Sage Journals: Discover world-class research

Abstract

Introduction:

Prediction scores for hematoma expansion in spontaneous intracerebral hemorrhage (ICH), such as the 9-point and BRAIN scores, were developed predominantly using planimetry to measure hematoma volume. In this study, we aim to investigate whether the ABC/2 formula, which is known to overestimate hematoma volume, can be reliably used as a substitute for planimetry in these prediction scores.

Patients and methods:

A total of 429 patients from four hospitals were retrospectively enrolled. CT scan and clinical data at admission and follow-up CT scan were collected. The 9-point and BRAIN scores were calculated using hematoma volume from ABC/2 and planimetry. Hematoma expansion was assessed using hematoma volume from planimetry.

Results:

The median hematoma volume measured by ABC/2 was 11.97 ml (interquartile range [IQR], 4.8–30.0), whereas the volume measured by planimetry was 11.70 ml (IQR, 4.9–26.6). The median measurement error between ABC/2 and planimetry was 0.30 ml (IQR, −0.72–2.87). ABC/2 overestimated hematoma volume in 244 patients (56.9%) compared to planimetry. In the 9-point score, the area under the curves (AUCs) for predicting hematoma expansion were 0.735 (95% confidence interval [CI], 0.675–0.796) with ABC/2 and 0.732 (95% CI, 0.672–0.793) with planimetry. In the BRAIN score, the AUCs were 0.753 (95% CI, 0.693–0.813) with ABC/2 and 0.745 (95% CI, 0.688–0.803) with planimetry.

Discussion and conclusion:

The 9-point and BRAIN scores using hematoma volume measured by ABC/2 and planimetry showed good performance in predicting hematoma expansion in ICH. ABC/2 volumetric estimation proved to be reliable for these scores.

Graphical abstract

Keywords

Intracerebral hemorrhage hematoma expansion prediction 9-point score BRAIN score ABC/2 planimetry

Introduction

Hematoma expansion is a significant event associated with early neurologic deterioration and poor patient outcomes in spontaneous intracerebral hemorrhage (ICH).^1,2 To date, many trials have focused on reducing the risk of hematoma expansion.^3
–6 Additionally, numerous studies have aimed to accurately predict hematoma expansion on admission by combining clinical information.^7
–26 Various methods, including scoring systems and state-of-the-art radiomics and machine learning analyses, have been proposed for the prediction.^{8,11,18,20,21,23
–26} Since radiomics and machine learning are not yet widely applicable, scoring systems, such as 9-point, BRAIN, and BAT scores, remain valuable.^8,11,18 These prediction scores are easily adoptable in clinical practice and are particularly useful in rural areas, where they assist in deciding whether a patient should be transferred to specialists.

The 9-point and BRAIN scores require hematoma volume to assign points.^8,11 The ABC/2 formula and the planimetry technique are widely used to measure hematoma volume. In the ABC/2 formula, hematoma volume is manually calculated using three mutually perpendicular lengths. The planimetry technique, on the other hand, calculates hematoma volume by multiplying segmented hematoma areas by a slice thickness using software. Planimetry can measure hematoma volume more precisely than ABC/2 that is known to overestimate hematoma volume.^27,28 To develop the 9-point score, planimetry was adopted to measure hematoma volume.⁸ In the BRAIN score, planimetry was employed in most cases, with ABC/2 used in a small number of cases.¹¹

Here, a question arises. Although the 9-point and BRAIN scores were developed predominantly using planimetry to measure hematoma volume, can they also predict hematoma expansion with the same performance when ABC/2, which may overestimate hematoma volume, is used as a substitute for planimetry? In real ICH patient management, ABC/2 is usually used for evaluating hematoma volume because planimetry is more time-consuming.

In this study, we investigate the performance of the 9-point and BRAIN scores using hematoma volume measured by ABC/2 and planimetry. Then, we aim to determine whether ABC/2 can be reliably used for these scores to predict hematoma expansion in spontaneous ICH.

