Outcome comparison between surgically treated brain arteriovenous malformation hemorrhage and spontaneous intracerebral hemorrhage

: BACKGROUND Case fatality and poor outcome rates are different between brain arteriovenous malformation associated intracerebral hemorrhage (bAVM-ICH) and spontaneous intracerebral hemorrhage (SICH). These outcome rates, however, have never been compared in patients who need neurosurgical evacuation of the intracerebral hemorrhage (ICH). OBJECTIVE To compare the short- and long-term functional outcome between surgically treated patients with bAVM-ICH and SICH. METHODS We collected data from surgically treated ICH patients at the department of neurosurgery, University hospital Zurich, from January 2015 to July 2018. We performed logistic regression analysis to compare the functional outcome between groups, adjusting for demographics, admission characteristics and stroke risk factors. RESULTS A total of 26 bAVM-ICH and 115 SICH patients were included in the final analysis. Patients with bAVM-ICH were younger and less likely to have hypertension without significant differences in ICH volume, hematoma location, intraventricular hemorrhage and other stroke risk factors. A significantly better functional outcome rate was seen in bAVM-ICH patients at short- and long-term follow-up. These differences remained significant after adjusting for confounders. CONCLUSIONS Patients with a bAVM who need surgical evacuation of an ICH have a more favorable outcome than surgically treated patients with spontaneous ICH, even after correction for confounding factors, such as younger age and less premorbid hypertension. - BACKGROUND: Case fatality and poor outcome rates are different between brain arteriovenous malformation L associated intracerebral hemorrhage (bAVM-ICH) and spontaneous intracerebral hemorrhage (SICH). These outcome rates, however, have never been compared in patients who need neurosurgical evacuation of the intracerebral hemorrhage (ICH). - OBJECTIVE: To compare the short- and long-term functional outcome between surgically treated patients with bAVM-ICH and SICH. - METHODS: We collected data from surgically treated ICH patients at the Department of Neurosurgery, University Hospital Zurich, from January 2015 to July 2018. We performed logistic regression analysis to compare the functional outcome between groups, adjusting for demographics, admission characteristics, and stroke risk factors. - RESULTS: A total of 26 bAVM-ICH and 115 SICH patients were included in the final analysis. Patients with bAVM-ICH were younger and less likely to have hypertension without significant differences in ICH volume, hematoma location, intraventricular hemorrhage, and other stroke risk factors. A significantly better functional outcome rate was seen in bAVM-ICH patients at short- and long-term follow-up. These differences remained significant after adjusting for confounders. - CONCLUSIONS: patients with a bAVM who need surgical evacuation of an ICH have a more favorable outcome than surgically treated patients with spontaneous ICH, even after correction for confounding factors, such as younger age and less premorbid hypertension.

