Abstract
Background and Objectives
Hematoma expansion (HE) is a major determinant of neurologic deterioration and poor outcome in intracerebral hemorrhage (ICH) and represents an appealing therapeutic target. We analyzed the prognostic effect of different degrees of HE.
Methods
This was a retrospective analysis of patients with ICH admitted at 8 academic institutions in Italy, Germany, Canada, China, and the United States. All patients underwent baseline and follow-up imaging for HE assessment. Relative HE (rHE) was classified as follows: none (<0%), mild (0%–33%), moderate (33.1%–66%), and severe (>66%). Absolute HE (aHE) was classified as none (<0 mL), mild (0–6.0 mL), moderate (6.1–12.5 mL), and severe (>12.5 mL). Predictors of poor functional outcome (90 days modified Rankin Scale 4–6) were explored with logistic regression.
Results
We included 2,163 patients, of whom 1,211 (56.0%) had poor outcome. The occurrence of severe aHE or rHE was more common in patients with unfavorable outcome (13.9% vs 6.5%, p < 0.001 and 18.3% vs 7.2%, p < 0.001 respectively). This association was confirmed in logistic regression (rHE odds ratio [OR] 1.98, 95% CI 1.38–2.82, p < 0.001; aHE OR 1.73, 95% CI 1.23–2.45, p = 0.002) while there was no association between mild or moderate HE and poor outcome. The association between severe HE and poor outcome was significant only in patients with baseline ICH volume below 30 mL.
Discussion
The strongest association between HE and outcome was observed in patients with smaller initial volume experiencing severe HE. These findings may inform clinical trial design and guide clinicians in selecting patients for antiexpansion therapies.
Introduction
Hematoma expansion (HE) is a major cause of neurologic deterioration and poor functional outcome in patients with intracerebral hemorrhage (ICH).1,2 Prevention of HE is, therefore, a compelling therapeutic approach to improve prognosis.3,4 Observational studies and randomized trials have traditionally analyzed HE as a dichotomous variable, combining absolute and relative ICH volume growth using previously proposed fixed thresholds.1,5,6 The use of a dichotomous definition of HE favored standardization and comparison between different studies but suffered from the inherent limitation of losing important information because of the arbitrary dichotomization of a continuous biological variable.7 Furthermore, the use of a dichotomous definition combining absolute and relative cutoffs might miss a potential interaction between baseline ICH volume and the prognostic influence of HE. Patients with small ICH volume may be labeled as “expanders” based on a large percentage volume change that is small in absolute terms and may have minimal clinical significance. Conversely, even large growth in volume may not clinically change the outcome in patients with sufficiently large hemorrhages on arrival. Overall, the traditional cutoffs for HE may not provide the information needed to correlate with changes in outcome. This suggests the need for a deeper and more detailed analysis of HE and a more granular assessment of the entire spectrum of hematoma growth.8 The aim of this study was to describe the prognostic effect of HE analyzed as a polychotomous variable and investigate whether this association is a function of baseline hematoma volume.
Methods
Standard Protocol Approvals, Registrations, and Patient Consents
All the study procedures were approved by the Institutional Review Boards of each participating institution: Arcispedale S. Anna, Ferrara, Italy (PN 26032009-15122011); IRCCS Mondino Foundation, Pavia, Italy (PN 0035588/22); ASST Spedali Civili, Brescia, Italy (PN 4067-08052020); IRCCS Istituto delle Scienze Neurologiche, Bologna, Italy (DLGS 196/2003); The First Affiliated Hospital of Chongqing Medical University, Chongqing, China (PN 2017-075); Massachusetts General Hospital, Boston, MA (PN 2006P000570); McMaster University/Population Health Research Institute, Hamilton, Ontario, Canada (PN 3253); and Charité Hospital, Berlin, Germany (PN EA1/035/20). Written informed consent was obtained by patients and caregivers or waived by the Institutional Review Boards.
Study Population
We retrospectively selected patients admitted for primary spontaneous nontraumatic ICH at 8 sites in Italy, Germany, Canada, China, and the United States. Patients meeting the following criteria were included in our analysis: (1) diagnosis of primary spontaneous nontraumatic ICH on noncontrast CT (NCCT), (2) age 18 years or older, and (3) availability of good quality follow-up NCCT acquired 24–72 hours from symptom onset or time last seen well. Patients were excluded in case of any of the following conditions: (1) traumatic brain injury, vascular malformation, or other intracranial pathology leading to secondary ICH; (2) primary, isolated intraventricular hemorrhage (IVH); (3) ischemic stroke with hemorrhagic transformation; and (4) surgical hematoma evacuation or any other head surgery before follow-up NCCT. The population selection process is summarized in Figure 1. All clinical and imaging variables were collected by trained investigators, blinded to the outcomes of interest.
