- Email: [email protected]
Determining the Optimal Shape-Related Indicator on Noncontrast Computed Tomography for Predicting Hematoma Expansion in Spontaneous Intracerebral Hemorrhage
Original Article
Determining the Optimal Shape-Related Indicator on Noncontrast Computed Tomography for Predicting Hematoma Expansion in Spontaneous Intracerebral Hemorrhage Zhiyuan Yu1, Jun Zheng1, Mou Li2, Xiaoze Wang2, Rui Guo1, Lu Ma1, Chao You1, Hao Li1
-
OBJECTIVE: Among several novel predictors on noncontrast computed tomography (CT) for hematoma expansion in spontaneous intracerebral hemorrhage (sICH), shape irregularity grade, satellite sign, and island sign are all related to irregular shape of hematoma. This study is aimed to compare the accuracy of these imaging markers for predicting hematoma expansion in the same cohort of sICH patients.
-
-
METHODS: This retrospective study enrolled sICH patients who underwent diagnostic computed tomography (CT) scans within 6 hours after onset and another follow-up CT scan within 24 hours after initial CT scan. Shape irregularity grade, satellite sign, and island sign were assessed according to the definitions in previous studies. The accuracy of these imaging indicators for predicting hematoma expansion was analyzed using receiver operator analysis.
INTRODUCTION
RESULTS: Finally, a total of 196 patients were included. Shape irregularity grade ‡3 was found in 87 (44.39%) patients, satellite sign was identified in 76 (38.78%) patients, and island sign was shown in only 41 (20.92%) patients. Only island sign remained an independent predictor for hematoma expansion in multivariate logistic regression. The sensitivity values of shape irregularity grade ‡3, satellite sign, and island sign were 0.52, 0.63, and 0.48, respectively. By contrast, the specificity values of these 3 predictors were 0.58, 0.69, and 0.85, respectively. Shape irregularity grade ‡3 had the smallest area under the curve (0.597), and island sign had the largest (0.676).
pontaneous intracerebral hemorrhage (sICH) is a devastating subtype of stroke with high mortality and disability.1 Although several clinical trials have been conducted in recent years, the management of sICH is still controversial.2-6 Hematoma expansion occurs in about one third of all sICH patients and is significantly related to worse outcome.7 Early identification of patients at high risk of hematoma expansion is critical for potential therapies.8 The spot sign on computed tomographic angiography (CTA) has been confirmed as a reliable marker for predicting hematoma expansion in several previous studies.9-12 However, some patients have only the results of noncontrast computed tomography (CT) in the first several hours after onset of sICH, and the CTA spot sign is not applicable in these patients. Several previous studies have been performed to look for predictors for hematoma expansion based on noncontrast CT.13-19 Among these predictors, shape irregularity grade, satellite sign, and island sign are all related to the shape of the hematoma, and their definitions partially overlap.9,17,18 Thus, it is important to determine the optimal shape-related predictor for hematoma expansion before building a reliable predicting scale for clinical practice. This study aimed to compare these 3 predictors based on hematoma shape and to try to determine the optimal predictor.
-
CONCLUSIONS: Island sign seems to be the optimal shape-related predictor for hematoma expansion in sICH patients and could be included in the future predictive model for hematoma expansion.
S
Key words Hematoma expansion - Island sign - Satellite sign - Shape irregularity - Spontaneous intracerebral hemorrhage
From the Departments of 1Neurosurgery and 2Radiology, West China Hospital, Sichuan University, Chengdu, Sichuan, China
Abbreviations and Acronyms AUC: Area under the curve CT: Computed tomography CTA: Computed tomographic angiography sICH: Spontaneous intracerebral hemorrhage
Journal homepage: www.WORLDNEUROSURGERY.org
-
WORLD NEUROSURGERY -: e1-e5, - 2018
To whom correspondence should be addressed: Hao Li, Ph.D. [E-mail: [email protected]] Citation: World Neurosurg. (2018). https://doi.org/10.1016/j.wneu.2018.09.172
Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2018 Elsevier Inc. All rights reserved.
www.WORLDNEUROSURGERY.org
e1
ORIGINAL ARTICLE ZHIYUAN YU ET AL.