Patients and methods

Study population

The institutional review boards granted approval for this study. Individual informed consent was waived by the institutional review boards due to the retrospective nature of this study. All study protocols and procedures were conducted in accordance with the Declaration of Helsinki. This article was prepared according to the standards for reporting of diagnostic accuracy (STARD) statement.

Consecutive patients aged ⩾ 18 years with acute ICH admitted to a university hospital and three regional base hospitals were retrospectively reviewed. The hospitals included Mie University Hospital (January 2017 to December 2021), Mie Chuo Medical Center (December 2013 to December 2021), Matsusaka Chuo General Hospital (January 2018 to September 2021), and Suzuka Kaisei Hospital (October 2017 to December 2021). Inclusion criteria were as follows: baseline CT scan within 24 h of onset; follow-up CT scan within 48 h after baseline CT scan; and CT scans with a thickness of 5.0 mm or less. Exclusion criteria were as follows: multiple hemorrhage; secondary cause of ICH (e.g. hemorrhagic transformation of ischemic infarction, aneurysm, arteriovenous malformation, arteriovenous fistula, and tumor); traumatic ICH; surgical hematoma evacuation before follow-up CT scan; and insufficient data.

A total of 429 patients were enrolled in this study (Figure 1): 100 (Mie University Hospital), 105 (Mie Chuo Medical Center), 116 (Matsusaka Chuo General Hospital), and 108 (Suzuka Kaisei Hospital). The following demographic and clinical data on admission were collected from the medical records: age, sex, previous history (ICH, cerebral infarction, ischemic heart disease, hypertension, diabetes mellitus, and dyslipidemia), anticoagulant use, antiplatelet use, systolic and diastolic blood pressures, Glasgow Coma Scale, hemoglobin, platelet count, white blood cell count, serum creatinine, serum total bilirubin, prothrombin time-international normalized ratio (PT-INR), and time from onset to baseline CT scan (Table 1).

Figure 1.

Flow diagram of patient enrollment.

Table 1.

Demographic and clinical characteristics of patients. .

Characteristics	N = 429
Age, y	72 (61–81)
Sex (male)	245 (57.1)
Medical history
Intracerebral hemorrhage	27 (6.3)
Cerebral infarction	48 (11.2)
Ischemic heart disease	26 (6.1)
Hypertension	268 (62.5)
Diabetes mellitus	107 (24.9)
Dyslipidemia	147 (34.3)
Anticoagulant use	44 (10.3)
Antiplatelet use	66 (15.4)
Systolic blood pressure, mmHg	186.0 ± 32.9
Diastolic blood pressure, mmHg	104.5 ± 23.4
Glasgow Coma Scale	14 (11–15)
Hemoglobin, mg/dl	13.7 ± 2.1
Platelet count, 10⁶/ml	214 (179–253)
White blood cell count, 10⁶/ml	7.60 (5.80–9.60)
Serum creatinine, mg/dl	0.72 (0.58–0.90)
Serum total bilirubin, mg/dl	0.7 (0.5–0.9)
Prothrombin time-international normalized ratio	0.97 (0.92–1.04)
Time from onset to baseline CT scan, h	2 (1–3)

Data are presented as n (%), mean ± standard deviation, or median (interquartile range).

Imaging acquisition

CT scanning was conducted at 120 kVp with a slice thickness of 0.25–5.0 mm. CT angiography was performed by injecting 50–100 ml of iodinated contrast medium at 3.5–5.0 ml/s. One hundred and thirty-four patients (31.2%) underwent CT angiography.

Inhospital management

Following the identification of ICH on the baseline CT scan, continuous blood pressure monitoring and the initiation of blood pressure-lowering treatment were undertaken. Throughout the period between the baseline and follow-up CT scans, calcium channel blockers were administered as antihypertensive agents, with a target systolic blood pressure of less than 140 mmHg or 180 mmHg. If patients were on anticoagulants, reversal agents were administered.

