with vs without CTA respectively).
The inter-rater agreement for SS and HD identification was good (κ=0.83, 95% confidence interval 0.76-0-89 and κ=0.87, 95% confidence interval 0.87-0-97for SS and HD respectively).
Table 2 shows the clinical and imaging differences between the four study groups. SS+HD+ patients were more severely affected, as highlighted by lower admission GCS, larger ICH volumes and higher rate of IVH and mortality at three months. Individuals in this group had a shorter time from onset to baseline NCCT and were also more likely to be on warfarin treatment. The hematoma expansion rate was significantly higher in SS+HD− (25.8%) or SS−HD+ (27.4%) when compared with SS−HD− (9.3%). SS+HD+ showed the highest rate of hematoma expansion (51.5%).
Table 2.
Univariate analysis stratified by spot sign (SS) and hypodensity (HD) presence
| SS−
HD− (n=460) |
SS+
HD− (n=62) |
SS−
HD+ (n=124) |
SS+
HD+ (n=99) |
p | ||
|---|---|---|---|---|---|---|
|
| ||||||
| Age, median (IQR), y | 74 (64–82) | 72 (58–80) | 71 (59–80) | 71 (61–81) | 0.127 | |
|
| ||||||
| Sex, male, n (%) | 237 (51.5) | 36 (58.1) | 69 (55.6) | 61 (61.6) | 0.262 | |
|
| ||||||
| History of hypertension, n (%) | 362 (78.7) | 52 (83.9) | 94 (75.8) | 77 (77.8) | 0.652 | |
|
| ||||||
| History of diabetes, n (%) | 100 (21.7) | 14 (22.6) | 26 (21.0) | 17 (17.2) | 0.774 | |
|
| ||||||
| Antiplatelet treatment, n (%) | 160 (34.8) | 23 (37.1) | 26 (21.0) | 28 (28.3) | 0.019 | |
|
| ||||||
| Anticoagulant treatment, n (%) | 41 (8.9) | 12 (19.4) | 25 (20.2) | 27 (27.3) | <0.001 | |
|
| ||||||
| SBP, mean (SD), mmHg | 177 (34) | 178 (34) | 176 (30) | 179 (39) | 0.791 | |
|
| ||||||
| DBP, mean (SD), mmHg | 93 (23) | 96 (29) | 94 (24) | 95 (22) | 0.929 | |
|
| ||||||
| GCS, median (IQR) | 15 (11–15) | 14 (8–15) | 13 (8–15) | 11 (6–15) | <0.001 | |
|
| ||||||
| Time from onset to NCCT | ||||||
|
| ||||||
| Median (IQR), h | 5.7 (3.8–10.1) | 2.3 (1.5–5.0) | 2.6 (1.1–5.1) | 2.2 (1.1–4.9) | <0.001 | |
|
|
||||||
| ≤ 6h, n (%) | 192 (41.7) | 37 (59.7) | 80 (64.5) | 70 (70.7) | <0.001 | |
|
|
||||||
| > 6h, n (%) | 157 (34.1) | 9 (14.5) | 20 (16.1) | 13 (13.1) | ||
|
|
||||||
| Unknown, n (%) | 111 (24.1) | 16 (25.8) | 24 (19.4) | 16 (16.2) | ||
|
|
||||||
| Baseline ICH volume, median (IQR), mL | 11 (4–25) | 22 (10–40) | 31 (16–60) | 39 (22–70) | <0.001 | |
|
|
||||||
| ICH location | ||||||
|
|
||||||
| Lobar, n (%) | 204 (44.3) | 27 (43.5) | 62 (50.0) | 51 (51.5) | 0.108 | |
|
|
||||||
| Deep, n (%) | 214 (46.5) | 30 (48.4) | 58 (46.8) | 46 (46.5) | ||
|
|
||||||
| Infratentorial, n (%) | 42 (9.1) | 5 (8.1) | 4 (3.2) | 2 (2.0) | ||
|
|
||||||
| Presence of IVH, n (%) | 168 (36.5) | 32 (51.6) | 56 (45.2) | 49 (49.5) | 0.014 | |
|
|
||||||
| ICH expansion, n (%) | 43 (9.3) | 16 (25.8) | 34 (27.4) | 55 (55.6) | <0.001 | |
|
|
||||||
| Mortality at 90 days, n (%) | 90 (19.6) | 28 (45.2) | 45 (36.6) | 51 (51.5) | <0.001 | |
IQR indicates interquartile range; ICH, intracerebral hemorrhage; SD, standard deviation; SBP, systolic blood pressure; DBP diastolic blood pressure; SS, spot sign; HD, hypodensities; NCCT, non contrast CT scan; GCS, Glasgow Coma Scale; IVH, intraventricular hemorrhage. ICH expansion was defined as hematoma growth>33% or >6 mL.
