This secondary analysis of a randomized clinical trial investigates whether the spot sign is associated with intracerebral hemorrhage expansion and whether intensive blood pressure reduction reduces hematoma expansion and improves outcomes.
Key Points
Question
Does the computed tomographic angiography spot sign identify the patients with intracerebral hemorrhage who are most likely to benefit from intensive blood pressure reduction?
Findings
In this secondary analysis of 133 patients with intracerebral hemorrhage nested in the Antihypertensive Treatment of Acute Cerebral Hemorrhage II randomized clinical trial, intensive blood pressure treatment did not improve outcomes in patients with a spot sign. The proportion of patients with a spot sign and a 90-day modified Rankin Scale score of 4 or greater was 20 of 27 in the intensive treatment group vs 10 of 20 in the standard treatment group.
Meaning
Our results showed no clinical benefit of intensive blood pressure reduction in patients with intracerebral hemorrhage and a spot sign.
Abstract
Importance
The computed tomographic angiography (CTA) spot sign is associated with intracerebral hemorrhage (ICH) expansion and may mark those patients most likely to benefit from intensive blood pressure (BP) reduction.
Objective
To investigate whether the spot sign is associated with ICH expansion across a wide range of centers and whether intensive BP reduction decreases hematoma expansion and improves outcome in patients with ICH and a spot sign.
Design, Setting, and Participants
SCORE-IT (Spot Sign Score in Restricting ICH Growth) is a preplanned prospective observational study nested in the Antihypertensive Treatment of Acute Cerebral Hemorrhage II (ATACH-II) randomized clinical trial. Participants included consecutive patients with primary ICH who underwent a CTA within 8 hours from onset at 59 sites from May 15, 2011, through December 19, 2015. Data were analyzed for the present study from July 1 to August 31, 2016.
Main Outcomes and Measures
Patients in ATACH-II were randomized to intensive (systolic BP target, <140 mm Hg) vs standard (systolic BP target, <180 mm Hg) BP reduction within 4.5 hours from onset. Expansion of ICH was defined as hematoma growth of greater than 33%, and an unfavorable outcome was defined as a 90-day modified Rankin Scale score of 4 or greater (range, 0-6). The association among BP reduction, ICH expansion, and outcome was investigated with multivariable logistic regression.
Results
A total of 133 patients (83 men [62.4%] and 50 women [37.6%]; mean [SD] age, 61.9 [13.1] years) were included. Of these, 53 (39.8%) had a spot sign, and 24 of 123 without missing data (19.5%) experienced ICH expansion. The spot sign was associated with expansion with sensitivity of 0.54 (95% CI, 0.34-0.74) and specificity of 0.63 (95% CI, 0.53-0.72). After adjustment for potential confounders, intensive BP treatment was not associated with a significant reduction of ICH expansion (relative risk, 0.83; 95% CI, 0.27-2.51; P = .74) or improved outcome (relative risk of 90-day modified Rankin Scale score ≥4, 1.24; 95% CI, 0.53-2.91; P = .62) in spot sign–positive patients.
Conclusions and Relevance
The predictive performance of the spot sign for ICH expansion was lower than in prior reports from single-center studies. No evidence suggested that patients with ICH and a spot sign specifically benefit from intensive BP reduction.
Trial Registration
clinicaltrials.gov Identifier: NCT01176565
Introduction
Intracerebral hemorrhage (ICH) is still the deadliest type of stroke, with a case fatality rate at 30 days of approximately 40% and severe disability in most of the survivors. Baseline ICH volume is the variable most strongly associated with an unfavorable outcome, and as many as one-third of patients with ICH experience significant hematoma growth. Because of its frequency and association with poor prognosis, hematoma expansion is an appealing target for acute ICH treatment. Stratification of ICH expansion risk may help target patients with the highest likelihood to benefit from antiexpansion treatment. An association of the computed tomographic angiography (CTA) spot sign has been found with hematoma growth in a range of single-center studies and 1 multicenter study and may improve our ability to stratify the risk for hematoma expansion in clinical practice. However, different criteria for spot sign definition have been reported, with heterogeneity in spot sign density, dimension, morphologic features, and CTA window. To our knowledge, the diagnostic performance of different spot sign definitions has never been directly compared, and only 1 spot sign definition was validated in a multicenter study. In the present study, we investigated whether the CTA spot sign is associated with hematoma growth across a wide range of centers in a large international clinical trial and whether it identifies patients with ICH more likely to benefit from intensive BP reduction. In particular, we tested the hypothesis that intensive BP control would reduce ICH expansion and improve outcome in spot sign–positive patients, whereas those without a spot sign would not benefit from intensive BP reduction. As a secondary analysis, we compared the diagnostic performance of the 2 most commonly used spot sign definitions in the association of hematoma expansion and an unfavorable outcome.
Methods
Participants
The SCORE-IT (Spot Sign Score in Restricting ICH Growth) study was a prospective preplanned analysis of patients enrolled in the Antihypertensive Treatment of Acute Cerebral Hemorrhage II (ATACH-II) randomized clinical trial. In ATACH-II, patients with ICH were randomly assigned to intensive (systolic BP target, <140 mm Hg) vs standard (systolic BP target, <180 mm Hg) BP treatment within 4.5 hours after stroke onset. Demographic and clinical characteristics were collected at enrollment in the clinical trial, as previously described. Patients enrolled in ATACH-II were eligible for SCORE-IT if they underwent a CTA. All aspects of the study were approved by the local institutional review boards of each participating center listed at the end of the article. Written informed consent was obtained from patients or family members.