Hematoma volume measurement by ABC/2 and planimetry

Only intraparenchymal hematoma was measured as hematoma volume; intraventricular hematoma was not included in the hematoma volume. For ABC/2 and planimetry, two different raters who are board-certified stroke specialists measured hematoma volume. Each rater evaluated approximately half of the total cases. Prior to conducting the measurements, the raters simultaneously assessed several cases as part of a training session to reach a consensus on the evaluation method.

In ABC/2, the following three lengths were specified: the maximum length measured on the slice with the largest hematoma area (A), the maximum length perpendicular to A (B), and the number of slices with hematoma multiplied by the slice thickness (C). Hematoma volume was calculated by multiplying A, B, and C and dividing by 2. ABC/2 was performed on the baseline CT scan to assign points to the 9-point and BRAIN scores.

In planimetry, hematoma was manually segmented in each slice using 3D slicer. Hematoma volume was computed by multiplying the segmented areas by the slice thickness. Planimetry was performed on the baseline CT scan to assign the points and on the follow-up CT scan to assess hematoma expansion.

Hematoma expansion was defined as a volume increase in planimetry between baseline and follow-up CT scans greater than 6 ml or 33% of baseline volume.^{8,11,18,19,29,30}

Qualitative assessment of CT findings at baseline

Hematoma locations were categorized as basal ganglia, thalamus, brainstem, cerebellum, and lobe. Hemorrhage in the basal ganglia, thalamus, brainstem, and cerebellum was defined as deep ICH, whereas hemorrhage in the lobe was defined as lobar ICH. Intraventricular extension of hematoma was evaluated. CT angiography spot sign was assessed with the following definition: ⩾1 focus of any size and morphology of contrast pooling (⩾120 Hounsfield units) within the hematoma, and discontinuous from normal or abnormal vasculature adjacent to the hematoma.^16,31 All assessments were conducted by two raters. In case of disagreement, the CT images were reassessed by both raters together and a consensus was reached.

9-point and BRAIN scores

The 9-point score was calculated from hematoma volume, CT angiography spot sign, anticoagulant use, and time from onset to baseline CT scan.⁸ The BRAIN score was calculated from hematoma volume, intraventricular extension of the hematoma, anticoagulant use, time from onset to baseline CT scan, and previous history of ICH.¹¹

Statistical analysis

Continuous variables were summarized as mean with standard deviation or median with interquartile range, depending on the distribution of the variable evaluated by the Shapiro-Wilk test. Categorical variables were summarized as counts with percentages.

Inter-rater reliability was evaluated for hematoma volume measurement and for the qualitative assessment of CT findings. For hematoma volume measurement, intra-class correlation coefficient (ICC) was calculated. Two raters measured identical 70 hematoma volumes that were randomly selected, and ICC was assessed for ABC/2 and planimetry, respectively. For the quantitative assessment of CT findings, Cohen’s kappa coefficient was calculated. It was evaluated for hematoma locations, intraventricular extension of hematoma, and angiography spot sign.

Wilcoxon signed-rank test and Pearson’s correlation coefficient were applied to analyze the relationship between ABC/2 and planimetry. The area under the receiver operating curve (AUC), sensitivity, and specificity in predicting hematoma expansion with the 9-point and BRAIN scores were calculated. Two AUCs were compared using DeLong test.

To investigate the influence of various factors on ABC/2 volumetric estimation, additional analyses were conducted. Patients were stratified according to hematoma volume as evaluated by planimetry (<15 ml, 15–30 ml, >30 ml), hematoma location (deep ICH, lobar ICH), and anticoagulant use. Measurement errors in hematoma volume between ABC/2 and planimetry were compared within each group using Mann-Whitney U test or Kruskal-Wallis rank sum test.

p values less than 0.05 were considered significant. All statistical analyses were performed using EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan).