Table 3 shows the results of the multivariable logistic regression analysis for ICH expansion. After adjustment for known predictors of hematoma expansion, the presence of SS or HD was independently associated with higher odds of ICH expansion (odds ratio (OR) 2.93, p = 0.002 for SS+HD− and OR 3.02, p<0.001 for SS−HD+) when compared to patients with neither of these markers. Patients with combined presence of SS and HD had the highest risk of ICH expansion (OR 9.50, p < 001 for SS+HD+). Figure 1 shows the predicted probability of ICH expansion stratified by SS and HD status.
Table 3.
Multivariable Logistic Regression for ICH expansion
| Model 1* | Model 2** | |||
|---|---|---|---|---|
| OR (95% CI) | p | OR (95% CI) | p | |
| SS− HD− | Reference | Reference | ||
| SS+ HD− | 2.93 (1.50–5,71) | 0.002 | 2.99 (1.53–5.85) | 0.001 |
| SS− HD+ | 3.02 (1.75–5.20) | <0.001 | 3.01 (1.74–5.21) | <0.001 |
| SS+ HD+ | 9.50 (5.38–16.78) | <0.001 | 9.63 (5.43–17.07) | <0.001 |
Adjusted for baseline ICH volume, time from onset to NCCT, anticoagulant treatment.
Adjusted for baseline ICH volume, time from onset to NCCT, anticoagulant treatment, antiplatelet treatment, GCS, presence of IVH.
ICH indicates intracerebral hemorrhage; SS, spot sign; HD, hypodensities; OR, odds ratio; CI, confidence interval; NCCT, non contrast CT scan; GCS, Glasgow Coma Scale; IVH, intraventricular hemorrhage. ICH expansion was defined as hematoma growth >33% or >6 mL.
Figure 1.
Predicted probability of ICH expansion.
SS indicates spot sign; HD, hypodensities. P value for trend was calculated with the Jonckheere Terpstra test. ICH expansion was defined as hematoma growth>33% or >6 mL.
These results remained significant when other potential confounders were included in the multivariable logistic regression analysis (table 3, model 2). We also confirmed all these findings in multiple sensitivity analyses: 1) using a different definition of ICH expansion (hematoma growth > 33% or > 12.5 mL), 2) excluding warfarin-associated hemorrhages, 3) restricting the logistic regression analysis to early presenters (time from onset to baseline NCCT ≤ 6h), 4) excluding subjects with unknown time from onset to baseline NCCT and 5) adjusting for hypertension, admission SBP and DBP in the multivariable logistic regression model.
Table 4 shows the test characteristics of SS and HD alone compared with potential combinations of the two markers.
Table 4.
Test Characteristics for hematoma expansion
| SS and HD analyzed separately | SS and HD combinations | p | |||
|---|---|---|---|---|---|
| SS+ | HD+ | SS+ and HD+ | SS+ and/or HD+ | ||
| Sensitivity (95% CI) | 0.48 (0.40–0.56) | 0.60 (0.52–0.68) | 0.37 (0.30–0.46) | 0.71 (0.63–0.78) | <0.001 |
| Specificity (95% CI) | 0.85 (0.82–0.88) | 0.78 (0.74–0.81) | 0.93 (0.90–0.95) | 0.70 (0.66–0.73) | <0.001 |
| PPV (95% CI) | 0.44 (0.40–0.50) | 0.40 (0.35–0.45) | 0.56 (0.47–0.64) | 0.37 (0.33–0.41) | 0.010 |
| NPV (95% CI) | 0.87 (0.85–0.89) | 0.89 (0.87–0.91) | 0.86 (0.84–0.87) | 0.91 (0.88–0.93) | 0.064 |
| Accuracy | 0.78 | 0.75 | 0.82 | 0.71 | <0.001 |
SS indicates spot sign; HD, hypodensities; PPV, positive predictive value; NPV, negative predictive value. Comparison of the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy percentages between the four groups was performed by using the χ 2 test. ICH expansion was defined as hematoma growth >33% or >6 mL.