Image Acquisition and Analysis
Axial non–contrast-enhanced computed tomographic (NCCT) and CTA images were obtained at each participant’s institution by using standard local protocols. We retrospectively collected the following CTA acquisition variables: tube current level (in milliamperes) and tube voltage level (in peak kilovoltage). Baseline and follow-up hematoma volumes at 24 hours were calculated on NCCT images with semiautomated computer-assisted volumetric analysis (AnalyzeDirect medical imaging software, version 11.0; AnalyzeDirect, Inc).
First-pass CTA images were independently reviewed by trained readers (A.M. and M.J.J.) for spot sign identification according to the 2 most commonly used spot sign definitions. Disagreements in spot sign readings were adjudicated by consensus under the supervision of a neuroradiologist (J.M.R.). For the purposes of this analysis, definition 1 was validated at Massachusetts General Hospital, Boston, whereas definition 2 was validated in the Prediction of Hematoma Growth and Outcome in Patients With Intracerebral Hemorrhage Using the CT-Angiography Spot Sign (PREDICT) study.
The diagnostic criteria used in spot sign definition 1 consisted of the presence of at least 1 focus of contrast extravasation in the hematoma, with any size or morphologic feature and density greater than 120 Hounsfield units. Definition 2 included intrahematoma contrast extravasation with a minimum diameter of 1.5 mm in any dimension, with spotlike or serpiginous shape, without any connection with vessels outside the hematoma, without corresponding hyperdensity at NCCT, and with attenuation at least double that of the background hematoma (in Hounsfield units). An illustrative comparison of both diagnostic criteria is shown in the eFigure in the Supplement.
All NCCT and CTA images were carefully reviewed to avoid the inclusion of spot sign mimics such as calcifications (visible on NCCT images) or the presence of connection between the spot sign and vessels outside the hematoma. The primary analysis on the effect of intensive BP control in spot sign–positive patients was performed using definition 1, as specified in the SCORE-IT study protocol. As a secondary analysis, CTA spot sign–positive patients according to definition 2 underwent analysis.
We had initially planned to include only CTA images obtained at those institutions that routinely perform CTA at admission in patients with ICH. However, during the trial, heterogeneity across all sites in CTA performance precluded using this inclusion criterion. The main analysis therefore included all CTA images that were obtained within 8 hours from symptom onset (operationally termed early CTA). As a secondary analysis, we included all CTA images irrespective of time performed.
Finally, we performed another analysis stratifying the risk for hematoma expansion with the spot sign score, a validated scoring system that accounts for the number, attenuation, and dimension of spot signs. The spot sign score ranges from 0 to 4 and, as the score increases, the likelihood of hematoma growth increases.
Randomization and Masking
Eligible patients were randomized at a 1:1 ratio via a centralized web-based randomization process to the intensive or standard antihypertensive treatment group. A minimization randomization scheme proposed by Pocock and Simon was used to control for age, baseline Glasgow Coma Scale score, presence of intraventricular hemorrhage, and clinical site. These baseline prognostic factors were identified a priori as those with imbalances that could potentially affect the primary outcome of the trial. Because of the nature of the treatment in this trial in which the infusion of nicardipine hydrochloride was titrated to target systolic BP, the treatment physicians and study personnel could not be blinded to treatment assignment. However, all sites were required to designate an individual who was blinded to treatment assignment and did not participate in the randomization or treatment of patients in the trial to be the blinded assessor of the 30- and 90-day outcomes.
Outcome Measures
The primary outcome of the present analysis was the proportion of spot sign–positive patients who experienced hematoma expansion and had a poor functional outcome at 90 days. Hematoma expansion was defined as absolute volume increase of greater than 33% from the baseline ICH volume. Functional outcome was measured with the modified Rankin Scale (mRS) score, and an unfavorable outcome was defined as an mRS score of 4 or greater (range, 0-6, with higher scores indicating greater disability and 6 indicating death) at 3 months.
Statistical Analysis
Continuous variables with normal distribution were expressed as mean (SD) and with nonnormal distribution as median (range), whereas categorical variables were expressed as count (percentage). Sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of the 2 spot sign definitions in predicting ICH expansion and poor long-term outcome are presented. The interrater reliability for the identification of any spot sign was determined using the Cohen κ statistic.
The association among BP treatment, ICH expansion, and outcome was investigated with a multivariable logistic regression analysis, stratified by spot sign status, and adjusted for known variables associated with hematoma growth and poor outcome in patients with ICH. Owing to the exploratory nature of this study, P < .05 was considered to be statistically significant, and all the analyses were performed using SAS software (version 9.4; SAS Institute Inc).
Results
The ATACH-II study was terminated early as advised by the monitoring board owing to futility. Of 1000 patients in the ATACH-II trial, 133 (13.3%) underwent a CTA within 8 hours after symptom onset (101 before and 32 after randomization) and were included in the present study (83 men [62.4%] and 50 women [37.6%]; mean [SD] age, 61.9 [13.1] years). The overall proportion of patients with at least 1 spot sign was 53 (39.8%) with definition 1 and 29 (21.8%) with definition 2. Functional outcome data at 90 days or follow-up NCCT for evaluation of ICH expansion were missing in 10 patients.