PREDICTING HEMATOMA EXPANSION IN SPONTANEOUS INTRACEREBRAL HEMORRHAGE
MATERIAL AND METHODS Study Design This retrospective study was based on the prospective sICH database in our department and was approved by the local biomedical ethics committee. The inclusion criteria were as follows: 1) Age 18 years, 2) diagnostic CT conducted 6 hours after onset of sICH, and 3) follow-up CT conducted 6 hours after diagnostic CT. The exclusion criteria included 1) secondary intracerebral hemorrhage, 2) no diagnostic CT or follow-up CT, and 3) surgical hematoma evacuation before follow-up CT. Patients were treated according to the current guidelines.2,20 Data Collection Baseline data of all included patients were obtained from the database, including gender, age, admission blood pressure, important medical history, and admission laboratory test results about coagulation function. Imaging Analysis Diagnostic CT was conducted at the emergency department. Follow-up CT was performed 24 hours after the initial diagnostic CT to assess the progress of the disease. All CT scans were obtained and stored in Digital Imaging and Communications in Medicine format. Two assessors blinded to the clinical data analyzed the imaging data independently. All disagreements between 2 assessors were solved by consensus. According to the study by Barras et al.,19 the shape irregularity grade was evaluated using a 5-point scale, ranging from grade 1 (most regular) to grade 5 (most irregular). The shape irregularity was divided into regular (grades 1e2) and irregular (grades 3e5). The definition of satellite sign was any small hematoma separate from the main hematoma in 1 slice(s).21 Island sign was defined as follows: 1) 3 small hematomas completely separate from main hematoma or 2) 4 small hematomas separate from or partly connected to main hematoma.17 Hematoma volume was measured using ABC/2 method.22 Hematoma expansion was considered if >12.5 mL absolute increase or >33% relative increase of hematoma volume was observed on follow-up CT.23 Statistical Analysis Baseline features were compared between patients with and without hematoma expansion. Continuous data were shown as mean standard deviation and analyzed using t test. Discontinuous data were shown as median and interquartile range and analyzed using Wilcoxon rank sum test. Categoric data were analyzed using c2 analysis. Multivariable logistic regression was performed to evaluate the association between these shape-related predictors and hematoma expansion. The accuracy of these imaging indicators for predicting hematoma expansion was analyzed using receiver operator analysis, and the area under the curves (AUCs) of these predictors were compared by Z test. The interreader reliability for identifying these predictors was evaluated by k values. RESULTS A total of 196 sICH patients, from February 2015 to January 2017, were finally included in this study. The mean age was 59.57
e2
www.SCIENCEDIRECT.com
12.53 years (range, 29e87 years). The baseline hematoma volume ranged from 0.33 to 111.71 mL (mean, 25.73 22.07 mL). The hematoma location included basal ganglia 114 (58.16%), thalamus 19 (9.7%), lobar 34 (17.35%), cerebellum 10 (5.1%), and brainstem 19 (9.7%). Hematoma expansion was found in 46 (23.47%) patients. Table 1 shows the baseline features of sICH patients in this study. Shape irregularity grade 3 was found in 87 (44.39%) patients, and satellite sign was identified in 76 (38.78%) patients. By contrast, island sign was shown in only 41 (20.92%) patients. The k values for shape irregularity grade, satellite sign, and island sign were 0.824, 0.882, and 0.867, respectively, which suggested good interreader reliability for these predictors.
Table 1. Baseline Characteristics Characteristic Mean age (y) Sex, male
Expander (n [ 46)
Nonexpander (n [ 150)
P Value
61 13
59 12
0.439
32
109
0.682
Admission SBP (mm Hg)
176 30
170 28
0.237
Admission DBP (mm Hg)
103 16
99 18
0.144
Hypertension
28
79
0.328
Diabetes mellitus
5
12
0.545
Alcohol consumption
20
49
0.179
Smoking
18
52
0.580
Previous stroke
4
10
0.640
Anticoagulation use
7
13
0.199
PLT (109/L)
161 65
154 56
0.478
PT (seconds)
11.1 0.8
11.5 2.5
0.330
APTT (seconds)
27.7 2.7
27.3 5.5
0.669
INR
0.94 0.07
0.97 0.21
0.331
Time to CT (hours)
3.93 1.421
4.36 1.293
0.060
Hematoma volume (mL)
32.5 19.1
23.7 22.6
0.017
Shape irregularity grade
0.074
Grade 1
4
27
Grade 2
18
60
Grade 3
6
24
Grade 4
4
19
Grade 5
14
20
Shape irregularity grade 3
24
63
0.224
Satellite sign
29
47
<0.001
Island sign
22
19
<0.001
IVH
17
51
0.712
Data are mean standard deviation or number of patients. SBP, systolic blood pressure; DBP, diastolic blood pressure; PLT, platelet count; PT, prothrombin time; APTT, activated partial thromboplastin time; INR, international normalized ratio; CT, computed tomography; IVH, intraventricular hemorrhage.
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2018.09.172
ORIGINAL ARTICLE ZHIYUAN YU ET AL.
PREDICTING HEMATOMA EXPANSION IN SPONTANEOUS INTRACEREBRAL HEMORRHAGE
Although shape irregularity grade 3 occurred more frequently in patients with hematoma expansion (52.17%) than in those without hematoma expansion (42.00%), no significant difference was found (P ¼ 0.224). By contrast, the prevalence of satellite sign was significantly higher in patients who experienced hematoma expansion (P < 0.001). Moreover, island sign appeared significantly more frequently in patients with hematoma expansion (P < 0.001). However, only island sign remained an independent predictor for hematoma expansion in multivariate logistic regression (Table 2). The results of this study showed that the sensitivity, specificity, and positive and negative predictive values of shape irregularity grade 3 were 0.52, 0.58, 0.27, and 0.80, respectively. By contrast, the sensitivity, specificity, and positive and negative predictive values of satellite sign were 0.63, 0.69, 0.38, and 0.86, respectively. In addition, the sensitivity, specificity, and positive and negative predictive values of island sign were 0.48, 0.87, 0.54, and 0.85, respectively. Among 3 shape-related predictors for hematoma expansion, shape irregularity grade 3 had the smallest AUC (0.597), and island sign had the largest AUC (0.676). In addition, the AUC of satellite sign was 0.659. However, no significant difference was found among AUCs of these predictors (shape irregularity grade 3 vs. satellite sign, P ¼ 0.144; shape irregularity grade 3 vs. island sign, P ¼ 0.051; island sign vs. satellite sign, P ¼ 0.661) (Figure 1). DISCUSSION This study compared the accuracy of the current 3 shape-related indicators on noncontrast CT for predicting hematoma expansion in sICH patients. The results suggest that the island sign seems to be the optimal shape-related predictor for hematoma expansion. Shape irregularity grade was first reported by Barras et al.19 and measured by a 5-point scale. In their study, shape irregularity grade was not found as an independent predictor for hematoma expansion in sICH. Some other studies also discussed the association between shape irregularity grade and hematoma expansion based on this 5-point scale later. However, their results were inconsistent.13,15,24,25 A study by Takeda et al.24 did not demonstrate that shape irregularity grade was independently related to hematoma expansion. By contrast, shape irregularity
Table 2. Multivariate Analysis Hematoma Expansion Value
OR
95% CI
P Value
Time to CT
0.822
0.627e1.079
0.158
Hematoma volume
1.001
0.983e1.020
0.902
Island sign
4.341
1.840e10.245
0.001
Shape irregularity grade
0.972
0.712e1.328
0.860
Satellite sign
2.000
0.889e4.503
0.094
OR, odds ratio; CI, confidence interval; CT, computed tomography.