Figure 1. Selection Flowchart.
ICH = intracerebral hemorrhage.
Clinical Variables
Age, sex, medical history of hypertension, antiplatelet and anticoagulant treatment, admission systolic blood pressure, Glasgow Coma Scale (GCS) score, and time from onset to baseline imaging were collected. Medical management followed the American Heart Association/American Stroke Association guidelines.9,10 Functional outcome at 3 months from the index event was measured with the modified Rankin Scale (mRS) and assessed through follow-up telephone calls, outpatient service evaluations, or querying the national social security databases for mortality data.
Image Acquisition and Analysis
All patients underwent baseline and follow-up NCCT scans with axial reconstruction and 3–5 mm slice thickness. Follow-up NCCT was routinely performed at 24 hours from symptom onset or at the time last seen well, and in cases with multiple early NCCT scans, HE occurrence was assessed based on imaging at 24 hours. The NCCT parameters were not standardized and followed the clinical practice local acquisition protocol at each participating institution. ICH volumes were calculated with semiautomated, computer-assisted planimetric softwares. NCCT images were also analyzed for IVH presence and determination of ICH location, classified as supratentorial lobar (involving the cortex, cortical-subcortical junction, and subcortical white matter), supratentorial deep (involving the thalamus, basal ganglia, internal capsule, and deep periventricular white matter), and infratentorial (involving the brainstem and cerebellum).11
HE Definition
Relative (follow-up ICH volume minus baseline ICH volume, divided by baseline ICH volume, %) and absolute (follow-up ICH volume minus baseline ICH volume, mL) HE were calculated. HE was analyzed as a tetrachotomous variable, with the following 4 categories: none, mild, moderate, and severe, based on previously proposed cutoffs.1,8 Relative HE (rHE) was classified as follows: none (<0%), mild (0%–33%), moderate (33.1%–66%), and severe (>66%). Figure 2 shows an example of different HE degrees. Absolute HE (aHE) was classified as follows: none (<0 mL), mild (0–6.0 mL), moderate (6.1–12.5 mL), and severe (>12.5 mL).
Figure 2. Illustrative Example of Different Degrees of Hematoma Expansion.
Baseline and follow-up imaging in patients with no (A), mild (B), moderate (C), and severe (D) hematoma expansion.
Statistical Analysis
Categorical variables were summarized as count (percentage) and compared with the χ2 test. Continuous variables were expressed as median (interquartile range) or mean (SD) based on their distribution, evaluated with the Shapiro-Wilk test, and compared using the Mann-Whitney or t test, respectively as appropriate. Poor functional outcome, defined as mRS 4–6 at 90 days from ICH onset, was the main outcome of interest.5 Predictors of poor outcome were investigated with multivariable binary logistic regression, including age, baseline ICH volume, admission GCS score, presence of IVH,12 and all variables with p < 0.1 in univariate analysis comparing patients with good and poor prognosis. In a sensitivity analysis, regression models also accounted for time from onset/last seen well to baseline imaging.
The following secondary analyses were also performed: (1) poor functional outcome was defined as mRS scale 3–6 at 90 days; (2) the logistic regression analyses were repeated after the exclusion of infratentorial hemorrhages; (3) the prognostic effect of HE was explored stratifying patients by baseline ICH volume, with the following cutoffs: <15 mL, 15–30 mL, and >30 mL12-14; and (4) the association between HE and outcome was tested using ordinal logistic regression. Finally, every patient's predicted probability of unfavorable prognosis was calculated based on individual data and logistic regression estimates and expressed as a continuous variable ranging from 0 to 1. This additional analysis was performed in the entire study population and in the subgroup of patients taking any antithrombotic medication on admission.
All the analyses were performed with the statistical package SPSS version 21.0, and statistical significance was set at p < 0.05. The study findings were reported following the Strengthening the Reporting of Observational Studies in Epidemiology.15
Data Availability
Requests to access the data set may be sent to the corresponding author.