DISCUSSION
The main finding of this retrospective analysis is that the integration of SS and HD provides additional yield in the stratification of ICH expansion risk. In particular, patients with combined presence of SS and HD are at highest risk of ICH expansion, with an ICH expansion rate more than 5 times higher than subjects with neither of these markers. We also directly compared the diagnostic performance of SS and HD, showing that HD has higher sensitivity for ICH expansion whereas specificity was superior for SS.
Optimal stratification of hematoma expansion risk is a critical component of patient selection in clinical trials aiming at limiting the extent of bleeding in the acute phase of ICH. In one large international clinical trial, less than 20% of ICH patients received an early CTA and intensive blood pressure reduction did not decrease the risk of ICH expansion (12). In addition, the predictive performance of the SS in this minority of patients was poor, suggesting the need for imaging predictors of hematoma growth that do not require a CTA. Several NCCT markers of hematoma expansion, such as the blend sign, black hole sign, and hematoma density and shape have been described (5,13–15). Unfortunately, the relationship between SS and NCCT markers of ICH expansion remains poorly characterized. One study has investigated the interaction and clinical yield of integrating the blend sign, a NCCT predictor of hematoma growth, with the SS (16). In this analysis the majority of patients with a SS were also blend sign positive and when included in the same logistic regression analysis the SS was no longer a significant predictor of ICH expansion. This may indicate that CTA SS is simply an epiphenomenon of the NCCT marker blend sign (or vice-versa) and the combination of the two radiological signs does not improve the prediction of hematoma growth in ICH patients. Conversely, we showed that SS and HD predict hematoma growth independently from each other and their integrated analysis identifies patients at highest risk of ICH expansion.
These results may be particularly relevant in the setting of study design of future clinical trials testing anti-expansion therapies. The presence of SS+HD+ may mark those patients most likely to benefit from anti-expansion therapies, therefore maximizing statistical power to detect a superior treatment. Conversely, the lack of SS and HD identifies patients with the lowest risk of ICH expansion and therefore less likely to derive benefit from acute therapies targeting hematoma growth.
The combined evidence of SS and HD predicted hematoma expansion with good PPV and excellent specificity. This restrictive selection strategy (less than 15% of ICH patients were SS+HD+ in our cohort) may be more appropriate for clinical trials testing anti-expansion therapies with potential to harm, such as hemostatic drugs with thrombotic side-effects. A more inclusive definition such as the presence of SS and/or HD in any combination identified patients at risk of hemorrhage growth with high sensitivity and excellent NPV. This approach appears ideal for anti-expansion therapies with low risk of harm, such as intensive blood pressure reduction or administration of tranexamic acid, in order to include as many patients as possible at risk of further bleeding (17–20).
Our findings may also have important implications for ICH care in clinical practice because hematoma expansion is one of the main determinants of clinical deterioration in the early phase of ICH natural history (21). Subjects at high risk of ICH growth could be selected for more intensive neurological monitoring. In addition, as minimally invasive surgery techniques become available, patients at low risk of hematoma expansion may be more suitable for surgical hematoma evacuation due to a lower risk of postoperative rebleeding.
Although HD and SS are robust and validated predictors of ICH expansion, the biological mechanisms underlying these imaging signs remains poorly characterized and deserve further studies. One possible explanation for HD pathophysiology is the direct relationship between the degree of blood coagulation and hematoma density on CT (22). HD may indeed represent regions of non-coagulated blood from an actively bleeding sites in hemorrhages at a very early stage of development. In line with this hypothesis, HD presence is associated with shorter time from onset to NCCT and warfarin-related coagulopathy (6). Similarly, previous studied reported a strong association between anticoagulant therapy and SS presence (23) and also described a higher rate of iodine contrast extravasation in patients experiencing hematoma growth, indirectly supporting the theory that SS reflects active bleeding (24). However, in SS+HD+ patients we observed a clear spatial correspondence between the two markers in less than half of the subjects (6,13), raising the hypothesis that SS and HD may not represent the same pathophysiological process.
Some limitations should be considered in the interpretation of our results. First, although based on a large, well characterized ICH cohort, our findings derive from a nonrandomized, single-center retrospective analysis and prospective confirmation is required. Prospective data collection for independent, external validation of our findings is currently ongoing. Second, a significant proportion of patients were excluded from the analysis because of missing NCCT or CTA images, introducing a potential bias. CTA and follow-up NCCT scan may have been avoided because of withdrawal of care, in particular in patients with warfarin-associated hemorrhages (and therefore at very high risk of hematoma growth), leading to potential underestimation of the association between SS, HD and hemorrhage expansion. This hypothesis is indirectly supported by the observed higher prevalence of infratentorial, warfarin-associated hemorrhages with intraventricular bleeding in the excluded study population. Early limitation of care may have included avoidance of anti-expansion therapies such as intensive BP lowering and coagulopathy reversal (25), explaining therefore the higher rate of ICH expansion and mortality in the excluded patients.