A total of 24 of 123 patients (19.5%) experienced hematoma expansion. The case-fatality rate at 90 days was 8.3% (11 patients), and 56 of 123 patients (45.5%) had a poor functional outcome at 3 months. The demographic and clinical characteristics of the study population, stratified by CTA timing, are summarized in Table 1. The Figure shows the cohort selection flowchart.
Table 1. Baseline Characteristics.
| Characteristic | ATACH-II Participants by CTA Timing | All ATACH-II Participants (N = 1000) |
||
|---|---|---|---|---|
| <8 h Before Randomization (n = 101) |
<8 h After Randomization (n = 32) |
>8 h From Randomization (n = 33) |
||
| Age, mean (SD), y | 61.4 (12.8) | 57.2 (11.7) | 62.8 (14.4) | 61.9 (13.1) |
| Male, No./total No. (%) | 64/101 (63.4) | 19/32 (59.4) | 14/33 (42.4) | 620/1000 (62) |
| Hypertension, No./total No. (%) | 77/96 (80.2) | 24/29 (82.8) | 26/33 (78.8) | 793/973 (81.5) |
| Admission SBP, mean (SD), mm Hg | 204.8 (25.8) | 204.1 (33.5) | 197.9 (28.2) | 200.6 (27.0) |
| Diabetes, No./total No. (%) | 16/97 (16.5) | 5/30 (16.7) | 4/33 (12.1) | 185/981 (18.9) |
| Creatinine level, median (range), mg/dL | 0.9 (0.1-1.8) | 0.9 (0.4-2.0) | 0.9 (0.5-1.7) | 0.9 (0.1-13.0) |
| eGFR<60 mL/min/1.73 m2, No./total No. (%) | 18/101 (17.8) | 7/32 (21.9) | 8/33 (24.2) | 195/1000 (19.5) |
| Randomized to SBP reduction to <140 mm Hg, No./total No. (%) | 57/101 (56.4) | 18/32 (56.3) | 17/33 (51.5) | 500/1000 (50) |
| Initial GCS score, No./total No. (%)a | ||||
| 3-11 | 12/101 (11.9) | 3/32 (9.4) | 8/33 (24.2) | 147/1000 (14.7) |
| 12-14 | 30/101 (29.7) | 5/32 (15.6) | 11/33 (33.3) | 294/1000 (29.4) |
| 15 | 59/101 (58.4) | 24/32 (75) | 14/33 (42.4) | 559/1000 (55.9) |
| Time from onset to arrival, mean (SD), min | 70.9 (41.3) | 55.6 (26.8) | 58.6 (37.0) | 72.0 (44.8) |
| Time from onset to baseline NCCT, mean (SD), min | 109.3 (54.1) | 78.3 (42.6) | 102.6 (112.0) | 98.9 (54.0) |
| Time from onset to CTA, mean (SD), min | 126.0 (53.9) | 281.4 (114.0) | 1169.5 (538.7 | |
| Time from onset to randomization, mean (SD), min | 206.5 (49.4) | 170.2 (66.7) | 190.2 (57.4) | 183.5 (56.9) |
| Time from onset to start of nicardipine hydrochloride therapy, mean (SD), min | 157.4 (70.1) | 134.8 (66.6) | 167.2 (138.4) | 157.0 (85.1) |
| ICH volume, median (range), mLbb | 10.5 (0.7-65.5) | 8.6 (1.9-85.2) | 15.5 (0.3-48.8) | 10.2 (0.1-85.2) |
| IVH present, No./total No. (%) | 30/100 (30) | 13/32 (40.6) | 14/32 (43.8) | 264/988 (26.7) |
| ICH expansion, No./total No. (%)c | 20/95 (21.1) | 4/28 (14.3) | 9/25 (36) | 189/876 (21.6) |
| Spot sign definition 1, No./total No. (%)d | 44/101 (43.6) | 9/32 (28.1) | 12/33 (36.4) | NA |
| Spot sign definition 2, No./total No. (%)e | 28/101 (27.7) | 1/32 (3.1) | 5/33 (15.2) | NA |
| CTA tube voltage, median (range), kVp | 120 (80-140) | 120 (80-140) | 120 (100-140) | NA |
| CTA tube current, median (range), mA | 350 (100-765) | 350 (85-714) | 450 (134-884) | NA |
| Mortality at 90 d, No./total No. (%) | 8/101 (7.9) | 3/32 (9.4) | 6/33 (18.2) | 67/1000 (6.7) |
| mRS score ≥4 at 90 d, No./total No. (%)f | 41/92 (44.6) | 15/31 (48.4) | 19/32 (59.4) | 367/961 (38.2) |
Abbreviations: ATACH-II, Antihypertensive Treatment of Acute Cerebral Hemorrhage II; CTA, computed tomographic angiography; eGFR, estimated glomerular filtration rate; GCS, Glasgow Coma Scale; ICH, intracerebral hemorrhage; IVH, intraventricular hemorrhage; kVp, peak kilovoltage; mA, milliamperes; mRS, modified Rankin Scale; NA, not applicable; NCCT, non–contrast-enhanced computed tomography; SBP, systolic blood pressure.
SI conversion factor: To convert creatinine to micromoles per liter, multiply by 88.4.