Results
A total of 2,163 patients met the inclusion criteria, of whom 1,211 (56.0%) had poor functional outcome at 90 days. The characteristics of the study population and the comparison between patients with good and poor outcome are summarized in Table 1. Patients with poor outcome were older and more severely affected on admission, as highlighted by larger ICH volume, lower GCS score, and higher rates of IVH and infratentorial hemorrhages. The occurrence of severe rHE or aHE was significantly higher in the subgroup of patients with unfavorable prognosis (13.9% vs 6.5%, p < 0.001 and 18.3% vs 7.2%, p < 0.001 and respectively). Severe rHE or severe aHE remained independently associated with higher odds of poor functional outcome after adjustment for potential confounders in multivariable logistic regression, as shown in Table 2. The inclusion of baseline imaging timing in logistic regression did not modify this finding (severe rHE, odds ratio [OR] 1.91, 95% CI 1.34–2.74, p < 0.001; severe aHE, OR 1.68, 95% CI 1.19–2.37, p = 0.003). Ordinal regression confirmed the association between severe HE and outcome (rHE OR 2.14, 95% CI 1.61–2.84, p < 0.001; aHE OR 1.61, 95% CI 1.23–2.10, p = 0.001). Conversely, mild or moderate HE were not associated with poor outcome when compared with patients with no HE.
Table 1.
Population Characteristics
| All (N = 2,163) | Functional outcome at 90 d | |||
| mRS 0–3 (n = 952) | mRS 4–6 (n = 1,211) | p Value | ||
| Age, y, median (IQR) | 71 (61–80) | 67 (57–76) | 75 (65–82) | <0.001 |
| Sex, male, n (%) | 1,085 (50.2) | 489 (51.4) | 596 (49.2) | 0.321 |
| History of hypertension, n (%) | 1,542 (71.3) | 617 (64.8) | 925 (76.4) | <0.001 |
| Antiplatelet treatment, n (%) | 564 (26.1) | 227 (23.8) | 337 (27.8) | 0.036 |
| Anticoagulant treatment, n (%) | 387 (17.9) | 106 (11.1) | 281 (23.2) | <0.001 |
| SBP, mean (IQR), mm Hg | 167 (150–190) | 163 (143–184) | 170 (150–195) | <0.001 |
| GCS, median (IQR) | 14 (10–15) | 14 (11–15) | 13 (8–15) | <0.001 |
| Time from onset/LSW to NCCT, h, median (IQR) | 4.9 (2.3–12.0) | 5.6 (2.8–13.5) | 4.3 (2.0–11.0) | <0.001 |
| ICH location, n (%) | <0.001 | |||
| Lobar | 882 (40.8) | 361 (37.9) | 521 (43.0) | |
| Deep | 1,094 (50.6) | 525 (55.1) | 569 (47.0) | |
| Infratentorial | 187 (8.6) | 66 (6.9) | 121 (10.0) | |
| IVH, n (%) | 794 (36.7) | 255 (26.8) | 539 (44.5) | <0.001 |
| Baseline ICH volume, mL, median (IQR) | 15.3 (6.2–36.0) | 11.7 (5.3–24.7) | 20.5 (7.3–45.6) | <0.001 |
| Follow-up ICH volume, mL, median (IQR) | 17.5 (6.8–43.2) | 13.0 (5.4–27.0) | 24.3 (8.8–57.2) | <0.001 |
| ICH expansion >33% and/or >6 mL, n (%) | 604 (27.9) | 214 (22.5) | 390 (32.2) | <0.001 |
| Relative hematoma growth, n (%) | <0.001 | |||
| None (<0%) | 821 (38.0) | 371 (39.0) | 450 (37.2) | |
| Mild (0%–33%) | 899 (41.6) | 416 (43.7) | 483 (39.9) | |
| Moderate (33.1%–66%) | 213 (9.8) | 103 (10.8) | 110 (9.1) | |
| Severe (>66%) | 230 (10.6) | 62 (6.5) | 168 (13.9) | |
| Absolute hematoma growth, n (%) | <0.001 | |||
| None (<0 mL) | 797 (36.8) | 355 (37.3) | 442 (36.5) | |
| Mild (0–6 mL) | 905 (41.8) | 460 (48.3) | 445 (36.7) | |
| Moderate (6.1–12.5 mL) | 170 (7.9) | 68 (7.1) | 102 (8.4) | |
| Severe (>12.5 mL) | 291 (13.5) | 69 (7.2) | 222 (18.3) | |
Abbreviations: GCS = Glasgow Coma Scale; IQR = interquartile range; IVH = intraventricular hemorrhage; LSW = last seen well; mRS = modified Rankin Scale; NCCT = noncontrast CT; SBP = systolic blood pressure.