Third, different definitions of SS have been reported and the use of other criteria for SS detection may change our results (3). In addition, we have analyzed only first pass CTA images for spot sign detection. Multiphase CTA may improve the recognition of spot sign and therefore modify our findings (26). Fourth, the subjects included were enrolled over a long time course with several changes in the NCCT and CTA scanners and acquisition protocols that may have influenced our analysis (9). Likewise, BP management in acute ICH has significantly changed in the last two decades and this may have had an impact on our results as well (27).
Fifth, the integration of other NCCT predictors of hematoma expansion with the SS or the combination of multiple NCCT markers may have different test characteristics. Finally, NCCT and CTA scans were analyzed by raters with extensive experience in imaging markers of ICH expansion. It remains to be determined whether SS and HD can be reliably and accurately detected by untrained clinicians that are not familiar with acute ICH neuroimaging, especially considering the lack of consensus criteria for identification of NCCT markers of expansion, with significant overlap between different signs.
CONCLUSION
The combined analysis of SS and HD improves the stratification of ICH expansion risk. This may help the selection of patients for clinical trials testing anti-expansion treatments.
Supplementary Material
Acknowledgments
Sources of funding
Dr. Anderson is supported by NIH-NINDS K23NS086873 award. The funding source did not have any involvement in study design; data collection, analysis, and interpretation; writing of the manuscript; or decision to submit the study for publication.
Footnotes
Disclosures:
Andrea Morotti reports no disclosures.
Gregoire Boulouis reports no disclosures.
Andreas Charidimou reports no disclosures.
Kristin Schwab reports no disclosures.
Christina Kourkoulis reports no disclosures.
Christopher D. Anderson reports NIH-NINDS funding and consulting for ApoPharma.
M. Edip Gurol reports no disclosures.
Anand Viswanathan reports no disclosures.
Javier M. Romero reports no disclosures.
Steven M. Greenberg reports no disclosures.
Jonathan Rosand reports reports NIH-NINDS funding, clinical trial monitoring for Pfizer and consulting for Boehringer Ingelheim.
Joshua N. Goldstein reports research grant and clinical trial monitoring from Pfizer, research grant from Portola, consulting for CSL Behring and Octapharma.
Contributor Information
Andrea Morotti, Stroke Unit, IRCCS Mondino Foundation, Pavia, Italy.
Gregoire Boulouis, Department of Neuroradiology, Université Paris Descartes, INSERM S894, DHU Neurovasc, Centre Hospitalier Sainte-Anne, Paris, France.
Andreas Charidimou, J. P. Kistler Stroke Research Center, Massachusetts General Hospital, Harvard Medical School, Boston. USA.
Kristin Schwab, J. P. Kistler Stroke Research Center, Massachusetts General Hospital, Harvard Medical School, Boston. USA.
Christina Kourkoulis, J. P. Kistler Stroke Research Center, Massachusetts General Hospital, Harvard Medical School, Boston. USA.
Christopher D. Anderson, J. P. Kistler Stroke Research Center, Massachusetts General Hospital, Harvard Medical School, Boston. USA.
M. Edip Gurol, J. P. Kistler Stroke Research Center, Massachusetts General Hospital, Harvard Medical School, Boston. USA.
Anand Viswanathan, J. P. Kistler Stroke Research Center, Massachusetts General Hospital, Harvard Medical School, Boston. USA.
Javier M. Romero, Neuroradiology Service, Department of Radiology, Massachusetts General Hospital, Harvard Medical School Boston, MA, USA.
Steven M. Greenberg, J. P. Kistler Stroke Research Center, Massachusetts General Hospital, Harvard Medical School, Boston. USA.
Jonathan Rosand, J. P. Kistler Stroke Research Center, Massachusetts General Hospital, Harvard Medical School, Boston. USA.
Joshua N. Goldstein, J. P. Kistler Stroke Research Center, Massachusetts General Hospital, Harvard Medical School, Boston. USADivision of Neurocritical Care and Emergency Neurology, Massachusetts General Hospital, Boston (JR, JNG).Department of Emergency Medicine, Massachusetts General Hospital, Boston (JNG).