Higher scores indicate better level of consciousness.
Hematoma volume was measured by a central reader. The rapid assessment of the hematoma volume by the site investigator was used to determine eligibility.
Hematoma expansion was defined as an increase of 33% or more in the hematoma volume from baseline to 24 hours.
Defined as the presence of at least 1 focus of contrast extravasation in the hematoma with any size or morphologic feature and density of greater than 120 Hounsfield units.
Defined as intrahematoma contrast extravasation with a minimum diameter of 1.5 mm in any dimension, with spotlike or serpiginous shape, not connected with vessels outside the hematoma, without corresponding hyperdensity on NCCT, and attenuation at least double that of the background hematoma.
Scores range from 0 to 6, with 6 indicating death.
Figure. Cohort Selection Flowchart and Trial Profile.
ATACH-II indicates Antihypertensive Treatment of Acute Cerebral Hemorrhage II; CTA, computed tomographic angiography; ICH, intracerebral hemorrhage; SBP, systolic blood pressure; and SCORE-IT, Spot Sign Score in Restricting ICH Growth.
Comparison of Different Spot Sign Definitions
The interrater reliability for the identification of any spot sign was 0.75 (95% CI, 0.69-0.81) for definition 1 and 0.79 (95% CI, 0.73-0.85) for definition 2. The number of spot sign–positive CTAs was significantly higher using definition 1 (Table 2), and all the spot sign–positive scans identified with definition 2 were also spot sign positive according to definition 1. The diagnostic performance of the 2 spot sign definitions in predicting ICH expansion and unfavorable outcome is summarized in Table 2.
Table 2. Comparison Between Different Spot Sign Definitionsa.
| Variable | Definition 1 | Definition 2 |
|---|---|---|
| Diagnostic criteria | ||
| CTA window, width/length | 200/110 | 177/77 |
| Density, HU | ≥120 | ≥ICH background ×2 |
| Shape | Any | Spot or serpiginous |
| Dimension, mm | Any | >1.5 |
| Exclusion of spot sign mimicsb | Same | Same |
| Spot signs, No. (%)c | ||
| None | 80 (60.2) | 104 (78.2) |
| Single | 23 (17.3) | 19 (14.3) |
| Multiple | 30 (22.6) | 10 (7.5) |
| ICH expansiond,e | ||
| Sensitivity (95% CI) | 0.54 (0.34-0.74) | 0.38 (0.18-0.57) |
| Specificity (95% CI) | 0.63 (0.53-0.72) | 0.81 (0.73-0.89) |
| PPV (95% CI) | 0.26 (0.14-0.38) | 0.32 (0.15-0.49) |
| NPV (95% CI) | 0.85 (0.77-0.93) | 0.84 (0.77-0.92) |
| Accuracy | 0.61 | 0.72 |
| mRS score ≥4 at 90 de,f | ||
| Sensitivity (95% CI) | 0.54 (0.41-0.67) | 0.32 (0.20-0.44) |
| Specificity (95% CI) | 0.75 (0.64-0.85) | 0.88 (0.80-0.96) |
| PPV (95% CI) | 0.64 (0.50-0.78) | 0.69 (0.51-0.87) |
| NPV (95% CI) | 0.66 (0.55-0.76) | 0.61 (0.51-0.71) |
| Accuracy | 0.65 | 0.62 |
Abbreviations: CTA, computed tomographic angiography; HU, Hounsfield unit; ICH, intracerebral hemorrhage; mRS, modified Rankin scale; NPV, negative predictive value; PPV, positive predictive value.
Includes 133 patients enrolled in the SCORE-IT (Spot Sign Score in Restricting ICH Growth) study.
Both definitions used lack of corresponding hyperdensity on non–contrast-enhanced computed tomography and lack of connection between the spot sign and vessels outside the hemorrhage.
P < .001.
Defined as volume increase of greater than 33%.
Data are represented as proportions.
Scores range from 0 to 6, with 6 indicating death.
Effect of Intensive BP Treatment on ICH Expansion and Outcome
Table 3 shows the differences between patients positive and negative for the spot sign, stratified by BP treatment (intensive vs standard BP lowering). The proportion of patients with a spot sign and a 90-day mRS score of 4 or greater was 20 of 27 (74.1%) in the intensive treatment group vs 10 of 20 (50%) in the standard treatment group. As expected, patients with a CTA spot sign had a larger baseline hematoma volume and a higher frequency of expansion. Intensive BP reduction was not associated with a significantly lower rate of hematoma expansion in spot sign–positive patients in univariate analysis (7 of 31 [22.6%] vs 6 of 19 [31.6%]; P = .52). We found no indication of an interactive effect between spot sign status and BP treatment. The univariate analysis was repeated including only CTA obtained before randomization using both spot sign definitions (eTables 1 and 2 in the Supplement).