WORLD NEUROSURGERY -: e1-e5, - 2018
Figure 1. Receiver operating characteristic (ROC) curves of shape irregularity measured by 5-point scale, satellite sign, and island sign. The area under the curve (AUC) of shape irregularity measured by 5-point scale ¼ 0. 597, the AUC of satellite sign ¼ 0.659, and the AUC of island sign ¼ 0.676 (shape irregularity vs. satellite sign, P ¼ 0.144; shape irregularity vs. island sign, P ¼ 0.051; island sign vs. satellite sign, P ¼ 0.661).
grade was identified as an independent predictor for hematoma expansion in the other 3 studies.13,15,25 A previous meta-analysis suggested that shape irregularity could have a limited value for predicting hematoma expansion based on the pooled sensitivity and specificity.26 In this study, we found that the sensitivity and specificity of shape irregularity grade 3 were relatively low, and its AUC was also relatively small. Thus, although hematoma shape might be related to hematoma expansion, shape irregularity grade measured by the 5-point scale of Barras et al.19 could not be an ideal implement to predict hematoma expansion. Satellite sign was first defined by Shimoda et al.21 as any small hematoma separate from the main hematoma and considered a predictor for unfavorable outcome in sICH patients. Our previous study confirmed that satellite sign could also predict hematoma expansion in sICH patients.18 Based on its definition, satellite sign also reflected irregular hematoma shape. However, previous studies did not compare it with other shape-related imaging markers. The results of this study showed that the sensitivity and specificity of satellite sign were higher than those of shape irregularity grade 3. In addition, its AUC was also larger than that of shape irregularity grade 3. Thus, satellite sign seemed to be a better predictor for hematoma expansion than shape irregularity grade. However, its specificity was lower than island sign, and its AUC was also smaller than island sign. Thus, satellite sign was not the optimal shape-related predictor for hematoma expansion. Island sign was a novel imaging indictor for predicting hematoma expansion and reflected the extremely irregular shape of hematoma, which was first reported by Li et al.17 In their study, island sign was confirmed as an independent indicator for predicting hematoma expansion, with ideal specificity (98.2%).17 However, the predictive value of island sign was not compared
www.WORLDNEUROSURGERY.org
e3
ORIGINAL ARTICLE ZHIYUAN YU ET AL.
PREDICTING HEMATOMA EXPANSION IN SPONTANEOUS INTRACEREBRAL HEMORRHAGE
with that of other predictors related to hematoma shape. In this study, we compared island sign with the other 2 shape-related predictors, shape irregularity grade and satellite sign. Although its sensitivity was lower than those of the other 2 predictors, it had the highest specificity. Moreover, island sign was the only independent predictor for hematoma expansion in multivariate logistic regression. In addition, island sign had the largest AUC among 3 shape-related predictors. Thus, island sign was the optimal predictor among the current 3 shape-related predictors for hematoma expansion. In recent years, several novel predictors on noncontrast CT have been reported. These new predictors usually reflect irregular shape (such as satellite sign and island sign) or heterogeneous density (such as blend sign, black hole sign, and hypodensities) of hematoma.14-18 For the better use of these findings in clinical practice, it is very important to develop a comprehensive predictive model that includes these imaging predictors. However, considering the similar definitions of these predictors, it is difficult to choose the eligible predictors. Thus, it is important to compare these similar imaging markers describing the same feature of hematoma. In our study, we found that island sign could be the optimal shape-related predictor for hematoma expansion, which could be used in the comprehensive predictive model base on noncontrast CT. Recently, Morotti et al.27 developed a new 5-point scale for predicting hematoma expansion, including blend sign, any hypodensities, and onset-to-CT time, which showed good
REFERENCES 1. Qureshi AI, Mendelow AD, Hanley DF. Intracerebral haemorrhage. Lancet. 2009;373:1632-1644. 2. Hemphill JC 3rd, 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;46:2032-2060. 3. Anderson CS, 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;368:2355-2365. 4. Qureshi AI, Palesch YY, Barsan WG, Hanley DF, Hsu CY, Martin RL, et al. Intensive bloodpressure lowering in patients with acute cerebral hemorrhage. N Engl J Med. 2016;375:1033-1043. 5. Hanley DF, Lane K, McBee N, Ziai W, Tuhrim S, Lees KR, et al. Thrombolytic removal of intraventricular haemorrhage in treatment of severe stroke: results of the randomised, multicentre, multiregion, placebo-controlled CLEAR III trial. Lancet. 2017;389:603-611. 6. Hanley DF, Thompson RE, Muschelli J, Rosenblum M, McBee N, Lane K, et al. Safety and efficacy of minimally invasive surgery plus alteplase in intracerebral haemorrhage evacuation (MISTIE): A randomised, controlled, open-label, phase 2 trial. Lancet Neurol. 2016;15:1228-1237.
e4
www.SCIENCEDIRECT.com
specificity. However, in their study, only 1 shape-related predictor, shape irregularity grade, was analyzed, and no shape-related predictor was included in the final scale. Considering the good performance of island sign in predicting hematoma expansion, further predictive models of hematoma expansion should include this important indicator reflecting irregular shape of hematoma. There are still several limitations in this study. First, it was a retrospective study in a single center and had a relatively small sample size. Second, the previous study suggested that hematoma volume could be associated with hematoma expansion.7 Although it was not independently related to hematoma expansion in multivariate analysis in our study, it could have been caused by the small sample size. Moreover, the association between these predictors and outcome in sICH patients was not discussed because of insufficient data. Furthermore, this study included only the currently reported shape-related predictors on noncontrast CT, and we are still unsure whether any better shaperelated predictor can be developed in the future. Further studies are needed to confirm our findings in this study. CONCLUSIONS The study compared the current 3 shape-related imaging markers on noncontrast CT for predicting hematoma expansion. Island sign seems to be the optimal shape-related predictor for hematoma expansion in sICH patients.