-OBJECTIVE: To compare the short-and long-term functional outcome between surgically treated patients with bAVM-ICH and SICH.
-METHODS: We collected data from surgically treated ICH patients at the Department of Neurosurgery, University Hospital Zurich, from January 2015 to July 2018. We performed logistic regression analysis to compare the functional outcome between groups, adjusting for demographics, admission characteristics, and stroke risk factors.
-RESULTS: A total of 26 bAVM-ICH and 115 SICH patients were included in the final analysis. Patients with bAVM-ICH were younger and less likely to have hypertension without significant differences in ICH volume, hematoma location, intraventricular hemorrhage, and other stroke risk factors. A significantly better functional outcome rate was seen in bAVM-ICH patients at short-and long-term follow-up. These differences remained significant after adjusting for confounders.
-CONCLUSIONS: patients with a bAVM who need surgical evacuation of an ICH have a more favorable outcome than surgically treated patients with spontaneous ICH, even after correction for confounding factors, such as younger age and less premorbid hypertension. INTRODUCTION I ntracerebral hemorrhage (ICH) is a devastating disease that can lead to high mortality and morbidity. The incidence of ICH is approximately 25 per 100,000 person-years, and it has a mortality of 40% within 1 month of presentation. ICH can be classified as primary or secondary, depending on the etiology of hemorrhage. Primary ICH is most commonly caused by chronic arterial hypertension, whereas secondary ICHs are caused by underlying lesions, such as aneurysms, vascular malformations, or tumors. 1 A brain arteriovenous malformation (bAVM) is a vascular malformation, which most often causes clinically significant ICH. Although the incidence of bAVM is rare, about 1.12e1.42 per 100,000 person-years, bAVM-associated ICH (bAVM-ICH) has a high impact on the socioeconomic and health care system. [2][3][4][5] The annual risk of bAVM-ICH is approximately 2.3%, which increases to a rate of 4.8% after a previous hemorrhage. [6][7][8] Reported case fatality and poor outcome rates after bAVM-ICH range between 5% and 25% and 10% and 30%, respectively. 3,[9][10][11] This is lower than reported outcomes after spontaneous ICH (SICH), with a mortality rate of 35%À52% in 30 days and 15%À30% independent functional outcome after 1 year. 3,[12][13][14][15][16][17] Recently, the arteriovenous malformationÀrelated intracerebral hemorrhage score was externally validated and showed better performance in predicting clinical outcome when compared with the ICH score. 18 As such, Key words -Arteriovenous malformation -Cerebral hemorrhage -Outcome Abbreviations and Acronyms bAVM: Brain arteriovenous malformation GCS: Glasgow Coma Scale ICH: Intracerebral hemorrhage mRS: Modified Rankin Scale SICH: Spontaneous intracranial hemorrhage treatment and management decisions, as well as providing information about the prognosis after bAVM-ICH, should probably not be based on the results from SICH.
There have been some studies on comparison of morbidity and mortality of bAVM-ICH and SICH and showed that patients with bAVM-ICH have a more favorable outcome than patients with SICH from other causes, independent of patient age and other known predictors of ICH outcome. 3,5 These studies included all various causes of ICH including intracranial aneurysms and the patients in both good and poor clinical admission status. 5,19 However, patients who need surgical evacuation of their ICH are generally in worse clinical condition at admission, which has influenced the clinical outcome. No studies that directly compared the clinical outcome of surgically treated bAVM-ICH with surgically treated SICH were found. We aimed to investigate whether surgically treated patients with bAVM-ICH fare better than SICH patients regarding short-and long-term functional outcomes.

Study Design and Data Collection
We performed a retrospective analysis of prospectively collected patient data of patients who underwent semiemergent (within 1 week after diagnosis) neurosurgical evacuation of their ICH at the department of neurosurgery, University Hospital Zurich from January 2015 to July 2018. Patients younger than 18 years and patients with ICH due to traumatic brain injury or cerebrovascular diseases other than bAVM, such as a ruptured intracranial aneurysm, were excluded.
Baseline data consisted of patient characteristics, hematoma parameters, and outcome variables, according to the review article of van Beijnum et al. 3 Patient characteristics included age, sex, Glasgow Coma Scale (GCS), and systolic and diastolic blood pressure at admission. GCS was divided into 3 groups according to arteriovenous malformationÀrelated intracerebral hemorrhage score. 20 Hematoma parameters included location (cortical vs. deep vs. infratentorial), side and size (measured by the ABC/2 method in native brain CT), and presence of intraventricular hemorrhage (IVH). 21 Outcome variables included case fatality and modified Rankin Scale (mRS, assessed by trained physicians) at 6 weeks, 6 months, and 12 months after treatment. A mRS of !3 was defined as a poor outcome.
Missing data were retrospectively collected from our hospital information system and telephone interviews with the patient or general practitioner. Ethical approval for this study was obtained from the local committee (BASEC number: PB_2017-00093). Informed consent for collecting data for scientific purposes was retrieved through a general consent form, which patients have signed.

Statistical Analysis
The differences in baseline characteristics between the 2 groups were evaluated with Pearson c 2 test for categorical variables.
Continuous variables were compared using parametric statistics when data showed normal distribution and nonparametric statistics when they did not. We compared outcomes using odds ratios (ORs) with 95% confidence intervals (CIs) at 6 weeks and 6 and 12 months after surgery. The differences were considered significant at a P value of < 0.05.
Multivariate logistic regression analysis was performed to evaluate the association between bAVM-ICH and SICH outcomes, with a backward elimination procedure to identify significant confounders. Cutoff values of P < 0.1 were used to select the initial set of variables to be included in the initial multivariable model. All statistical analyses were performed with the SPSS statistical software package (IBM SPSS Statistics, version 25).