Table 3. Univariate Analysis Including All CTA Within 8 Hours From Onseta.
| Characteristic | Spot Sign Negative Definition 1 | Spot Sign Positive Definition 1 | ||
|---|---|---|---|---|
| SBP Reduction <140 mm Hg (n = 44) |
SBP Reduction <180 mm Hg (n = 36) |
SBP Reduction <140 mm Hg (n = 31) |
SBP Reduction <180 mm Hg (n = 22) |
|
| Age, mean (SD), y | 61.5 (13.6) | 60.2 (10.5) | 60.9 (13.6) | 57.6 (13.0) |
| Male, No./total No. (%) | 26/44 (59.1) | 26/36 (72.2) | 19/31 (61.3) | 12/22 (54.5) |
| Hypertension, No./total No. (%) | 39/42 (92.9) | 27/35 (77.1) | 23/28 (82.1) | 12/20 (60) |
| Admission SBP, mean (SD), mm Hg | 204.6 (20.6) | 205.2 (30.2) | 206.0 (35.4) | 201.9 (24.7) |
| Diabetes, No. (%) | 9/41 (22) | 6/35 (17.1) | 2/30 (6.7) | 4/21 (19.0) |
| Initial GCS score, No./total No. (%)b | ||||
| 3-11 | 1/44 (2.3) | 2/36 (5.6) | 9/31 (29) | 3/22 (13.6) |
| 12-14 | 14/44 (31.8) | 8/36 (22.2) | 10/31 (32.3) | 3/22 (13.6) |
| 15 | 29/44 (65.9) | 26/36 (72.2) | 12/31 (38.7) | 16/22 (72.7) |
| ICH location, No./total No. (%) | ||||
| Lobar | 6/44 (13.6) | 6/36 (16.7) | 8/31 (25.8) | 6/22 (27.3) |
| Deep | 38/44 (86.4) | 30/36 (83.3) | 23/31 (74.2) | 16/22 (72.7) |
| IVH present, No./total No. (%) | 14/44 (31.8) | 9/35 (25.7) | 14/31 (45.2) | 6/22 (27.3) |
| Time from onset to arrival, mean (SD), minc | 62.7 (37.1) | 72.3 (38.7) | 69.0 (41.5) | 65.4 (39.9) |
| Time from onset to baseline NCCT, mean (SD), min | 99.0 (58.3) | 98.2 (47.4) | 112.1 (52.0) | 98.5 (53.9) |
| Time from onset to CTA, mean (SD), min | 173.0 (104.7) | 186.6 (120.9) | 149.1 (61.9) | 126.3 (76.2) |
| Time from onset to randomization, mean (SD), min | 201.8 (55.8) | 198.7 (57.5) | 202.1 (55.5) | 182.3 (55.8) |
| Time from onset to start of nicardipine hydrochloride therapy, mean (SD), min | 141.6 (67.4) | 172.9 (66.5) | 151.8 (71.5) | 137.2 (73.5) |
| Baseline ICH volume, median (range), mLd | 7.3 (0.7-49.7) | 6.6 (1.4-34.1) | 13.7 (1.4-85.2) | 16.4 (2.7-65.5) |
| Baseline ICH + IVH volume, median (range), mL | 12.2 (0.7-49.7) | 10.4 (1.4-61.1) | 18.2 (1.4-86.7) | 18.5 (2.7-65.5) |
| Follow-up ICH volume, median (range), mL | 7.2 (0.8-60.7) | 7.0 (1.3-74.4) | 13.5 (1.3-65.8) | 23.5 (4.4-69.3) |
| Follow-up ICH + IVH volume, media (range), mL | 9.1 (0.8-60.7) | 11.5 (1.3-118.7) | 20.2 (1.3-65.8) | 24.1 (4.4-69.3) |
| Any ICH volume increase, No./total No. (%) | 20/40 (50) | 20/33 (60.6) | 20/31 (64.5) | 13/19 (68.4) |
| Absolute ICH volume increase, mean (SD), mL | 4.3 (10.5) | 7.3 (16.6) | 6.7 (7.8) | 6.6 (8.1) |
| Any ICH + IVH volume increase, No./Total No. (%) | 19/40 (47.5) | 17/33 (51.5) | 16/31 (51.6) | 12/19 (63.2) |
| Absolute ICH + IVH volume increase, mean (SD), mL | 5.4 (10.7) | 13.2 (31.0) | 9.1 (7.9) | 8.4 (10.1) |
| ICH expansion, No./total No. (%)e | 6/40 (15) | 5/33 (15.2) | 7/31 (22.6) | 6/19 (31.6) |
| mRS score ≥4 at 90 d, No./total No. (%)f | 15/40 (37.5) | 11/36 (30.6) | 20/27 (74.1) | 10/20 (50) |
| Mortality at 90 d, No./total No. (%) | 5/44 (11.4) | 2/36 (5.6) | 3/31 (9.7) | 1/22 (4.5) |
Abbreviations: CTA, computed tomographic angiography; GCS, Glasgow Coma Scale; ICH, intracerebral hemorrhage; IVH, intraventricular hemorrhage; mRS, modified Rankin Scale; NCCT, non–contrast-enhanced computed tomography; SBP, systolic blood pressure.
Includes 133 patients enrolled in the SCORE-IT (Spot Sign Score in Restricting ICH Growth) study. Percentages have been rounded and may not total 100.
Higher scores indicate better level of consciousness.
One patient was hospitalized at the time of stroke and was excluded because time to arrival was negative.
Hematoma volume was measured by a central reader. The rapid assessment of the hematoma volume by the site investigator was used to determine eligibility.
Hematoma expansion was defined as an increase of 33% or more in the hematoma volume from baseline to 24 hours.
Scores range from 0 to 6, with 6 indicating death.
