Aspirin does not Affect the Hematoma Volume and Growth in Severe Spontaneous Intracranial Hematoma


 Baseline hematoma volume (HV) and hematoma growth (HG) affect the prognosis of spontaneous intracranial hematoma (ICH), but there is still a lack of evidence on the effects of aspirin (acetylsalicylic acid, ASA) on HV and HG in patients with severe ICH. This study retrospectively analyzed patients with severe ICH who met the inclusion and exclusion criterias in Beijing Tiantan Hospital, Capital Medical University between January 1, 2015 and July 31, 2019. Severe ICH patients were divided into ASA group and nASA group according to ASA usage, and the effects of prior ASA on HV and HG were evaluated respectively. And possible risk factors for HG were screened. Totally, 221 patients with severe ICH were consecutively enrolled into this study. There were 72 (32.58%) patients in ASA group and 149 patients (67.42%) in nASA group. Although the baseline HV of the nASA group was significantly higher than that of the ASA group (45.51±29.76 VS 32.67±25.85, p=0.001), after adjusting for confounding factors, the prior ASA did not significantly affect the baseline HV (p = 0.057). Similarly, although the incidence of HG in nASA group was higher than that in ASA group (36.2% VS 25.0%, p = 0.095), ASA did not significantly affect the occurrence of HG (p =0.057). In addition, we found that admission high blood pressure and GCS were risk factors for HG. Prior ASA does not increase the baseline HV and the incidence of HG in severe ICH patients. In addition, admission high blood pressure and GCS were risk factors for HG.


Introduction
Spontaneous intracranial hemorrhage (ICH) has a poor outcome than ischemic stroke [1,2], especially severe ICH which usually leads to severe disability or death [3,4]. According to researches, hematoma growth (HG) after the onset of ICH was an important poor prognostic factor. [5][6][7] Aspirin (acetylsalicylic acid, ASA) is the most widely used antiplatelet drug. Its e cacy has been proven in primary and secondary stroke prevention. [8,9] However, the effective inhibition of platelet function by ASA may adversely affect certain bleeding disorders. [10][11][12] With the increasing population of receiving prior ASA at the time of ICH onset, the impact of prior ASA on hematoma volume (HV) and HG in ICH patients has aroused discussion among researchers. [13] Although some observational studies [14][15][16] have analyzed the effects of prior ASA on HV and HG before, there is still no consensus yet. Evidence in this area is still scarce, especially for patients with severe ICH. Therefore, we designed a retrospective cohort study to explore the effects of prior ASA on HV and HG in severe ICH patients.

Materials And Methods
This retrospective study was approved by the Institutional Review Board of Beijing Tiantan Hospital, Capital Medical University (reference number: KY2019-096-02).

Study Population
Severe ICH patients admitted to Beijing Tiantan Hospital, Capital Medical University between January 1, 2015 and July 31, 2019 were continuously enrolled into this study referred to the inclusion and exclusion criteria. The informed consent had all been signed by enrolled patients or their family members.

Inclusion and Exclusion Criteria
Inclusion criteria: (i) patients were diagnosed with severe ICH; (ii) patients were older than 18 years.
Exclusion criteria: (i) sever ICH was caused by aneurysm, moyamoya disease, vascular malformation, tumor, cerebral venous thrombosis or hemorrhagic transformation of ischemic infarction; (ii) patients were with hereditary coagulopathy; (iii) patients were with coagulation dysfunction attributed to malignant tumor, liver or renal dysfunction; (iv) patients taken other antiplatelet drugs other than only ASA or receiving anticoagulant therapy (such as, vitamin K antagonist (Warfarin), dabigatran or rivaroxaban) in 7 days; (v) The time from onset of symptoms to admission computed tomography (CT) was more than 24 hours.

Data Collection
Severe ICH patients' baseline characteristics were collected from the electronic medical records of Beijing Tiantan Hospital, Capital Medical University. Including (i) demographic characteristics: age, gender; (ii) vascular risk factors: smoking, alcohol, admission high blood pressure, admission low blood pressure; (iii) medical history: hypertension, diabetes, coronary heart disease, ischemic stroke history, cerebral hemorrhage history, history of ASA; (iv) radiographic characteristics of admission CT and follow-up CT in 6 days after ICH onset; (v) functional status: admission Glasgow Coma Scale (GCS).
The admission CT was regarded as baseline CT and the follow-up CT was regarded as control CT. All patients' CT was analyzed by two independent neurosurgeons who were blinded to the clinical and anamnestic histories. The following data was collected from the admission CT: (i) hemorrhage side (ii) hemorrhage localization; hemorrhage localization was divided into shallow and deep. When hemorrhage was located in basal ganglia, thalamus, internal capsule, cerebellum and brainstem, it was de ned as deep; when IH was located in other areas, it was de ned as shallow. (iii) whether combined with ventricular hematoma; (iv) whether combined with subarachnoid hemorrhage; (v) whether combined with hydrocephalus; (vi) whether combined with blend sign; [17] (vii) whether combined with black hole sign; [18] (viii) whether combined with satellite sign; [19] (ix) whether combined with irregular shape; [20] (x) whether combined with heterogeneous hematoma density [20]; (xi) HV (mL, including intraventricular hematoma); HV was measured by accumulating the product of slice thickness multiplied by the hematoma area of each CT slice with hematoma. The hematoma area of each CT slice was automatically measured by CT software (Neusoft PACS/RIS version 5.5.5.70613) after manually sketching the range of the hematoma. HV change was calculated by subtracting the baseline HV from the ICH volume on the control CT scan.