7. Brouwers HB, Chang Y, Falcone GJ, Cai X, Ayres AM, Battey TW, et al. Predicting hematoma expansion after primary intracerebral hemorrhage. JAMA Neurol. 2014;71:158-164. 8. Boulouis G, Dumas A, Betensky RA, Brouwers HB, Fotiadis P, Vashkevich A, et al. Anatomic pattern of intracerebral hemorrhage expansion: Relation to CT angiography spot sign and hematoma center. Stroke. 2014;45:1154-1156. 9. Wada R, Aviv RI, Fox AJ, Sahlas DJ, Gladstone DJ, Tomlinson G, et al. CT angiography "spot sign" predicts hematoma expansion in acute intracerebral hemorrhage. Stroke. 2007;38:1257-1262.
significant hematoma expansion. Stroke. 2015;46: 3111-3116. 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;46:2119-2123. 15. 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; 73:961-968.
10. 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 CTangiography spot sign (PREDICT): A prospective observational study. Lancet Neurol. 2012;11:307-314.
16. Li Q, Zhang G, Xiong X, Wang XC, Yang WS, Li KW, et al. Black hole sign: Novel imaging marker that predicts hematoma growth in patients with intracerebral hemorrhage. Stroke. 2016;47: 1777-1781.
11. Du FZ, Jiang R, Gu M, He C, Guan J. The accuracy of spot sign in predicting hematoma expansion after intracerebral hemorrhage: A systematic review and meta-analysis. PLoS One. 2014;9:e115777.
17. Li Q, Liu QJ, Yang WS, Wang XC, Zhao LB, Xiong X, et al. Island sign: An imaging predictor for early hematoma expansion and poor outcome in patients with intracerebral hemorrhage. Stroke. 2017;48:3019-3025.
12. Han JH, Lee JM, Koh EJ, Choi HY. The spot sign predicts hematoma expansion, outcome, and mortality in patients with primary intracerebral hemorrhage. J Korean Neurosurg Soc. 2014;56: 303-309. 13. Blacquiere D, Demchuk AM, Al-Hazzaa M, Deshpande A, Petrcich W, Aviv RI, et al. Intracerebral hematoma morphologic appearance on noncontrast computed tomography predicts
18. Yu Z, Zheng J, Ali H, Guo R, Li M, Wang X, et al. Significance of satellite sign and spot sign in predicting hematoma expansion in spontaneous intracerebral hemorrhage. Clin Neurol Neurosurg. 2017;162:67-71. 19. Barras CD, Tress BM, Christensen S, MacGregor L, Collins M, Desmond PM, et al. Density and shape as CT predictors of
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2018.09.172
ORIGINAL ARTICLE ZHIYUAN YU ET AL.
PREDICTING HEMATOMA EXPANSION IN SPONTANEOUS INTRACEREBRAL HEMORRHAGE
intracerebral hemorrhage growth. Stroke. 2009;40: 1325-1331.
after intracerebral hemorrhage. Neurology. 2006; 66:1175-1181.
20. Steiner T, Al-Shahi Salman R, Beer R, Christensen H, Cordonnier C, Csiba L, et al. European Stroke Organisation (ESO) guidelines for the management of spontaneous intracerebral hemorrhage. Int J Stroke. 2014;9:840-855.
24. Takeda R, Ogura T, Ooigawa H, Fushihara G, Yoshikawa S, Okada D, et al. A practical prediction model for early hematoma expansion in spontaneous deep ganglionic intracerebral hemorrhage. Clin Neurol Neurosurg. 2013;115:1028-1031.
21. Shimoda Y, Ohtomo S, Arai H, Okada K, Tominaga T. Satellite sign: A poor outcome predictor in intracerebral hemorrhage. Cerebrovasc Dis. 2017;44:105-112.
25. 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; 89:548-554.
22. Kothari RU, Brott T, Broderick JP, Barsan WG, Sauerbeck LR, Zuccarello M, et al. The ABCs of measuring intracerebral hemorrhage volumes. Stroke. 1996;27:1304-1305. 23. Davis SM, Broderick J, Hennerici M, Brun NC, Diringer MN, Mayer SA, et al. Hematoma growth is a determinant of mortality and poor outcome
26. Yu Z, Zheng J, Xu Z, Li M, Wang X, Lin S, et al. 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. World Neurosurg. 2017; 108:347-355.
WORLD NEUROSURGERY -: e1-e5, - 2018
27. Morotti A, Dowlatshahi D, Boulouis G, Al-Ajlan F, Demchuk AM, Aviv RI, et al. Predicting intracerebral hemorrhage expansion with noncontrast computed tomography: The BAT score. Stroke. 2018;49:1163-1169.