After adjustment for potential confounders in the multivariable regression analysis, intensive BP reduction in spot sign–positive patients was not associated with reduced ICH growth or a better outcome at 3 months (Table 4). Similar results were found using spot sign definition 2 (Table 4). We also obtained the same results when we restricted all the analyses to the subgroup of patients who underwent CTA before randomization (n = 101).
Table 4. Effect of Intensive BP Treatment on ICH Expansion and Outcome Stratified by Spot Sign Status Using Early CTA.
| Spot Sign Finding | ICH Expansiona | 90-d mRS Score ≥4b | ||
|---|---|---|---|---|
| RR (95% CI) | P Value | RR (95% CI) | P Value | |
| Positive, definition 1c | 0.83 (0.27-2.51) | .74 | 1.24 (0.53-2.91) | .62 |
| Positive, definition 2d | 0.49 (0.12-2.10) | .34 | 1.17 (0.38-3.60) | .78 |
| Negative, definition 1c | 1.03 (0.31-3.38) | .96 | 0.88 (0.38-2.03) | .76 |
| Negative, definition 2d | 1.45 (0.49-4.30) | .50 | 1.13 (0.56-2.30) | .73 |
Abbreviations: BP, blood pressure; CTA, computed tomographic angiography; ICH, intracerebral hemorrhage; mRS, modified Rankin Scale; RR, relative risk.
Adjusted for baseline ICH volume and time from onset to baseline non–contrast-enhanced computed tomography (NCCT). Hematoma expansion was defined as an increase of 33% or more in the hematoma volume from baseline to 24 hours.
Adjusted for age, baseline ICH volume, admission Glasgow Coma Scale score, and presence of intraventricular hemorrhage.
Defined as the presence of at least 1 focus of contrast extravasation in the hematoma with any size or morphologic feature and density of greater than 120 Hounsfield units.
Defined as intrahematoma contrast extravasation with a minimum diameter of 1.5 mm in any dimension, with spotlike or serpiginous shape, not connected with vessels outside the hematoma, without corresponding hyperdensity on NCCT, and attenuation at least double that of the background hematoma.
A secondary analysis including all patients with a CTA performed within 36 hours from onset (n = 166) showed the same results. Lowering of systolic BP below 140 mm Hg in patients with a spot sign did not significantly reduce the risk for hematoma expansion (relative risk, 0.68; 95% CI, 0.27-1.72; P = .41) and was not associated with improved long-term functional outcome (relative risk of mRS score, ≥4 at 90 days, 1.16; 95% CI, 0.57-2.36; P = .68).
We found no association between BP treatment and risk for hematoma growth also using a different definition of ICH expansion (relative hematoma growth >33% or absolute hematoma growth >6 mL). Finally, patients with a high spot sign score did not benefit from intensive BP reduction, as shown in eTable 3 in the Supplement.
Discussion
This prospective study of the CTA spot sign in ICH is, to our knowledge, the largest performed in terms of the number of hospitals involved and the largest study investigating the role of the CTA spot sign in selecting patients with ICH for antiexpansion treatment. We found that intensive BP lowering did not significantly reduce the rate of expansion and did not result in reduced mortality or severe disability in patients with a spot sign.
Some possible explanations for our findings include the following: (1) BP reduction does not truly influence hematoma expansion, and therefore a predictor of expansion would not help guide this therapy; (2) BP reduction reduces expansion, but the CTA spot sign is an inadequate tool to identify the patients most likely to benefit; and (3) the CTA spot sign could successfully perform this role, but we were underpowered to detect this effect in our analysis. For the first possibility, major trials have failed to convincingly demonstrate that BP lowering reduces hematoma expansion. The differences in actual BP achieved may have been inadequate to detect a true effect, but the process of ICH expansion is complex, and in most patients expansion may not be modifiable by BP reduction. For the second possibility, the suboptimal performance of the CTA spot sign in identifying patients at high risk for hematoma growth in this study may also explain our results. Compared with previous reports using the same diagnostic criteria, both definitions of spot sign showed lower sensitivity, specificity, and accuracy for prediction of ICH growth in our analysis. These discrepancies with previous studies may arise from heterogeneity and lack of standardization in the CTA acquisition protocol. The CTA acquisition variables may indeed influence the diagnostic performance of spot sign in identifying patients who will experience expansion. For the third possibility, we may have had inadequate statistical power to detect an effect of intensive BP reduction in patients at high risk for hematoma growth. All these potential explanations are not mutually exclusive, and the lack of statistical power precluded a thorough evaluation of the first 2 possibilities.
Finally, the cohort in this study showed some important differences from the general population with ICH. The patients included in our analysis were younger and had smaller hematomas, and more than half had a normal Glasgow Coma Scale score. Case fatality at 90 days was significantly lower (8.3%) compared with that of real-world studies in patients with ICH (30%-40%), leaving little room for outcome improvement with BP management (ceiling effect). In addition, the prevalence of the spot sign in this population was relatively high; using definition 1, we detected at least 1 spot sign in nearly 40% of the patients, whereas the prevalence of this imaging marker in previous studies using the same definition ranged from 17% to 23%. This difference may be owing to the fact that only those presenting within an early time frame (4.5 hours) were eligible for the ATACH-II trial, and the spot sign is more prevalent in early presenters despite the small baseline hematoma volume and the lack of oral anticoagulant treatment–associated ICH cases possibly having otherwise minimized spot sign detection. The great heterogeneity in the CTA acquisition protocol may also account for the high rate of spot sign–positive scans. The presence of poor-quality CTA images may indeed have led to the detection of false spot signs.