Results

Table 2 summarizes the assessments of baseline and follow-up CT scans. Hematoma expansion was observed in 75 patients (17.5%). The ICCs for hematoma volume measurements with ABC/2 and planimetry were 0.920 (95% confidence interval [CI], 0.875–0.950) and 0.951 (95% CI, 0.839–0.978), respectively. Cohen’s kappa coefficients for the CT findings of hematoma locations, intraventricular extension of hematoma, and angiography spot sign were 0.914 (95% CI, 0.882–0.946), 0.868 (95% CI, 0.820–0.916), and 0.723 (95% CI, 0.591–0.855), respectively.

Table 2.

Baseline and follow-up CT analyses. .

	N = 429
Hematoma location
Basal ganglia	168 (39.2)
Thalamus	136 (31.7)
Brain stem	22 (5.1)
Cerebellum	24 (5.6)
Lobe	79 (18.4)
CT markers at baseline
Intraventricular hematoma extension	174 (40.6)
CT angiography spot sign*	31 (23.1)
Hematoma expansion	75 (17.5)

Data are presented as n (%).

CT angiography was performed in 134 patients.

The median hematoma volume measured by ABC/2 was 11.97 ml (interquartile range (IQR), 4.8–30.0), while the volume measured by planimetry was 11.70 ml (IQR, 4.9–26.6). A significant difference was observed by Wilcoxon signed-rank test (p < 0.001). The median measurement error between ABC/2 and planimetry was 0.30 ml (IQR, −0.72–2.87). ABC/2 overestimated hematoma volume in 244 patients (56.9%) compared to planimetry. Pearson’s correlation coefficient between ABC/2 and planimetry was 0.965 (95% CI, 0.958–0.971), showing a very strong correlation (Figure 2).

Figure 2.

Pearson’s correlation coefficient between ABC/2 and planimetry for hematoma volume.

The 9-point score derived from ABC/2 and planimetry is summarized in Table 3. The AUCs for predicting hematoma expansion were 0.735 (95% CI, 0.675–0.796) with ABC/2 and 0.732 (95% CI, 0.672–0.793) with planimetry (Figure 3(a)). No significant difference was found between the AUCs by DeLong test (p = 0.647). With a cutoff value of 4, sensitivity and specificity for ABC/2 were 0.627 and 0.768, and those for planimetry were 0.600 and 0.785.

Table 3.

9-point score based on ABC/2 and planimetry.

Score	ABC/2			Planimetry
Score	Total N = 429	Expansion N = 75	Expansion rate (%)	Total N = 429	Expansion N = 75	Expansion rate (%)
0	3	0	0.0	3	0	0.0
1	26	0	0.0	27	0	0.0
2	72	6	8.3	72	6	8.3
3	199	22	11.1	206	24	11.7
4	65	19	29.2	60	18	30.0
5	50	19	38.0	47	18	38.3
6	4	2	50.0	7	4	57.1
7	9	6	66.7	6	4	66.7
8	1	1	100.0	1	1	100.0
9	0	–	–	0	–	–

Figure 3.

(a) The receiver operating curve (ROC) in predicting hematoma expansion with the 9-point score. The area under the ROCs (AUCs) were 0.735 for ABC/2 and 0.732 for planimetry and (b) the ROC with the BRAIN score. The AUCs were 0.754 for ABC/2 and 0.745 for planimetry.

The BRAIN score is summarized in Table 4. The AUCs were 0.753 (95% CI, 0.693–0.813) for ABC/2 and 0.745 (95% CI, 0.688–0.803) for planimetry (Figure 3(b)). DeLong test showed no significant difference between the AUCs (p = 0.445). With a cutoff value of 10, sensitivity and specificity for ABC/2 were 0.787 and 0.568, and those for planimetry were 0.773 and 0.593.

Table 4.

BRAIN score based on ABC/2 and planimetry.