Table 2.
Prognostic Impact of Hematoma Expansion
| OR (95% CI) | p Value | |
| rHE | ||
| Outcome: mRS 4–6 at 90 d | ||
| None (<0%) | Reference | |
| Mild (0%–33%) | 0.93 (0.75–1.15) | 0.483 |
| Moderate (33.1%–66%) | 0.77 (0.55–1.08) | 1.129 |
| Severe (>66%) | 1.98 (1.38–2.82) | <0.001 |
| Outcome: mRS 3–6 at 90 d | ||
| None (<0%) | Reference | |
| Mild (0%–33%) | 1.02 (0.82–1.28) | 0.834 |
| Moderate (33.1%–66%) | 0.84 (0.59–1.19) | 0.314 |
| Severe (>66%) | 2.62 (1.72–4.01) | <0.001 |
| aHE | ||
| Outcome: mRS 4–6 at 90 d | ||
| None (<0 mL) | Reference | |
| Mild (0–6 mL) | 0.84 (0.68–1.04) | 0.116 |
| Moderate (6.1–12.5 mL) | 0.92 (0.64–1.34) | 0.676 |
| Severe (>12.5 mL) | 1.73 (1.23–2.45) | 0.002 |
| Outcome: mRS 3–6 at 90 d | ||
| None (<0 mL) | Reference | |
| Mild (0–6 mL) | 0.96 (0.77–1.19) | 0.688 |
| Moderate (6.1–12.5 mL) | 1.09 (0.72–1.64) | 0.670 |
| Severe (>12.5 mL) | 1.64 (1.11–2.41) | 0.013 |
Abbreviations: aHE = absolute hematoma expansion; GCS = Glasgow Coma Scale; ICH = intracerebral hemorrhage; IVH = intraventricular hemorrhage; OR = odds ratio; rHE = relative hematoma expansion.
Multivariable logistic regression was adjusted for age, GCS, ICH volume, ICH location, IVH presence, and variables with p < 0.1 in univariate analysis.
The same results were obtained in secondary analyses defining poor outcome as mRS 3–6 at 3 months from the index event, as reported in Table 2, and restricting the analysis to patients with supratentorial ICH (data not shown).
To examine whether the effect of HE is a function of initial hematoma volume, we stratified our analysis by baseline hematoma volume, and the results are presented in Table 3. Both aHE and rHE were independent prognostic predictors in patients with small (<15 mL) and medium (15–30 mL) baseline ICH volume, whereas in patients with baseline ICH volume greater than 30 mL, even the occurrence of severe HE did not increase the odds of unfavorable prognosis.
Table 3.
Association Between HE and Poor Outcome Stratified by Baseline ICH Volume
| OR (95% CI) | p Value | |
| ICH volume <15 mL (n = 1,064) | ||
| rHE | ||
| None (<0%) | Reference | |
| Mild (0%–33%) | 1.10 (0.81–1.48) | 0.554 |
| Moderate (33.1%–66%) | 0.72 (0.43–1.21) | 0.220 |
| Severe (>66%) | 1.71 (1.07–2.74) | 0.025 |
| aHE | ||
| None (<0 mL) | Reference | |
| Mild (0–6 mL) | 1.02 (0.77–1.35) | 0.923 |
| Moderate (6.1–12.5 mL) | 0.68 (0.32–1.41) | 0.295 |
| Severe (>12.5 mL) | 4.10 (1.80–9.36) | 0.001 |
| ICH volume 15–30 mL (n = 447) | ||
| rHE | ||
| None (<0%) | Reference | |
| Mild (0%–33%) | 0.82 (0.51–1.33) | 0.422 |
| Moderate (33.1%–66%) | 0.77 (0.36–1.66) | 0.505 |
| Severe (>66%) | 3.53 (1.53–8.15) | 0.003 |
| aHE | ||
| None (<0 mL) | Reference | |
| Mild (0–6 mL) | 0.80 (0.49–1.30) | 0.363 |
| Moderate (6.1–12.5 mL) | 0.74 (0.36–1.51) | 0.402 |
| Severe (>12.5 mL) | 2.54 (1.17–5.50) | 0.018 |
| ICH volume >30 mL (n = 652) | ||
| rHE | ||
| None (<0%) | Reference | |
| Mild (0%–33%) | 0.78 (0.51–1.21) | 0.273 |
| Moderate (33.1%–66%) | 0.74 (0.41–1.34) | 0.326 |
| Severe (>66%) | 1.81 (0.77–4.23) | 0.171 |
| aHE | ||
| None (<0 mL) | Reference | |
| Mild (0–6 mL) | 0.60 (0.37–0.97) | 0.035 |
| Moderate (6.1–12.5 mL) | 0.91 (0.48–1.72) | 0.776 |
| Severe (>12.5 mL) | 0.92 (0.55–1.52) | 0.740 |
Abbreviations: aHE = absolute hematoma expansion; GCS = Glasgow Coma Scale; ICH = intracerebral hemorrhage; IVH = intraventricular hemorrhage; OR = odds ratio; rHE = relative hematoma expansion.