De nition and outcomes
Severe ICH was de ned as supratentorial hematoma volume >30ml, infratentorial hematoma volume >10ml or midline shift >1cm. [4] Prior ASA was de ned as taking ASA continuously for more than 7 days before the onset of ICH and the interruption time less than 7 days. [4,21,22] HG was de ned as a proportional increase ≥ 33% or an absolute HV increase ≥ 6 mL at any control CT scan in 6 days after ICH onset. [23] Statistical Analysis In this study, all statistical methods were done by SPSS statistical software (IBM, version 26). In groups comparison, categorical variables were analyzed by chi-square test or Fisher exact test and continuous variables were analyzed by Mann-Whitney U test. Regression analysis was used to adjusted for confounding variables. Univariate analysis was used to analyze the relationship between variables and HG; multivariate logistics regression analysis was used to screen risk factors for HG. Two-side p value 0.05 was considered statistically signi cant.

Study population
Totally, 1548 adult patients were diagnosed with ICH between January 1, 2015 and July 31, 2019 in Beijing Tiantan Hospital, Capital Medical University; 354 patients were excluded due to a diagnosis of aneurysm, arteriovenous malformation, or moyamoya disease etc.; 883 patients were excluded because they were not severe ICH; 43 patients were excluded because the time from the onset of ICH to their rst CT scan was more than 24 hours; 27 patients were excluded because they took other antiplatelet or drugs (such as, vitamin K antagonist (Warfarin), clopidogrel) in 7 days; 9 patients were excluded because they were accompanied by coagulopathy, liver or kidney dysfunction; 11 patients were excluded duo to missing data. Finally, 221sever ICH patients were consecutively enrolled into this study.

ASA Prevalence and Characteristics
There were 50 females and 171 males in this study. Among the study population, there were 72 (32.58%) patients taking ASA within 7 days before sever ICH onset. According to ASA usage, the study population were divided into the ASA group and the nASA group. Their baseline characteristics were summarized in table 1. Compared with nASA group, patients in ASA group were older and more likely to develop diabetes, coronary heart disease, ischemic stroke, and cerebral hemorrhage. There was no signi cant statistical difference in other variables between nASA group and ASA group.

Association between prior ASA and HV, HG
The results of analysis patients' admission CT showed baseline HV (45.51±29.76) of severe ICH patients in nASA group was signi cantly higher than that (32.67±25.85) of ASA group (p = 0.001), as shown in table 2. However, after adjusting for factors of age, diabetes, coronary heart disease, ischemic stroke, cerebral hemorrhage and hydrocephalus, the impact of ASA on baseline HV did not reach statistical differences (p =0.057, 95% con dence interval (CI), -10.042 (-20.376, 0.292)), as shown in table 3. Table 2 showed although the statistical difference was not reached, the ratio (54/149, 36.2%) of HG in nASA group was higher than that (18/72, 25.0%) in ASA group. Similarly, after adjusting for factors of age, diabetes, coronary heart disease, ischemic stroke, cerebral hemorrhage and hydrocephalus, ASA did not signi cantly affect HG of sever ICH patients (p =0.057, 95% CI, -10.042 (-20.376, 0.292)), as shown in table 3.

Risk Factors of HG
The study population were divided into HG group and nHG group according to the change of HV; their characteristics were summarized in table 4. In univariate analysis, HG were signi cantly correlated with admission high blood pressure, hydrocephalus, HV and GCS (p 0.05), as shown in table 4.
To screen risk factors of HG, variables with p 0.1 in univariate analysis (shown in table 4) were included into multivariate analysis. After adjusting for factors of admission high blood pressure, coronary heart disease, ASA, hydrocephalus, HV and GCS, the predictive indicators of HG were admission high blood pressure and GCS, as shown in table 5.