Score	ABC/2			Planimetry
Score	Total N = 429	Expansion N = 75	Expansion rate (%)	Total N = 429	Expansion N = 75	Expansion rate (%)
0	12	0	0.0	11	0	0.0
1	4	0	0.0	4	0	0.0
2	13	0	0.0	13	0	0.0
3	12	0	0.0	12	0	0.0
4	23	0	0.0	24	0	0.0
5	57	7	12.3	63	6	9.5
6	25	3	12.0	23	3	13.0
7	31	2	6.5	32	2	6.3
8	14	3	21.4	16	3	18.8
9	26	1	3.8	29	3	10.3
10	42	5	11.9	40	8	20.0
11	30	7	23.3	31	7	22.6
12	53	13	24.5	51	14	27.5
13	19	4	21.1	18	4	22.2
14	33	10	30.3	30	7	23.3
15	3	1	33.3	4	1	25.0
16	16	7	43.8	14	7	50.0
17	5	3	60.0	4	2	50.0
18	5	4	80.0	4	3	75.0
19	1	1	100.0	1	1	100.0
20	5	4	80.0	5	4	80.0

Hematoma volumes by ABC/2 and planimetry and their measurement errors are summarized in Table 5, stratified by hematoma volume as determined by planimetry (<15 ml, 15–30 ml, >30 ml), hematoma location (deep ICH, lobar ICH), and anticoagulant use. Larger hematoma volume and lobar ICH were significantly associated with higher measurement error, whereas anticoagulant use was not. PT-INR on admission in patients with and without anticoagulant use were 1.32 (IQR, 1.10–1.66) and 0.96 (IQR, 0.92–1.02), respectively.

Table 5.

Hematoma volumes and measurement errors. .

		Hematoma volume (ml)		Measurement error (ml)	p value
		ABC/2	Planimetry	Measurement error (ml)	p value
Hematoma volume by planimetry					<0.001
<15 ml	N = 251	6.16 (2.98–10.37)	6.00 (2.90–9.80)	−0.08 (−0.72–0.78)
15–30 ml	N = 81	22.87 (19.14–28.13)	20.30 (17.60–25.00)	1.64 (−0.40–5.00)
>30 ml	N = 97	60.30 (43.59–95.65)	52.80 (40.20–91.20)	6.25 (−1.10–14.84)
Hematoma location					0.010
Deep	N = 350	10.12 (4.18–21.01)	9.50 (4.30–18.80)	0.21 (−0.66–1.96)
Lobar	N = 79	55.35 (29.70–95.69)	45.10 (27.50–91.80)	2.18 (−1.04–10.00)
Anticoagulant use					0.960
Yes	N = 44	10.42 (3.55–22.20)	9.70 (3.18–23.50)	0.42 (0.66–2.05)
No	N = 385	12.10 (4.90–32.60)	12.00 (5.30–27.50)	0.30 (−0.72–2.88)

Data are presented as median (interquartile range).

Measurement error represents the difference in hematoma volume between ABC/2 and planimetry.

The p value was calculated for the difference in measurement error within each group.

Discussion

The 9-point and BRAIN scores using hematoma volume measured by ABC/2 and planimetry were effective in predicting hematoma expansion in spontaneous ICH. The ABC/2 volumetric estimation can be reliably used for these scores in clinical practice.

ABC/2 overestimated hematoma volume in more than half of the cases in this study, which was consistent with previous reports.^27,28 However, the difference in median hematoma volume between ABC/2 and planimetry was subtle (11.97 ml vs 11.70 ml). In addition, the measurement error between them was 0.30 ml, which was very small. In both the 9-point and BRAIN scores, hematoma volume is divided into three categories: < 30 ml, 30–60 ml, and > 60 ml in the 9-point score; and 10 ml, 10–20 ml, and > 20 ml in the BRAIN score. Therefore, only hematoma volume around 30 ml and 60 ml in the 9-point score and around 10 ml and 20 ml in the BRAIN score influence the point assignment differences. Especially in the 9-point score, since the same point is assigned to hematoma volume less than 30 ml, many cases were not affected by the small difference in hematoma volume measurement between ABC/2 and planimetry. The receiver operating curves were almost overlapping (Figure 3(a)), and the score distributions were very similar (Table 3). On the other hand, in the BRAIN score, the influence of hematoma volume on the point assignment was greater in many cases than in the 9-point score because of the lower volume setting in the hematoma volume component. However, when points from the other components were included, the AUCs and score distributions became relatively similar (Figure 3(b) and Table 4).