Multivariable logistic regression was adjusted for age, GCS, ICH volume, ICH location, IVH presence, and variables with p < 0.1 in univariate analysis. Poor outcome was defined as 90 days modified Rankin Scale 4–6.
Finally, we confirmed the association between HE and outcome analyzing rHE and aHE as continuous variables (data not shown). Figure 3 illustrates the predicted probability of poor outcome and its relationship with different degrees of HE. A significantly higher risk of poor outcome in patients with severe HE was also observed in the subgroup of patients taking antithrombotics on admission (n = 857, 39.6%). Particularly, the mean predicted probability of poor outcome was 0.73 and 0.80 in patients with antithrombotic treatment experiencing severe rHE and aHE, respectively (both p < 0.001 for trend).
Figure 3. Predicted Probability of Poor Outcome.
Poor outcome was defined as modified Rankin Scale 4–6 at 3 months. (A) Relative hematoma expansion. (B) Absolute hematoma expansion.
Discussion
The main finding of this retrospective international multicenter analysis is that the relationship between HE and poor outcome was not uniform across different degrees of HE. There may be a threshold effect, in which HE needs to be severe enough to worsen outcome. We also showed an interaction between baseline ICH volume and the prognostic effect of HE, as even severe HE was not associated with poor outcome in patients with baseline ICH volume above a critical threshold (30 mL). Our findings indicate that the use of a dichotomous definition of HE might dilute the true prognostic effect of severe HE. Therefore, therapeutic strategies to minimize HE might benefit only those patients destined for severe HE, rather than any HE. It may be that there is value in reanalyzing previous trials that analyzed HE as a dichotomous variable to analyze HE as more of a spectrum.8 The results of our analysis might also inform ongoing and future trials as we showed that HE, even in its most severe form, was not significantly associated with worse prognosis in patients with baseline ICH volume above 30 mL. This observation might indicate that ICH volume above a certain cutoff has already determined severe disability, and preventing more bleeding may have diminishing influence on functional outcome. Although most of the ongoing trials exclude patients with baseline ICH volume above 60–70 mL,16-18 the threshold at which HE has a smaller effect on prognosis appears lower. This observation shows the value of focusing on patients with small-to-moderate ICH size in future studies attempting to show clinical benefit and return to functional independence through reduction of HE, despite their overall lower risk of HE compared with larger bleedings.13,19 In our study population, less than 1 in 4 patients presented within 3 hours from onset/last seen well with a baseline ICH volume below 30 mL. Among this subgroup, less than 1 in 5 patients experienced severe HE. These findings indirectly confirm the established challenges into acute ICH trials20,21 and indicate that accurate selection is needed to identify patients at risk of severe HE and, therefore, more likely to benefit from medical therapies targeting active bleeding.22,23 Further research is also needed to accurately predict severe HE, as most of the currently available prediction tools are based on a dichotomic definition of HE (commonly defined as >6 mL and or >33% growth) that does not take into account different degrees of HE severity.22 If the main goal of a clinical trial is to return patients to functional independence, patients with smaller hematoma volume, short time from onset to imaging, and high risk of severe HE appear the population most likely to derive clinical benefit from antiexpansion therapies. Conversely, trials with the main goal of preventing further bleeding and analyzing HE as the primary outcome might have higher statistical power enrolling patients with larger baseline ICH volume.2,24 The use of combined outcome measures, ideally with a predefined hierarchy, might be a valuable approach to capture both HE and clinical outcome.25