Baseline Characteristics
In total, 31 bAVM-ICH and 115 SICH patients were extracted from the database. Five bAVM-ICH patients were excluded since they had their surgical treatment more than 1 week after hemorrhage. One SICH patient was lost to follow-up after the 6-weeks' outcome assessment due to emigration and is therefore not included in the 6-months' and 12-months' outcome analysis. The final analysis included 26 bAVM-ICH and 115 SICH patients.
Baseline characteristics are shown in Table 1.
Patients with bAVM-ICH were significantly younger (45.5 AE 17.9 vs. 61.6 AE 14.3 years, P < 0.001) and less likely to have premorbid hypertension (32.3% vs. 54.8%, P ¼ 0.026) compared with SICH patients. There were no significant differences in sex, ICH volume, hematoma location, intraventricular hemorrhage, and other stroke risk factors, such as diabetes mellitus and hyperlipidemia between studied groups. Table 2 and Figure 1 present the results of the outcome analysis.

Functional Outcome and Case Fatality
Multivariate analysis was done including the following parameters: age, side, hypertension, and diabetes mellitus. The proportion of SICH patients who were dead or dependent (mRS ! 3) was significantly higher than bAVM-ICH patients at short-and long-term follow-up. After stratified analysis, the rate of death or dependence remained significantly higher in SICH patients at all follow-up timepoints. There was no difference in case fatality between groups at any follow-up timepoint ( Table 3).

DISCUSSION
In our study, we found that patients with bAVM-ICH showed better functional outcome compared with SICH patients at shortterm (6 weeks) and long-term (6 and 12 months) follow-up. We concluded that bAVM patients who need semiemergent surgical evacuation of their ICH have a better prognosis than surgically treated patients with SICH. Our data suggest that treatment and management decisions, as well as information about the prognosis after bAVM-ICH, should not be based on results of SICH research.
Our study consolidated findings from previous studies that have found a better outcome after bAVM-ICH compared with SICH. [2][3][4][5] Our study, however, only included ICH patients who needed surgical evacuation of the ICH, indicating patients in a more severe clinical condition. This is reflected in the worse GCS at Patients with bAVM-ICH are younger than SICH patients and bear fewer risk factors, which can affect outcome after ICH, such as hypertension and diabetes mellitus. Nevertheless, even after adjusting for those confounders the patients with bAVM-ICH remained having a more favorable functional outcome. [22][23][24] One reason for a better prognosis in bAVM patients is that the malformation mostly ruptures within the nidus itself or in the venous side of malformation, leading to a bleeding with lower pressure instead of acute arterial hemorrhage due to hypertension. 5,[22][23][24] Secondly, a significant part, or even the complete, bAVM is located in a sulcus, leading to a sulcal hemorrhage with less damage of the brain parenchyma than SICH, except for the All parameters are presented in numbers with percentages between brackets, unless otherwise stated. bAVM-ICH, brain arteriovenous malformationÀassociated intracerebral hemorrhage; SICH, spontaneous intracerebral hemorrhage; SD, standard deviation; GCS, Glasgow Coma Scale; SBP, systolic blood pressure; DBP, diastolic blood pressure. compressive effect of the hematoma itself. Another reason might be the higher proportion of infratentorial hemorrhages in the SICH group, which is associated with a worse clinical outcome. Other studies regarding this topic have shown results with more favorable rates for bAVM-ICH patients at discharge or 1 month after hemorrhage. 3,5 The difference with these studies is explained by the fact that patients in our study were admitted in worse clinical condition and needed emergent hematoma evacuation.
One of the limitations of the current study is that, since it is a single-center study, the sample size of bAVM-ICH is relatively small, which may have introduced selection bias. In addition, the indication for surgical evacuation of the ICH was set by the treating physician, which might also have introduced a selection bias. In addition, missing data were retrospectively collected from our hospital information system and telephone interview with the patient or the general practitioner, which can lead to recall bias, since they may remember patients' conditions wrongly at different time points.

CONCLUSION
Brain AVM patients who need surgical evacuation of an intracerebral hematoma have a more favorable outcome than surgically treated patients with spontaneous ICH at short-and long-term follow-up (6 weeks and 12 months), even after correction for confounding factors, such as younger age and less premorbid hypertension. A thorough radiologic investigation of the underlying pathology of intracerebral hematoma is mandatory to assess functional outcome after surgical evacuation. We suggest that treatment and management decisions, as well as information about the prognosis after bAVM-ICH, should not be based on results of SICH research.

ACKNOWLEDGMENTS
Elisabeth Jehli and Johannes Sarnthein developed the quality registration database of the Department of Neurosurgery in which patient data are collected prospectively.