Conflict of interest statement: This work was supported by Outstanding Subject Development 135 Project of West China Hospital, Sichuan University (grant number ZY2016102). Received 27 May 2018; accepted 24 September 2018 Citation: World Neurosurg. (2018). https://doi.org/10.1016/j.wneu.2018.09.172 Journal homepage: www.WORLDNEUROSURGERY.org Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2018 Elsevier Inc. All rights reserved.
www.WORLDNEUROSURGERY.org
e5
Determining the Optimal Shape-Related Indicator on Noncontrast Computed Tomography for Predicting Hematoma Expansion in Spontaneous Intracerebral Hemorrhage Zhiyuan Yu1, Jun Zheng1, Mou Li2, Xiaoze Wang2, Rui Guo1, Lu Ma1, Chao You1, Hao Li1
-
OBJECTIVE: Among several novel predictors on noncontrast computed tomography (CT) for hematoma expansion in spontaneous intracerebral hemorrhage (sICH), shape irregularity grade, satellite sign, and island sign are all related to irregular shape of hematoma. This study is aimed to compare the accuracy of these imaging markers for predicting hematoma expansion in the same cohort of sICH patients.
-
-
METHODS: This retrospective study enrolled sICH patients who underwent diagnostic computed tomography (CT) scans within 6 hours after onset and another follow-up CT scan within 24 hours after initial CT scan. Shape irregularity grade, satellite sign, and island sign were assessed according to the definitions in previous studies. The accuracy of these imaging indicators for predicting hematoma expansion was analyzed using receiver operator analysis.
INTRODUCTION
RESULTS: Finally, a total of 196 patients were included. Shape irregularity grade ‡3 was found in 87 (44.39%) patients, satellite sign was identified in 76 (38.78%) patients, and island sign was shown in only 41 (20.92%) patients. Only island sign remained an independent predictor for hematoma expansion in multivariate logistic regression. The sensitivity values of shape irregularity grade ‡3, satellite sign, and island sign were 0.52, 0.63, and 0.48, respectively. By contrast, the specificity values of these 3 predictors were 0.58, 0.69, and 0.85, respectively. Shape irregularity grade ‡3 had the smallest area under the curve (0.597), and island sign had the largest (0.676).
pontaneous intracerebral hemorrhage (sICH) is a devastating subtype of stroke with high mortality and disability.1 Although several clinical trials have been conducted in recent years, the management of sICH is still controversial.2-6 Hematoma expansion occurs in about one third of all sICH patients and is significantly related to worse outcome.7 Early identification of patients at high risk of hematoma expansion is critical for potential therapies.8 The spot sign on computed tomographic angiography (CTA) has been confirmed as a reliable marker for predicting hematoma expansion in several previous studies.9-12 However, some patients have only the results of noncontrast computed tomography (CT) in the first several hours after onset of sICH, and the CTA spot sign is not applicable in these patients. Several previous studies have been performed to look for predictors for hematoma expansion based on noncontrast CT.13-19 Among these predictors, shape irregularity grade, satellite sign, and island sign are all related to the shape of the hematoma, and their definitions partially overlap.9,17,18 Thus, it is important to determine the optimal shape-related predictor for hematoma expansion before building a reliable predicting scale for clinical practice. This study aimed to compare these 3 predictors based on hematoma shape and to try to determine the optimal predictor.
-
CONCLUSIONS: Island sign seems to be the optimal shape-related predictor for hematoma expansion in sICH patients and could be included in the future predictive model for hematoma expansion.
S
Key words Hematoma expansion - Island sign - Satellite sign - Shape irregularity - Spontaneous intracerebral hemorrhage
From the Departments of 1Neurosurgery and 2Radiology, West China Hospital, Sichuan University, Chengdu, Sichuan, China
Abbreviations and Acronyms AUC: Area under the curve CT: Computed tomography CTA: Computed tomographic angiography sICH: Spontaneous intracerebral hemorrhage
Journal homepage: www.WORLDNEUROSURGERY.org
-
WORLD NEUROSURGERY -: e1-e5, - 2018
To whom correspondence should be addressed: Hao Li, Ph.D. [E-mail: [email protected]] Citation: World Neurosurg. (2018). https://doi.org/10.1016/j.wneu.2018.09.172
Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2018 Elsevier Inc. All rights reserved.
www.WORLDNEUROSURGERY.org
e1
ORIGINAL ARTICLE ZHIYUAN YU ET AL.
PREDICTING HEMATOMA EXPANSION IN SPONTANEOUS INTRACEREBRAL HEMORRHAGE
MATERIAL AND METHODS Study Design This retrospective study was based on the prospective sICH database in our department and was approved by the local biomedical ethics committee. The inclusion criteria were as follows: 1) Age 18 years, 2) diagnostic CT conducted 6 hours after onset of sICH, and 3) follow-up CT conducted 6 hours after diagnostic CT. The exclusion criteria included 1) secondary intracerebral hemorrhage, 2) no diagnostic CT or follow-up CT, and 3) surgical hematoma evacuation before follow-up CT. Patients were treated according to the current guidelines.2,20 Data Collection Baseline data of all included patients were obtained from the database, including gender, age, admission blood pressure, important medical history, and admission laboratory test results about coagulation function. Imaging Analysis Diagnostic CT was conducted at the emergency department. Follow-up CT was performed 24 hours after the initial diagnostic CT to assess the progress of the disease. All CT scans were obtained and stored in Digital Imaging and Communications in Medicine format. Two assessors blinded to the clinical data analyzed the imaging data independently. All disagreements between 2 assessors were solved by consensus. According to the study by Barras et al.,19 the shape irregularity grade was evaluated using a 5-point scale, ranging from grade 1 (most regular) to grade 5 (most irregular). The shape irregularity was divided into regular (grades 1e2) and irregular (grades 3e5). The definition of satellite sign was any small hematoma separate from the main hematoma in 1 slice(s).21 Island sign was defined as follows: 1) 3 small hematomas completely separate from main hematoma or 2) 4 small hematomas separate from or partly connected to main hematoma.17 Hematoma volume was measured using ABC/2 method.22 Hematoma expansion was considered if >12.5 mL absolute increase or >33% relative increase of hematoma volume was observed on follow-up CT.23 Statistical Analysis Baseline features were compared between patients with and without hematoma expansion. Continuous data were shown as mean standard deviation and analyzed using t test. Discontinuous data were shown as median and interquartile range and analyzed using Wilcoxon rank sum test. Categoric data were analyzed using c2 analysis. Multivariable logistic regression was performed to evaluate the association between these shape-related predictors and hematoma expansion. The accuracy of these imaging indicators for predicting hematoma expansion was analyzed using receiver operator analysis, and the area under the curves (AUCs) of these predictors were compared by Z test. The interreader reliability for identifying these predictors was evaluated by k values. RESULTS A total of 196 sICH patients, from February 2015 to January 2017, were finally included in this study. The mean age was 59.57
e2
www.SCIENCEDIRECT.com
12.53 years (range, 29e87 years). The baseline hematoma volume ranged from 0.33 to 111.71 mL (mean, 25.73 22.07 mL). The hematoma location included basal ganglia 114 (58.16%), thalamus 19 (9.7%), lobar 34 (17.35%), cerebellum 10 (5.1%), and brainstem 19 (9.7%). Hematoma expansion was found in 46 (23.47%) patients. Table 1 shows the baseline features of sICH patients in this study. Shape irregularity grade 3 was found in 87 (44.39%) patients, and satellite sign was identified in 76 (38.78%) patients. By contrast, island sign was shown in only 41 (20.92%) patients. The k values for shape irregularity grade, satellite sign, and island sign were 0.824, 0.882, and 0.867, respectively, which suggested good interreader reliability for these predictors.