Some limitations of our study should be acknowledged. First, systolic blood pressure reductions and fluctuations, especially in the first 24 hours, might modify the risk of HE, and we could only account for admission blood pressure values.26 Second, selection bias might have occurred as follow-up imaging was not mandatory, and some institutions did not enroll patients from intensive care units. These factors might have led to the exclusion of more severely affected patients. Third, coagulopathy reversal was not standardized across sites, and we were not able to adjust for the timing and efficacy of coagulopathy treatment.27 Fourth, clinical outcomes were collected at 90 days, and emerging evidence suggests longer recovery trajectories up to 6–12 months in patients with ICH. Thus, our study is unable to detect separation in recovery trajectories or differences in clinical outcome between the different HE categories beyond the 90-day time window.28,29 Finally, withdrawal or limitation of care is a known predictor of poor outcome in ICH, and we could not account for this factor as well in multivariable analyses.30 In conclusion, we found that only severe HE had an independent prognostic influence in patients with acute ICH while there was no association between mild-to-moderate HE and poor outcome. The relationship between HE and poor outcome was significant only in patients with baseline ICH volume below 30 mL and severe HE. These findings raise the opportunity to reconsider previously tested therapeutic approaches and might inform ongoing and future trials targeting HE.
Glossary
- aHE
absolute HE
- GCS
Glasgow Coma Scale
- HE
hematoma expansion
- ICH
intracerebral hemorrhage
- IVH
intraventricular hemorrhage
- mRS
modified Rankin Scale
- NCCT
noncontrast CT
- OR
odds ratio
- rHE
relative HE
Appendix. Authors
| Name | Location | Contribution |
| Andrea Morotti, MD | ASST Spedali Civili, Brescia, Italy | Study concept and design, data analysis and interpretation; statistical analysis; manuscript drafting and critical revision |
| Gregoire Boulouis, MD, PhD | University Hospital of Tours, France | Data acquisition, analysis and interpretation; critical revision |
| Jawed Nawabi, MD | Charitè Hospital, Berlin, Germany | Data acquisition, analysis and interpretation; critical revision |
| Qi Li, MD, PhD | The First Affiliated Hospital of Chongqing Medical University, China | Data acquisition, analysis and interpretation; critical revision |
| Andreas Charidimou, MD, PhD | Boston University Medical Center, MA | Data acquisition, analysis and interpretation; critical revision |
| Marco Pasi, MD, PhD | University Hospital of Tours, France | Data acquisition, analysis and interpretation; critical revision |
| Frieder Schlunk, MD | Charitè Hospital, Berlin, Germany | Data acquisition, analysis and interpretation; critical revision |
| Ashkan Shoamanesh, MD | McMaster University, Hamilton, Ontario, Canada | Data acquisition, analysis and interpretation; critical revision |
| Aristeidis H. Katsanos, MD | McMaster University, Hamilton, Ontario, Canada | Data acquisition, analysis and interpretation; critical revision |
| Federico Mazzacane, MD | IRCCS Mondino, Pavia, Italy | Data acquisition, analysis and interpretation; critical revision |
| Giorgio Busto, MD | Ospedale Universitario Careggi, Firenze, Italy | Data acquisition, analysis and interpretation; critical revision |
| Francesco Arba, MD, PhD | Ospedale Universitario Careggi, Firenze, Italy | Data acquisition, analysis and interpretation; critical revision |
| Laura Brancaleoni, MD | IRCCS Istituto delle Scienze Neurologiche di Bologna, Italy | Data acquisition, analysis and interpretation; critical revision |
| Sebastiano Giacomozzi, MD | IRCCS Istituto delle Scienze Neurologiche di Bologna, Italy | Data acquisition, analysis and interpretation; critical revision |
| Luigi Simonetti, MD | IRCCS Istituto delle Scienze Neurologiche di Bologna, Italy | Data acquisition, analysis and interpretation; critical revision |
| Andrew D. Warren, BS | Harvard Medical School, Boston, MA | Data acquisition, analysis and interpretation; critical revision |
| Michele Laudisi, MD | Ospedale Universitario S. Anna, Ferrara, Italy | Data acquisition, analysis and interpretation; critical revision |