Although the scoring system, such as the 9-point and BRAIN scores, is a conventional method for predicting hematoma expansion, it is still valuable because of its ease of application in clinical practice. In major hospitals in developed countries, the artificial intelligence-based systems for predicting the clinical course of patients will be introduced in the not-so-distant future. Many studies have addressed the prediction of hematoma expansion using radiomics, machine learning, and neural networks; these methods will be integrated into medical record or imaging systems.^{20,21,23
–26} However, in developing countries or rural areas, the introduction of the latest technology is usually delayed. In this study, the BRAIN score based on ABC/2 showed good AUC and sensitivity. It will continue to help clinicians for some time yet as an easy-to-use method for clinical decision making to select patients with ICH for admission to the intensive care unit or referral to specialists.

To date, 10 scoring systems have been reported to predict hematoma expansion.³² Of these, the following four scoring systems included hematoma volume for prediction: 9-point score, BRAIN score, HEAVN score, and Basal Ganglia Score.^8,11,33,34 The 9-point and BRAIN scores were developed using prospective cohort data and validated with external data, whereas the HEAVN and Basal Ganglia Scores were developed with single-center retrospective data and have not been externally validated.^{8,11,30,33
–35} Therefore, the latter two were deemed ineligible for evaluation and were not included in this study.

In the additional analyses to identify factors influencing the measurement error between ABC/2 and planimetry, larger hematoma volume and lobar hemorrhage were significantly associated with higher measurement error. This finding was consistent with previous reports.^27,28 Anticoagulant use was not associated with measurement error, which may be due to the relatively smaller hematoma volume observed in patients on anticoagulants. The effect of anticoagulant use on hematoma volume at admission depends on the PT-INR, with PT-INR ⩽ 3.0 having no effect.^36,37 In this study, the median PT-INR in patients on anticoagulants was 1.32, which did not result in increased hematoma volume at admission.

There are several limitations to note. First, the results of this study were based on the retrospective cohort. Prospective data collection will provide more reliable results. Second, this study included only patients from Japan, which introduces a potential bias in patient background. However, it focused on technical aspects of hematoma volume measurement rather than the disease itself and had a sufficient sample size with diverse hematoma locations and volumes. Therefore, the results may be generalizable and applicable to patients worldwide. Third, the proportion of patients who underwent CT angiography was lower in this study (31.2%) compared to the original article that developed the 9-point score (46.1%). Given that the CT angiography spot sign is one of the strongest predictors of hematoma expansion, it would have been desirable to include more patients with CT angiography. Forth, the number of patients with a score of 6 or more in the 9-point score and with a score of 17 or more in the BRAIN score was small. More patients with the higher scores are required for more accurate investigations.

Conclusion

The 9-point and BRAIN scores for predicting hematoma expansion in spontaneous ICH using hematoma volume measured by ABC/2 and planimetry showed good performance. ABC/2 volumetric estimation proved to be reliable for measuring hematoma volume in these scoring systems.

Footnotes

Acknowledgements

None.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Ethical approval

This study was approved by the following institutional review boards: Mie University Hospital institutional review board (REC number: T2023-7), Mie Chuo Medical Center institutional review board (REC number: MCERB-202321), Matsusaka Chuo General Hospital institutional review board (REC number: 325), and Suzuka Kaisei Hospital institutional review board (REC number: 2020-05).

Informed consent

Individual informed consent was waived by the institutional review boards due to the retrospective nature of this study.

Guarantor

ST.

Contributorship

ST reviewed the literature and designed the study. FI, KT, TA, and HS were involved in protocol development and ethical approval. ST, TY, TK, and MI collected and interpreted the data. YK and YN performed the data analysis. OUA, AH, and DF validated the data analysis. ST wrote the first draft of the manuscript. All authors reviewed, edited, and approved the final version of the manuscript.