Discussion
In the current study, we retrospectively evaluated the effects of prior ASA on HV and HG in severe ICH patients. As demonstrated here, prior ASA did not signi cantly affect the baseline HV and HG in severe ICH patients, which was consistent with the results of previous reports [24,25]. Besides, the results showed admission high blood pressure and GCS were risk factors for HG in severe ICH patients.
Previous studies were con icting in assessing the effects of ASA on HV and HG. In 2016, a prospective study and meta-analysis showed that prior antiplatelet therapy (APT) increased the baseline HV and increased the risk of HG in ICH patients.
[16] In addition, a systematic review of more than 5000 ICH patients also showed APT was an independent predictor of HG.
[26] Yet, these two studies had a selection bias in their enrolled population and their result may have certain limitations. Otherwise, increasing evidences suggested APT had no signi cant effect on baseline HV and HG in ICH patients. [24,25] In 2018, Verena van Ginneken et al. did not nd signi cant effects of APT on baseline HV (p = 0.32) or HG (OR 0.8, 95% CI 0.4-1.9). [25] Our conclusions consistent with that. Recently, a study about effects of ASA on cerebral hemorrhage in mice also suggested the application of ASA did not increase the size of cerebral hemorrhage in mice. [27] Moreover, although combination APT was associated with a poor prognosis, single APT did not affect the prognosis of ICH patients, according to reports.
[28] These evidences showed that prior ASA is safe in ICH patients.
Unlike current study, Michael Moussouttas et al. found although there was no statistical signi cance, ICH patients receiving prior APT had a greater absolute HV and a higher proportion of HG compared with that not receiving prior APT. [24] In current study, the baseline HV of sever ICH patients in ASA group was signi cantly smaller than that in nASA group and the ratio of HG in ASA group was lower than that in nASA group. The underlying reasons of this difference may be the HV calculation method or the APT withdrawal effect. Previous studies [14,15,26,29] mostly used A×B×C/2 method [30]to calculate HV.
Although this method has high sensitivity and speci city when measuring cerebral parenchymal hematoma, its accuracy will be reduced when measuring intraventricular HV. [30] In this study, intraventricular hematoma was included in total HV, which may be related to the above difference. According to reports, studies have shown that the sudden withdrawal of APT may increase the sensitivity of platelets to some stimuli, thereby enhancing blood coagulation and preventing further bleeding after the occurrence of ICH. [31] Recently, Santosh Murthy et al. [14] and Maximilian I. Sprügel et al. [32] also observed similar results, although they did not provide a reasonable explanation for it.
HG is associated with mortality of ICH patients [7] and effective control of risk factors for HG may reduce its occurrence. Previous studies have reported some independent risk factors for HG, such as low hemoglobin [33], serum magnesium level [34], spot sign[6] etc.. As current study showed admission high blood pressure and GCS were also risk factors for HG in severe ICH patients. These ndings were consistent with results of previous researches [1,26,35]. Particularly, admission high blood pressure was critical to the outcome of ICH patients. A large number of prospective studies have demonstrated that effectively reducing hypertension can signi cantly improve the prognosis of ICH patients.[36-38] It is worth noting that some special signs such as blend sign [17], black hole sign [41], satellite sign [42], hematoma shape and density [20] etc.of non-enhanced CT can be used as reliable risk factors for HG, which has attracted the attention of researchers. [39,40] Unfortunately, no similar differences were observed between HG and nHG groups in this study population. The reason for this may be that nonsevere ICH patients were excluded in this study, thereby reducing the difference between the two groups. According to reports, early HG mainly occurred within the rst 7 to 8 hours after the onset of ICH. It is reported that early HG mainly occurred within the rst 7 to 8 hours after the onset of ICH[43], our study extended the time to receive CT to hospital to 24 hours. This means that some ICH patients in this study already had HG before they received CT in the hospital, which may also reduce the difference between the two groups.
This study has several advantages. The population only included patients with severe ICH, which can better re ect the effects of ASA on severe ICH. Besides, our study excluded ICH patients taking anticoagulant drugs and antiplatelet drugs other than aspirin, thereby avoiding the in uence of other drugs on the results of the study. This research also has some shortcomings. The time of occurrence of HG was not included in the analysis. In addition, our study is a single-center retrospective study, which has certain limitations.

Conclusion
In conclusion, prior ASA does not increase the baseline HV and the incidence of HG in severe[] ICH patients. Admission high blood pressure and GCS were risk factors for HG.

Declarations Con icts of Interest
Authors declare no con icts of interest.

Availability of data and material
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Ethics approval
This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Committee of Beijing Tiantan Hospital, Capital Medical University (reference number: KY2019-096-02).

Consent to participate
Informed consent was obtained from all individual participants included in the study.

Consent for publication
The participant has consented to the submission of the results.

Authors' contributions
All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Junhua Yang, Kaiwen Wang and Shaohua Mo. The rst draft of the manuscript was written by Junghua Yang, and all authors commented on previous versions of the manuscript. All authors read and approved the nal manuscript.