| Anna Cavallini, MD | IRCCS Mondino, Pavia, Italy | Data acquisition, analysis and interpretation; critical revision |
| M. Edip Gurol, MD, MSc | Harvard Medical School, Boston, MA | Data acquisition, analysis and interpretation; critical revision |
| Anand Viswanathan, MD, PhD | Harvard Medical School, Boston, MA | Data acquisition, analysis and interpretation; critical revision |
| Andrea Zini, MD | IRCCS Istituto delle Scienze Neurologiche di Bologna, Italy | Data acquisition, analysis and interpretation; critical revision |
| Ilaria Casetta, MD, PhD | Ospedale Universitario S. Anna, Ferrara, Italy | Data acquisition, analysis and interpretation; critical revision |
| Enrico Fainardi, MD, PhD | Ospedale Universitario Careggi, Firenze, Italy | Data acquisition, analysis and interpretation; critical revision |
| Steven M. Greenberg, MD, PhD | Harvard Medical School, Boston, MA | Data acquisition, analysis and interpretation; critical revision |
| Alessandro Padovani, MD, PhD | Clinica Neurologica, Università degli Studi di Brescia, Italy. | Data acquisition, analysis and interpretation; critical revision |
| Jonathan Rosand, MD, MSc | Harvard Medical School, Boston, MA | Data acquisition, analysis and interpretation; critical revision; study supervision |
| Joshua N. Goldstein, MD, PhD | Harvard Medical School, Boston, MA | Manuscript drafting; data acquisition, analysis and interpretation; critical revision; study supervision |
Study Funding
No targeted funding reported.
Disclosure
A.H. Katsanos is supported from a McMaster University Department of Medicine Career Research Award. J. Rosand receives research support from NIH and the American Heart Association. He has consulted with Takeda and the National Football League. J.N. Goldstein has received research support from NIH, Pfizer, Takeda, and Octapharma. He has received consulting support from Astrazeneca, CSL Behring, NControl, and Cayuga. All other authors report no relevant disclosures. Go to Neurology.org/N for full disclosures.
References
- 1.Dowlatshahi D, Demchuk AM, Flaherty ML, Ali M, Lyden PL, Smith EE. Defining hematoma expansion in intracerebral hemorrhage: relationship with patient outcomes. Neurology. 2011;76(14):1238-1244. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Al Shahi Salman R, Frantzias J, Lee RJ, 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(10):885-894. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Greenberg SM, Ziai WC, Cordonnier C, et al. 2022 Guideline for the management of patients with spontaneous intracerebral hemorrhage: a guideline from the American Heart Association/American Stroke Association. Stroke. 2022;53(7):e282-e361. [DOI] [PubMed] [Google Scholar]
- 4.Broderick JP, Grotta JC, Naidech AM, et al. The story of intracerebral hemorrhage: from recalcitrant to treatable disease. Stroke. 2021;52(5):1905-1914. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Qureshi AI, Palesch YY, Barsan WG, et al. 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]
- 6.Anderson C, Heeley E, Huang Y, et al. Rapid blood-pressure lowering in patients with acute intracerebral hemorrhage. N Engl J Med. 2013;368(25):2355-2365. [DOI] [PubMed] [Google Scholar]
- 7.Altman DG, Royston P. The cost of dichotomising continuous variables. BMJ. 2006;332(7549):1080. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Yogendrakumar V, Ramsay T, Menon BK, Qureshi AI, Saver JL, Dowlatshahi D. Hematoma expansion shift analysis to assess acute intracerebral hemorrhage treatments. Neurology. 2021;97(8):e755-e764. [DOI] [PubMed] [Google Scholar]
- 9.Morgenstern LB, Hemphill JC, Anderson C, et al. Guidelines for the management of spontaneous intracerebral hemorrhage: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke. 2010;41(9):2108-2129. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Hemphill JC, Greenberg SM, Anderson CS, 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(7):2032-2060. [DOI] [PubMed] [Google Scholar]