References
- 1.Qureshi AI, Mendelow AD, Hanley DF. Intracerebral haemorrhage. Lancet. 2009 May 9;373:1632–44. doi: 10.1016/S0140-6736(09)60371-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Steiner T, Bösel J. Options to restrict hematoma expansion after spontaneous intracerebral hemorrhage. Stroke. 2010;41:402–9. doi: 10.1161/STROKEAHA.109.552919. [DOI] [PubMed] [Google Scholar]
- 3.Demchuk AM, Dowlatshahi D, Rodriguez-Luna D, Molina CA, Blas YS, Dzialowski I, et al. Prediction of haematoma growth and outcome in patients with intracerebral haemorrhage using the CT-angiography spot sign (PREDICT): A prospective observational study. Lancet Neurol. 2012;11:307–14. doi: 10.1016/S1474-4422(12)70038-8. [DOI] [PubMed] [Google Scholar]
- 4.Goldstein JN, Fazen LE, Snider R, Schwab K, Greenberg SM, Smith EE, et al. Contrast extravasation on CT angiography predicts hematoma expansion in intracerebral hemorrhage. Neurology. 2007;68:889–94. doi: 10.1212/01.wnl.0000257087.22852.21. [DOI] [PubMed] [Google Scholar]
- 5.Morotti A, Boulouis G, Romero JM, Brouwers HB, Jessel MJ, Vashkevich A, et al. Blood pressure reduction and noncontrast CT markers of intracerebral hemorrhage expansion. Neurology. 2017;89:548–54. doi: 10.1212/WNL.0000000000004210. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Boulouis G, Morotti A, Brouwers HB, Charidimou A, Jessel MJ, Auriel E, et al. Association Between Hypodensities Detected by Computed Tomography and Hematoma Expansion in Patients With Intracerebral Hemorrhage. JAMA Neurol. 2016;73:961–8. doi: 10.1001/jamaneurol.2016.1218. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Morotti A, Phuah C-L, Anderson CD, Jessel MJ, Schwab K, Ayres AM, et al. Leukocyte Count and Intracerebral Hemorrhage Expansion. Stroke. 2016;47:1473–8. doi: 10.1161/STROKEAHA.116.013176. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Hemphill JC, Greenberg SM, Anderson CS, Becker K, Bendok BR, Cushman M, et al. Guidelines for the Management of Spontaneous Intracerebral Hemorrhage: A Guideline for Healthcare Professionals From the American Heart Association/American Stroke Association. Stroke. 2015;46:2032–60. doi: 10.1161/STR.0000000000000069. [DOI] [PubMed] [Google Scholar]
- 9.Morotti A, Romero JM, Jessel MJ, Brouwers HB, Gupta R, Schwab K, et al. Effect of CTA Tube Current on Spot Sign Detection and Accuracy for Prediction of Intracerebral Hemorrhage Expansion. AJNR Am J Neuroradiol. 2016;37:1781–86. doi: 10.3174/ajnr.A4810. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Dowlatshahi D, Demchuk AM, Flaherty ML, Ali M, Lyden PL, Smith EE. Defining hematoma expansion in intracerebral hemorrhage: Relationship with patient outcomes. Neurology. 2011;76:1238–44. doi: 10.1212/WNL.0b013e3182143317. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Brouwers HB, Greenberg SM. Hematoma expansion following acute intracerebral hemorrhage. Cerebrovasc Dis. 2013;35:195–201. doi: 10.1159/000346599. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Morotti A, Brouwers HB, Romero JM, Jessel MJ, Vashkevich A, Schwab K, et al. Intensive Blood Pressure Reduction and Spot Sign in Intracerebral Hemorrhage. JAMA Neurol. 2017;74:950–60. doi: 10.1001/jamaneurol.2017.1014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Boulouis G, Morotti A, Charidimou A, Dowlatshahi D, Goldstein JN. Noncontrast Computed Tomography Markers of Intracerebral Hemorrhage Expansion. Stroke. 2017;48:1120–5. doi: 10.1161/STROKEAHA.116.015062. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Li Q, Zhang G, Huang YJ, Dong MX, Lv FJ, Wei X, et al. Blend Sign on Computed Tomography: Novel and Reliable Predictor for Early Hematoma Growth in Patients With Intracerebral Hemorrhage. Stroke. 2015;46:2119–23. doi: 10.1161/STROKEAHA.115.009185. [DOI] [PubMed] [Google Scholar]