ORCID iDs

Satoru Tanioka

Adam Hilbert

References

Delcourt

Huang

Arima

, et al. Hematoma growth and outcomes in intracerebral hemorrhage: the INTERACT1 study. Neurology 2012; 79: 314–319.

Leira

Davalos

Silva

, et al. Early neurologic deterioration in intracerebral hemorrhage: predictors and associated factors. Neurology 2004; 63: 461–467.

Anderson

Heeley

Huang

, et al. Rapid blood-pressure lowering in patients with acute intracerebral hemorrhage. N Engl J Med 2013; 368: 2355–2365.

Anderson

Huang

Wang

, et al. Intensive blood pressure reduction in acute cerebral haemorrhage trial (INTERACT): a randomised pilot trial. Lancet Neurol 2008; 7: 391–399.

Qureshi

Palesch

Barsan

, et al. Intensive blood-pressure lowering in patients with acute cerebral hemorrhage. N Engl J Med 2016; 375: 1033–1043.

Tsivgoulis

Katsanos

Butcher

, et al. Intensive blood pressure reduction in acute intracerebral hemorrhage: a meta-analysis. Neurology 2014; 83: 1523–1529.

Selariu

Zia

Brizzi

, et al. Swirl sign in intracerebral haemorrhage: definition, prevalence, reliability and prognostic value. BMC Neurol 2012; 12: 109.

Brouwers

Chang

Falcone

, et al. Predicting hematoma expansion after primary intracerebral hemorrhage. JAMA Neurol 2014; 71: 158–164.

Blacquiere

Demchuk

Al-Hazzaa

, et al. Intracerebral hematoma morphologic appearance on noncontrast computed tomography predicts significant hematoma expansion. Stroke 2015; 46: 3111–3116.

10.

Zhang

Huang

, 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.

11.

Wang

Arima

Al-Shahi Salman

, et al. Clinical prediction algorithm (BRAIN) to determine risk of hematoma growth in acute intracerebral hemorrhage. Stroke 2015; 46: 376–381.

12.

Boulouis

Morotti

Brouwers

, et al. Association between hypodensities detected by computed tomography and hematoma expansion in patients with intracerebral hemorrhage. JAMA Neurol 2016; 73: 961–968.

13.

Boulouis

Morotti

Charidimou

, et al. Noncontrast computed tomography markers of intracerebral hemorrhage expansion. Stroke 2017; 48: 1120–1125.

14.

Sporns

Schwake

Schmidt

, 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: 131–135.

15.

Al-Shahi Salman

Frantzias

Lee

, et al. Absolute risk and predictors of the growth of acute spontaneous intracerebral haemorrhage: a systematic review and meta-analysis of individual patient data. Lancet Neurol 2018; 17: 885–894.

16.

Sun

Liu

, et al. Iodine sign as a novel predictor of hematoma expansion and poor outcomes in primary intracerebral hemorrhage patients. Stroke 2018; 49: 2074–2080.

17.

Morotti

Boulouis

Charidimou

, et al. Integration of computed tomographic angiography spot sign and noncontrast computed tomographic hypodensities to predict hematoma expansion. Stroke 2018; 49: 2067–2073.

18.

Morotti

Dowlatshahi

Boulouis

, et al. Predicting intracerebral hemorrhage expansion with noncontrast computed tomography: the BAT score. Stroke 2018; 49: 1163–1169.

19.

Poli

Leuci

Costa

, et al. Validation and comparison of noncontrast CT scores to predict intracerebral hemorrhage expansion. Neurocrit Care 2020; 32: 804–811.

20.

Teng

Ren

Zhang

, et al. Artificial intelligence can effectively predict early hematoma expansion of intracerebral hemorrhage analyzing noncontrast computed tomography image. Front Aging Neurosci 2021; 13: 632138.

21.

Zhong

Jin

Song

, et al. Deep learning for automatically predicting early haematoma expansion in Chinese patients. Stroke Vasc Neurol 2021; 6: 610–614.