Table 1. Baseline Characteristics Characteristic Mean age (y) Sex, male
Expander (n [ 46)
Nonexpander (n [ 150)
P Value
61 13
59 12
0.439
32
109
0.682
Admission SBP (mm Hg)
176 30
170 28
0.237
Admission DBP (mm Hg)
103 16
99 18
0.144
Hypertension
28
79
0.328
Diabetes mellitus
5
12
0.545
Alcohol consumption
20
49
0.179
Smoking
18
52
0.580
Previous stroke
4
10
0.640
Anticoagulation use
7
13
0.199
PLT (109/L)
161 65
154 56
0.478
PT (seconds)
11.1 0.8
11.5 2.5
0.330
APTT (seconds)
27.7 2.7
27.3 5.5
0.669
INR
0.94 0.07
0.97 0.21
0.331
Time to CT (hours)
3.93 1.421
4.36 1.293
0.060
Hematoma volume (mL)
32.5 19.1
23.7 22.6
0.017
Shape irregularity grade
0.074
Grade 1
4
27
Grade 2
18
60
Grade 3
6
24
Grade 4
4
19
Grade 5
14
20
Shape irregularity grade 3
24
63
0.224
Satellite sign
29
47
<0.001
Island sign
22
19
<0.001
IVH
17
51
0.712
Data are mean standard deviation or number of patients. SBP, systolic blood pressure; DBP, diastolic blood pressure; PLT, platelet count; PT, prothrombin time; APTT, activated partial thromboplastin time; INR, international normalized ratio; CT, computed tomography; IVH, intraventricular hemorrhage.
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2018.09.172
ORIGINAL ARTICLE ZHIYUAN YU ET AL.
PREDICTING HEMATOMA EXPANSION IN SPONTANEOUS INTRACEREBRAL HEMORRHAGE
Although shape irregularity grade 3 occurred more frequently in patients with hematoma expansion (52.17%) than in those without hematoma expansion (42.00%), no significant difference was found (P ¼ 0.224). By contrast, the prevalence of satellite sign was significantly higher in patients who experienced hematoma expansion (P < 0.001). Moreover, island sign appeared significantly more frequently in patients with hematoma expansion (P < 0.001). However, only island sign remained an independent predictor for hematoma expansion in multivariate logistic regression (Table 2). The results of this study showed that the sensitivity, specificity, and positive and negative predictive values of shape irregularity grade 3 were 0.52, 0.58, 0.27, and 0.80, respectively. By contrast, the sensitivity, specificity, and positive and negative predictive values of satellite sign were 0.63, 0.69, 0.38, and 0.86, respectively. In addition, the sensitivity, specificity, and positive and negative predictive values of island sign were 0.48, 0.87, 0.54, and 0.85, respectively. Among 3 shape-related predictors for hematoma expansion, shape irregularity grade 3 had the smallest AUC (0.597), and island sign had the largest AUC (0.676). In addition, the AUC of satellite sign was 0.659. However, no significant difference was found among AUCs of these predictors (shape irregularity grade 3 vs. satellite sign, P ¼ 0.144; shape irregularity grade 3 vs. island sign, P ¼ 0.051; island sign vs. satellite sign, P ¼ 0.661) (Figure 1). DISCUSSION This study compared the accuracy of the current 3 shape-related indicators on noncontrast CT for predicting hematoma expansion in sICH patients. The results suggest that the island sign seems to be the optimal shape-related predictor for hematoma expansion. Shape irregularity grade was first reported by Barras et al.19 and measured by a 5-point scale. In their study, shape irregularity grade was not found as an independent predictor for hematoma expansion in sICH. Some other studies also discussed the association between shape irregularity grade and hematoma expansion based on this 5-point scale later. However, their results were inconsistent.13,15,24,25 A study by Takeda et al.24 did not demonstrate that shape irregularity grade was independently related to hematoma expansion. By contrast, shape irregularity
Table 2. Multivariate Analysis Hematoma Expansion Value
OR
95% CI
P Value
Time to CT
0.822
0.627e1.079
0.158
Hematoma volume
1.001
0.983e1.020
0.902
Island sign
4.341
1.840e10.245
0.001
Shape irregularity grade
0.972
0.712e1.328
0.860
Satellite sign
2.000
0.889e4.503
0.094
OR, odds ratio; CI, confidence interval; CT, computed tomography.