- 11.Falcone GJ, Biffi A, Brouwers HB, et al. Predictors of hematoma volume in deep and lobar supratentorial intracerebral hemorrhage. JAMA Neurol. 2013;70(8):988-994. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Hemphill JC, Farrant M, Neill TA. Prospective validation of the ICH Score for 12-month functional outcome. Neurology. 2009;73(14):1088-1094. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Li Q, Yang WS, Shen YQ, et al. Benign intracerebral hemorrhage: a population at low risk for hematoma growth and poor outcome. J Am Heart Assoc. 2019;8(8):e011892. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Hanley DF, Thompson RE, Muschelli J, 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(12):1228-1237. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbrouckef JP. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet. 2007;370(9596):1453-1457. [DOI] [PubMed] [Google Scholar]
- 16.Naidech AM, Grotta J, Elm Jet al. Recombinant factor VIIa for hemorrhagic stroke treatment at earliest possible time (FASTEST): protocol for a phase III, double-blind, randomized, placebo-controlled trial. Int J Stroke. 2022;17:806-809. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Yassi N, Zhao H, Churilov L, et al. Tranexamic acid for intracerebral haemorrhage within 2 hours of onset: protocol of a phase II randomised placebo-controlled double-blind multicentre trial. Stroke Vasc Neurol. 2022;7(2):158-165. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Jiang C, Wang J, Wang J, Zhang J. Rationale and design of a randomized, double-blind trial evaluating the efficacy of tranexamic acid on hematoma expansion and peri-hematomal edema in patients with spontaneous intracerebral hemorrhage within 4.5 h after symptom onset: the THE-ICH trial protocol. J Stroke Cerebrovasc Dis. 2020;29(10):105136. [DOI] [PubMed] [Google Scholar]
- 19.Dowlatshahi D, Smith EE, Flaherty ML, Ali M, Lyden P, Demchuk AM. Small intracerebral haemorrhages are associated with less haematoma expansion and better outcomes. Int J Stroke. 2011;6(3):201-206. [DOI] [PubMed] [Google Scholar]
- 20.Fonville AF, Samarasekera N, Hutchison A, Perry D, Roos YB, Salman RAS. Eligibility for randomized trials of treatments specifically for intracerebral hemorrhage community-based study. Stroke. 2013;44(10):2729-2734. [DOI] [PubMed] [Google Scholar]
- 21.Selim M, Hanley D, Steiner T, et al. Recommendations for clinical trials in ICH: the second hemorrhagic stroke academia industry roundtable. Stroke. 2020;51(4):1333-1338. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Morotti A, Boulouis G, Dowlatshahi D, et al. Intracerebral haemorrhage expansion: definitions, predictors, and prevention. Lancet Neurol. 2023;22(2):159-171. [DOI] [PubMed] [Google Scholar]
- 23.Casolla B, Cordonnier C. Is hyperselection of patients the right strategy? JAMA Neurol. 2019;76(12):1426-1427. [DOI] [PubMed] [Google Scholar]
- 24.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]
- 25.Mayer SA, Frontera JA, Jankowitz B, et al. Recommended primary outcomes for clinical trials evaluating hemostatic agents in patients with intracranial hemorrhage: a consensus statement. JAMA Netw Open. 2021;4(9):e2123629. [DOI] [PubMed] [Google Scholar]
- 26.Moullaali TJ, Wang X, Martin RH, et al. Blood pressure control and clinical outcomes in acute intracerebral haemorrhage: a preplanned pooled analysis of individual participant data. Lancet Neurol. 2019;18(9):857-864. [DOI] [PubMed] [Google Scholar]
- 27.Kuramatsu JB, Gerner ST, Schellinger PD, et al. Anticoagulant reversal, blood pressure levels, and anticoagulant resumption in patients with anticoagulation-related intracerebral hemorrhage. JAMA. 2015;313(8):824. [DOI] [PubMed] [Google Scholar]
- 28.Shah VA, Thompson TE, Yenokyan G, et al. One-year outcome trajectories and factors associated with functional recovery among survivors of intracerebral and intraventricular hemorrhage with initial severe disability. JAMA Neurol. 2022;79(9):856-868. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Foster L, Robinson L, Yeatts SD, et al. Effect of deferoxamine on trajectory of recovery after intracerebral hemorrhage: a post hoc analysis of the i-DEF Trial. Stroke. 2022;53(7):2204-2210. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Zahuranec DB, Brown DL, Lisabeth LD, et al. Early care limitations independently predict mortality after intracerebral hemorrhage. Neurology. 2007;68(20):1651-1657. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Requests to access the data set may be sent to the corresponding author.