- 15.Blacquiere D, Demchuk AM, Al-Hazzaa M, Deshpande A, Petrcich W, Aviv RI, et al. Intracerebral Hematoma Morphologic Appearance on Noncontrast Computed Tomography Predicts Significant Hematoma Expansion. Stroke. 2015;46:3111–6. doi: 10.1161/STROKEAHA.115.010566. [DOI] [PubMed] [Google Scholar]
- 16.Sporns PB, Schwake M, Schmidt R, Kemmling A, Minnerup J, Schwindt W, et al. Computed Tomographic Blend Sign Is associated With Computed Tomographic Angiography Spot Sign and Predicts Secondary Neurological Deterioration After Intracerebral Hemorrhage. Stroke. 2017;48:131–35. doi: 10.1161/STROKEAHA.116.014068. [DOI] [PubMed] [Google Scholar]
- 17.Boulouis G, Morotti A, Goldstein JN, Charidimou A. Intensive blood pressure lowering in patients with acute intracerebral haemorrhage: clinical outcomes and haemorrhage expansion. Systematic review and meta-analysis of randomised trials. J Neurol Neurosurg Psychiatry. 2017;88:339–45. doi: 10.1136/jnnp-2016-315346. [DOI] [PubMed] [Google Scholar]
- 18.Qureshi AI, Palesch YY, Barsan WG, Hanley DF, Hsu CY, Martin RL, et al. Intensive Blood-Pressure Lowering in Patients with Acute Cerebral Hemorrhage. N Engl J Med 2016. 2016;375:1033–43. doi: 10.1056/NEJMoa1603460. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Anderson C, Heeley E, Huang Y, Wang J, Stapf C, Delcourt C, et al. Rapid Blood-Pressure Lowering in Patients with Acute Intracerebral Hemorrhage. N Engl J Med. 2013;368:2355–65. doi: 10.1056/NEJMoa1214609. [DOI] [PubMed] [Google Scholar]
- 20.Sprigg N, Renton CJ, Dineen RA, Kwong Y, Bath PMW. Tranexamic Acid for Spontaneous Intracerebral Hemorrhage: A Randomized Controlled Pilot Trial (ISRCTN50867461) J Stroke Cerebrovasc Dis. 2014;23:1312–8. doi: 10.1016/j.jstrokecerebrovasdis.2013.11.007. [DOI] [PubMed] [Google Scholar]
- 21.Balami JS, Buchan AM. Complications of intracerebral haemorrhage. Lancet Neurol. 2012;11(1):101–18. doi: 10.1016/S1474-4422(11)70264-2. [DOI] [PubMed] [Google Scholar]
- 22.New PF, Aronow S. Attenuation measurements of whole blood and blood fractions in computed tomography. Radiology. 1976;121:635–40. doi: 10.1148/121.3.635. [DOI] [PubMed] [Google Scholar]
- 23.Radmanesh F, Falcone GJ, Anderson CD, Battey TWK, Ayres AM, Vashkevich A, et al. Risk factors for computed tomography angiography spot sign in deep and lobar intracerebral hemorrhage are shared. Stroke. 2014;45:1833–5. doi: 10.1161/STROKEAHA.114.005276. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Brouwers HB, Battey TWK, Musial HH, Ciura VA, Falcone GJ, Ayres AM, et al. Rate of Contrast Extravasation on Computed Tomographic Angiography Predicts Hematoma Expansion and Mortality in Primary Intracerebral Hemorrhage. Stroke. 2015;46:2498–503. doi: 10.1161/STROKEAHA.115.009659. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Kuramatsu JB, Gerner ST, Schellinger PD, Glahn J, Endres M, Sobesky J, et al. Anticoagulant Reversal, Blood Pressure Levels, and Anticoagulant Resumption in Patients With Anticoagulation-Related Intracerebral Hemorrhage. Jama. 2015;313:824–36. doi: 10.1001/jama.2015.0846. [DOI] [PubMed] [Google Scholar]
- 26.Rodriguez-Luna D, Coscojuela P, Rodriguez-Villatoro N, Juega JM, Boned S, Muchada M, et al. Multiphase CT Angiography Improves Prediction of Intracerebral Hemorrhage Expansion. Radiology. 2017;285:932–40. doi: 10.1148/radiol.2017162839. [DOI] [PubMed] [Google Scholar]
- 27.Majidi S, Suarez JI, Qureshi AI. Management of Acute Hypertensive Response in Intracerebral Hemorrhage Patients After ATACH-2 Trial. Neurocrit Care. 2017;27:249–258. doi: 10.1007/s12028-016-0341-z. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.