22.

Guo

, et al. External validation study on the value of deep learning algorithm for the prediction of hematoma expansion from noncontrast CT scans. BMC Med Imaging 2022; 22: 45.

23.

Tang

Zhu

, et al. Deep learning-based prediction of hematoma expansion using a single brain computed tomographic slice in patients with spontaneous intracerebral hemorrhages. World Neurosurg 2022; 165: e128–e136.

24.

Tanioka

Yago

Tanaka

, et al. Machine learning prediction of hematoma expansion in acute intracerebral hemorrhage. Sci Rep 2022; 12: 12452.

25.

Xiong

Huang

, et al. Using radiomics and convolutional neural networks for the prediction of hematoma expansion after intracerebral hemorrhage. Int J Gen Med 2023; 16: 3393–3402.

26.

Tanioka

Aydin

Hilbert

, et al. Prediction of hematoma expansion in spontaneous intracerebral hemorrhage using a multimodal neural network. Sci Rep 2024; 14: 16465.

27.

Huttner

Steiner

Hartmann

, et al. Comparison of ABC/2 estimation technique to computer-assisted planimetric analysis in warfarin-related intracerebral parenchymal hemorrhage. Stroke 2006; 37: 404–408.

28.

Webb

Ullman

Morgan

, et al. Accuracy of the ABC/2 score for intracerebral hemorrhage: systematic review and analysis of MISTIE, CLEAR-IVH, and CLEAR III. Stroke 2015; 46: 2470–2476.

29.

Dowlatshahi

Demchuk

Flaherty

, et al. Defining hematoma expansion in intracerebral hemorrhage: relationship with patient outcomes. Neurology 2011; 76: 1238–1244.

30.

Huynh

Aviv

Dowlatshahi

, et al. Validation of the 9-point and 24-point hematoma expansion prediction scores and derivation of the PREDICT A/B scores. Stroke 2015; 46: 3105–3110.

31.

Delgado Almandoz

Yoo

Stone

, 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: 2994–3000.

32.

Yogendrakumar

Moores

Sikora

, et al. Evaluating hematoma expansion scores in acute spontaneous intracerebral hemorrhage: a systematic scoping review. Stroke 2020; 51: 1305–1308.

33.

Huang

Zhang

Yang

A reliable grading system for prediction of hematoma expansion in intracerebral hemorrhage in the basal ganglia. Biosci Trends 2018; 12: 193–200.

34.

Miyahara

Noda

Yamaguchi

, et al. New prediction score for hematoma expansion and neurological deterioration after spontaneous intracerebral hemorrhage: a hospital-based retrospective cohort study. J Stroke Cerebrovasc Dis 2018; 27: 2543–2550.

35.

VanDerWerf

Kurowski

Siegler

, et al. Combination of intra-hematomal hypodensity on CT and BRAIN scoring improves prediction of hemorrhage expansion in ICH. Neurocrit Care 2018; 29: 40–46.

36.

Flibotte

Hagan

O’Donnell

, et al. Warfarin, hematoma expansion, and outcome of intracerebral hemorrhage. Neurology 2004; 63: 1059–1064.

37.

Flaherty

Tao

Haverbusch

, et al. Warfarin use leads to larger intracerebral hematomas. Neurology 2008; 71: 1084–1089.

Reliability of ABC/2 volumetric estimation in spontaneous intracerebral hemorrhage for hematoma expansion prediction scores

Abstract

Introduction:

Patients and methods:

Results:

Discussion and conclusion:

Keywords

Introduction

Patients and methods

Study population

Imaging acquisition

Inhospital management

Hematoma volume measurement by ABC/2 and planimetry

Qualitative assessment of CT findings at baseline

9-point and BRAIN scores

Statistical analysis

Results

Discussion

Conclusion

Footnotes

Acknowledgements

Declaration of conflicting interests

Funding

Ethical approval

Informed consent

Guarantor

Contributorship

ORCID iDs

References