WORLD NEUROSURGERY -: e1-e5, - 2018
Figure 1. Receiver operating characteristic (ROC) curves of shape irregularity measured by 5-point scale, satellite sign, and island sign. The area under the curve (AUC) of shape irregularity measured by 5-point scale ¼ 0. 597, the AUC of satellite sign ¼ 0.659, and the AUC of island sign ¼ 0.676 (shape irregularity vs. satellite sign, P ¼ 0.144; shape irregularity vs. island sign, P ¼ 0.051; island sign vs. satellite sign, P ¼ 0.661).
grade was identified as an independent predictor for hematoma expansion in the other 3 studies.13,15,25 A previous meta-analysis suggested that shape irregularity could have a limited value for predicting hematoma expansion based on the pooled sensitivity and specificity.26 In this study, we found that the sensitivity and specificity of shape irregularity grade 3 were relatively low, and its AUC was also relatively small. Thus, although hematoma shape might be related to hematoma expansion, shape irregularity grade measured by the 5-point scale of Barras et al.19 could not be an ideal implement to predict hematoma expansion. Satellite sign was first defined by Shimoda et al.21 as any small hematoma separate from the main hematoma and considered a predictor for unfavorable outcome in sICH patients. Our previous study confirmed that satellite sign could also predict hematoma expansion in sICH patients.18 Based on its definition, satellite sign also reflected irregular hematoma shape. However, previous studies did not compare it with other shape-related imaging markers. The results of this study showed that the sensitivity and specificity of satellite sign were higher than those of shape irregularity grade 3. In addition, its AUC was also larger than that of shape irregularity grade 3. Thus, satellite sign seemed to be a better predictor for hematoma expansion than shape irregularity grade. However, its specificity was lower than island sign, and its AUC was also smaller than island sign. Thus, satellite sign was not the optimal shape-related predictor for hematoma expansion. Island sign was a novel imaging indictor for predicting hematoma expansion and reflected the extremely irregular shape of hematoma, which was first reported by Li et al.17 In their study, island sign was confirmed as an independent indicator for predicting hematoma expansion, with ideal specificity (98.2%).17 However, the predictive value of island sign was not compared
www.WORLDNEUROSURGERY.org
e3
ORIGINAL ARTICLE ZHIYUAN YU ET AL.
PREDICTING HEMATOMA EXPANSION IN SPONTANEOUS INTRACEREBRAL HEMORRHAGE
with that of other predictors related to hematoma shape. In this study, we compared island sign with the other 2 shape-related predictors, shape irregularity grade and satellite sign. Although its sensitivity was lower than those of the other 2 predictors, it had the highest specificity. Moreover, island sign was the only independent predictor for hematoma expansion in multivariate logistic regression. In addition, island sign had the largest AUC among 3 shape-related predictors. Thus, island sign was the optimal predictor among the current 3 shape-related predictors for hematoma expansion. In recent years, several novel predictors on noncontrast CT have been reported. These new predictors usually reflect irregular shape (such as satellite sign and island sign) or heterogeneous density (such as blend sign, black hole sign, and hypodensities) of hematoma.14-18 For the better use of these findings in clinical practice, it is very important to develop a comprehensive predictive model that includes these imaging predictors. However, considering the similar definitions of these predictors, it is difficult to choose the eligible predictors. Thus, it is important to compare these similar imaging markers describing the same feature of hematoma. In our study, we found that island sign could be the optimal shape-related predictor for hematoma expansion, which could be used in the comprehensive predictive model base on noncontrast CT. Recently, Morotti et al.27 developed a new 5-point scale for predicting hematoma expansion, including blend sign, any hypodensities, and onset-to-CT time, which showed good
REFERENCES 1. Qureshi AI, Mendelow AD, Hanley DF. Intracerebral haemorrhage. Lancet. 2009;373:1632-1644. 2. Hemphill JC 3rd, 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;46:2032-2060. 3. Anderson CS, 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;368:2355-2365. 4. Qureshi AI, Palesch YY, Barsan WG, Hanley DF, Hsu CY, Martin RL, et al. Intensive bloodpressure lowering in patients with acute cerebral hemorrhage. N Engl J Med. 2016;375:1033-1043. 5. Hanley DF, Lane K, McBee N, Ziai W, Tuhrim S, Lees KR, et al. Thrombolytic removal of intraventricular haemorrhage in treatment of severe stroke: results of the randomised, multicentre, multiregion, placebo-controlled CLEAR III trial. Lancet. 2017;389:603-611. 6. Hanley DF, Thompson RE, Muschelli J, Rosenblum M, McBee N, Lane K, et al. Safety and efficacy of minimally invasive surgery plus alteplase in intracerebral haemorrhage evacuation (MISTIE): A randomised, controlled, open-label, phase 2 trial. Lancet Neurol. 2016;15:1228-1237.
e4
www.SCIENCEDIRECT.com
specificity. However, in their study, only 1 shape-related predictor, shape irregularity grade, was analyzed, and no shape-related predictor was included in the final scale. Considering the good performance of island sign in predicting hematoma expansion, further predictive models of hematoma expansion should include this important indicator reflecting irregular shape of hematoma. There are still several limitations in this study. First, it was a retrospective study in a single center and had a relatively small sample size. Second, the previous study suggested that hematoma volume could be associated with hematoma expansion.7 Although it was not independently related to hematoma expansion in multivariate analysis in our study, it could have been caused by the small sample size. Moreover, the association between these predictors and outcome in sICH patients was not discussed because of insufficient data. Furthermore, this study included only the currently reported shape-related predictors on noncontrast CT, and we are still unsure whether any better shaperelated predictor can be developed in the future. Further studies are needed to confirm our findings in this study. CONCLUSIONS The study compared the current 3 shape-related imaging markers on noncontrast CT for predicting hematoma expansion. Island sign seems to be the optimal shape-related predictor for hematoma expansion in sICH patients.