Our findings have important implications for future studies aimed at reducing hematoma growth in patients with ICH. First, if CTA is to be used, a protocol optimized for spot sign detection must be performed, which will likely help with accuracy and specificity. Second, we noted some differences in the diagnostic performance of the 2 spot sign definitions. Definition 1 had a higher sensitivity and therefore may be more suitable for selecting patients for antiexpansion treatments with minimal risk to capture as many patients as possible at high risk for hematoma growth. However, definition 2 appears to show higher specificity and may be preferred in future trials selecting patients for treatment with hemostatic therapies that may be associated with risks such as thrombotic complications. Several NCCT imaging markers have been recently reported as potentially associated with hematoma growth. Because of the low cost and wide availability, such NCCT findings may be an option for future studies to stratify the risk for ICH growth and select patients for expansion-targeted treatments.
Limitations
The main limitation of our study is the small proportion of patients who underwent CTA as part of their diagnostic workup. This analysis had an 11% power to detect a 10% difference in the risk for unfavorable outcomes between the two treatment groups at the P < .05 level. The presence of a ceiling effect owing to a mildly affected study population further reduced our power to detect a significant effect of BP treatment on ICH growth and outcome. In addition, the absolute difference in systolic BP in the 2 treatment arms was smaller compared with the expected values based on the suggested cutoffs (140 vs 180 mm Hg). The mean minimum systolic BP at 2 hours was 129 mm Hg in the intensive treatment group vs 141 mm Hg in the standard treatment arm. Finally, the delay between baseline NCCT and start of BP treatment and ICH expansion may have occurred in this time frame, especially in patients at high risk of ICH expansion (early presenters with a spot sign).
Conclusions
We report 2 major findings in this study. For the first time, to our knowledge, we have prospectively validated that a CTA spot sign is associated with hematoma expansion, but the diagnostic accuracy of this marker was lower compared with findings in previous single-center studies. Second, even in patients with a CTA spot sign (and thus a higher risk for expansion), we found no evidence that intensive BP reduction lowers the risk for expansion or improves outcome.
eFigure. Illustrative Comparison Between the 2 Different Spot Sign Definitions
eTable 1. Univariate Analysis With All CTA Before Randomization and Spot Sign Definition 1 (n = 101)
eTable 2. Univariate Analysis With All CTA Before Randomization and Spot Sign Definition 2 (n = 101)
eTable 3. Effect of Intensive Blood Pressure Treatment on ICH Expansion and Outcome, Stratified by Spot Sign Score
References
- 1.van Asch CJ, Luitse MJ, Rinkel GJ, van der Tweel I, Algra A, Klijn CJ. Incidence, case fatality, and functional outcome of intracerebral haemorrhage over time, according to age, sex, and ethnic origin: a systematic review and meta-analysis. Lancet Neurol. 2010;9(2):167-176. [DOI] [PubMed] [Google Scholar]
- 2.Broderick JP, Brott TG, Duldner JE, Tomsick T, Huster G. Volume of intracerebral hemorrhage: a powerful and easy-to-use predictor of 30-day mortality. Stroke. 1993;24(7):987-993. [DOI] [PubMed] [Google Scholar]
- 3.Brouwers HB, Chang Y, Falcone GJ, et al. Predicting hematoma expansion after primary intracerebral hemorrhage. JAMA Neurol. 2014;71(2):158-164. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Dowlatshahi D, Demchuk AM, Flaherty ML, Ali M, Lyden PL, Smith EE; VISTA Collaboration . Defining hematoma expansion in intracerebral hemorrhage: relationship with patient outcomes. Neurology. 2011;76(14):1238-1244. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Steiner T, Bösel J. Options to restrict hematoma expansion after spontaneous intracerebral hemorrhage. Stroke. 2010;41(2):402-409. [DOI] [PubMed] [Google Scholar]
- 6.Romero JM, Brouwers HB, Lu J, et al. Prospective validation of the computed tomographic angiography spot sign score for intracerebral hemorrhage. Stroke. 2013;44(11):3097-3102. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Goldstein JN, Fazen LE, Snider R, et al. Contrast extravasation on CT angiography predicts hematoma expansion in intracerebral hemorrhage. Neurology. 2007;68(12):889-894. [DOI] [PubMed] [Google Scholar]
- 8.Demchuk AM, Dowlatshahi D, Rodriguez-Luna D, et al. ; PREDICT/Sunnybrook ICH CTA study group . 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(4):307-314. [DOI] [PubMed] [Google Scholar]
- 9.Del Giudice A, D’Amico D, Sobesky J, Wellwood I. Accuracy of the spot sign on computed tomography angiography as a predictor of haematoma enlargement after acute spontaneous intracerebral haemorrhage: a systematic review. Cerebrovasc Dis. 2014;37(4):268-276. [DOI] [PubMed] [Google Scholar]
- 10.Goldstein J, Brouwers H, Romero J, et al. SCORE-IT: the spot sign score in restricting ICH growth—an ATACH-II ancillary study. J Vasc Interv Neurol. 2012;5(suppl):20-25. [PMC free article] [PubMed] [Google Scholar]