7. Brouwers HB, Chang Y, Falcone GJ, Cai X, Ayres AM, Battey TW, et al. Predicting hematoma expansion after primary intracerebral hemorrhage. JAMA Neurol. 2014;71:158-164. 8. Boulouis G, Dumas A, Betensky RA, Brouwers HB, Fotiadis P, Vashkevich A, et al. Anatomic pattern of intracerebral hemorrhage expansion: Relation to CT angiography spot sign and hematoma center. Stroke. 2014;45:1154-1156. 9. Wada R, Aviv RI, Fox AJ, Sahlas DJ, Gladstone DJ, Tomlinson G, et al. CT angiography "spot sign" predicts hematoma expansion in acute intracerebral hemorrhage. Stroke. 2007;38:1257-1262.
significant hematoma expansion. Stroke. 2015;46: 3111-3116. 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;46:2119-2123. 15. 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; 73:961-968.
10. 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 CTangiography spot sign (PREDICT): A prospective observational study. Lancet Neurol. 2012;11:307-314.
16. Li Q, Zhang G, Xiong X, Wang XC, Yang WS, Li KW, et al. Black hole sign: Novel imaging marker that predicts hematoma growth in patients with intracerebral hemorrhage. Stroke. 2016;47: 1777-1781.
11. Du FZ, Jiang R, Gu M, He C, Guan J. The accuracy of spot sign in predicting hematoma expansion after intracerebral hemorrhage: A systematic review and meta-analysis. PLoS One. 2014;9:e115777.
17. Li Q, Liu QJ, Yang WS, Wang XC, Zhao LB, Xiong X, et al. Island sign: An imaging predictor for early hematoma expansion and poor outcome in patients with intracerebral hemorrhage. Stroke. 2017;48:3019-3025.
12. Han JH, Lee JM, Koh EJ, Choi HY. The spot sign predicts hematoma expansion, outcome, and mortality in patients with primary intracerebral hemorrhage. J Korean Neurosurg Soc. 2014;56: 303-309. 13. Blacquiere D, Demchuk AM, Al-Hazzaa M, Deshpande A, Petrcich W, Aviv RI, et al. Intracerebral hematoma morphologic appearance on noncontrast computed tomography predicts
18. Yu Z, Zheng J, Ali H, Guo R, Li M, Wang X, et al. Significance of satellite sign and spot sign in predicting hematoma expansion in spontaneous intracerebral hemorrhage. Clin Neurol Neurosurg. 2017;162:67-71. 19. Barras CD, Tress BM, Christensen S, MacGregor L, Collins M, Desmond PM, et al. Density and shape as CT predictors of
WORLD NEUROSURGERY, https://doi.org/10.1016/j.wneu.2018.09.172
ORIGINAL ARTICLE ZHIYUAN YU ET AL.
PREDICTING HEMATOMA EXPANSION IN SPONTANEOUS INTRACEREBRAL HEMORRHAGE
intracerebral hemorrhage growth. Stroke. 2009;40: 1325-1331.
after intracerebral hemorrhage. Neurology. 2006; 66:1175-1181.
20. Steiner T, Al-Shahi Salman R, Beer R, Christensen H, Cordonnier C, Csiba L, et al. European Stroke Organisation (ESO) guidelines for the management of spontaneous intracerebral hemorrhage. Int J Stroke. 2014;9:840-855.
24. Takeda R, Ogura T, Ooigawa H, Fushihara G, Yoshikawa S, Okada D, et al. A practical prediction model for early hematoma expansion in spontaneous deep ganglionic intracerebral hemorrhage. Clin Neurol Neurosurg. 2013;115:1028-1031.
21. Shimoda Y, Ohtomo S, Arai H, Okada K, Tominaga T. Satellite sign: A poor outcome predictor in intracerebral hemorrhage. Cerebrovasc Dis. 2017;44:105-112.
25. 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; 89:548-554.
22. Kothari RU, Brott T, Broderick JP, Barsan WG, Sauerbeck LR, Zuccarello M, et al. The ABCs of measuring intracerebral hemorrhage volumes. Stroke. 1996;27:1304-1305. 23. Davis SM, Broderick J, Hennerici M, Brun NC, Diringer MN, Mayer SA, et al. Hematoma growth is a determinant of mortality and poor outcome
26. Yu Z, Zheng J, Xu Z, Li M, Wang X, Lin S, et al. 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. World Neurosurg. 2017; 108:347-355.
WORLD NEUROSURGERY -: e1-e5, - 2018
27. Morotti A, Dowlatshahi D, Boulouis G, Al-Ajlan F, Demchuk AM, Aviv RI, et al. Predicting intracerebral hemorrhage expansion with noncontrast computed tomography: The BAT score. Stroke. 2018;49:1163-1169.
Conflict of interest statement: This work was supported by Outstanding Subject Development 135 Project of West China Hospital, Sichuan University (grant number ZY2016102). Received 27 May 2018; accepted 24 September 2018 Citation: World Neurosurg. (2018). https://doi.org/10.1016/j.wneu.2018.09.172 Journal homepage: www.WORLDNEUROSURGERY.org Available online: www.sciencedirect.com 1878-8750/$ - see front matter ª 2018 Elsevier Inc. All rights reserved.
www.WORLDNEUROSURGERY.org
e5