- 11.Qureshi AI, Palesch YY, Barsan WG, et al. ; ATACH-2 Trial Investigators and the Neurological Emergency Treatment Trials Network . Intensive blood-pressure lowering in patients with acute cerebral hemorrhage. N Engl J Med. 2016;375(11):1033-1043. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Qureshi AI, Palesch YY. Antihypertensive Treatment of Acute Cerebral Hemorrhage (ATACH) II: design, methods, and rationale. Neurocrit Care. 2011;15(3):559-576. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Thompson AL, Kosior JC, Gladstone DJ, et al. ; PREDICTS/Sunnybrook ICH CTA Study Group . Defining the CT angiography “spot sign” in primary intracerebral hemorrhage. Can J Neurol Sci. 2009;36(4):456-461. [DOI] [PubMed] [Google Scholar]
- 14.Gazzola S, Aviv RI, Gladstone DJ, et al. Vascular and nonvascular mimics of the CT angiography “spot sign” in patients with secondary intracerebral hemorrhage. Stroke. 2008;39(4):1177-1183. [DOI] [PubMed] [Google Scholar]
- 15.Dowlatshahi D, Brouwers HB, Demchuk AM, et al. Predicting intracerebral hemorrhage growth with the spot sign: the effect of onset-to-scan time. Stroke. 2016;47(3):695-700. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Pocock SJ, Simon R. Sequential treatment assignment with balancing for prognostic factors in the controlled clinical trial. Biometrics. 1975;31(1):103-115. [PubMed] [Google Scholar]
- 17.Hemphill JC III, Bonovich DC, Besmertis L, Manley GT, Johnston SC. The ICH score: a simple, reliable grading scale for intracerebral hemorrhage. Stroke. 2001;32(4):891-897. [DOI] [PubMed] [Google Scholar]
- 18.Anderson CS, Heeley E, Huang Y, et al. ; INTERACT2 Investigators . Rapid blood-pressure lowering in patients with acute intracerebral hemorrhage. N Engl J Med. 2013;368(25):2355-2365. [DOI] [PubMed] [Google Scholar]
- 19.Delgado Almandoz JE, Yoo AJ, Stone MJ, et al. Systematic characterization of the computed tomography angiography spot sign in primary intracerebral hemorrhage identifies patients at highest risk for hematoma expansion: the spot sign score. Stroke. 2009;40(9):2994-3000. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Waaijer A, Prokop M, Velthuis BK, Bakker CJG, de Kort GAP, van Leeuwen MS. Circle of Willis at CT angiography: dose reduction and image quality—reducing tube voltage and increasing tube current settings. Radiology. 2007;242(3):832-839. [DOI] [PubMed] [Google Scholar]
- 21.Chakraborty S, Alhazzaa M, Wasserman JK, et al. Dynamic characterization of the CT angiographic “spot sign”. PLoS One. 2014;9(3):e90431. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Morotti A, Romero JM, Jessel MJ, et al. Effect of CTA tube current on spot sign detection and accuracy for prediction of intracerebral hemorrhage expansion [published online May 19, 2016]. AJNR Am J Neuroradiol. doi: 10.3174/ajnr.A4810 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Delgado Almandoz JE, Yoo AJ, Stone MJ, et al. The spot sign score in primary intracerebral hemorrhage identifies patients at highest risk of in-hospital mortality and poor outcome among survivors. Stroke. 2010;41(1):54-60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Park SY, Kong MH, Kim JH, Kang DS, Song KY, Huh SK. Role of “spot sign” on CT angiography to predict hematoma expansion in spontaneous intracerebral hemorrhage. J Korean Neurosurg Soc. 2010;48(5):399-405. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Brouwers HB, Falcone GJ, McNamara KA, et al. CTA spot sign predicts hematoma expansion in patients with delayed presentation after intracerebral hemorrhage. Neurocrit Care. 2012;17(3):421-428. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Radmanesh F, Falcone GJ, Anderson CD, et al. Risk factors for computed tomography angiography spot sign in deep and lobar intracerebral hemorrhage are shared. Stroke. 2014;45(6):1833-1835. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Boulouis G, Morotti A, Brouwers HB, et al. Association between hypodensities detected by computed tomography and hematoma expansion in patients with intracerebral hemorrhage. JAMA Neurol. 2016;73(8):961-968. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Li Q, Zhang G, Huang Y-J, et al. Blend sign on computed tomography: novel and reliable predictor for early hematoma growth in patients with intracerebral hemorrhage. Stroke. 2015;46(8):2119-2123. [DOI] [PubMed] [Google Scholar]
- 29.Blacquiere D, Demchuk AM, Al-Hazzaa M, et al. ; PREDICT/Sunnybrook ICH CTA Study Group . Intracerebral hematoma morphologic appearance on noncontrast computed tomography predicts significant hematoma expansion. Stroke. 2015;46(11):3111-3116. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eFigure. Illustrative Comparison Between the 2 Different Spot Sign Definitions
eTable 1. Univariate Analysis With All CTA Before Randomization and Spot Sign Definition 1 (n = 101)
eTable 2. Univariate Analysis With All CTA Before Randomization and Spot Sign Definition 2 (n = 101)
eTable 3. Effect of Intensive Blood Pressure Treatment on ICH Expansion and Outcome, Stratified by Spot Sign Score