Baseline characteristics, in-hospital outcomes and time variances of intracerebral hemorrhage in China: Based on a national, large-scale registry cohort

To the Editor: Intracerebral hemorrhage (ICH) is associated with significantly higher mortality in China[1] relative to ischemic stroke. China has a higher incidence of ICH than Europe and the America;[2] however, few studies have detailed the disease burden of ICH in China, especially data regarding clinical characteristics during hospitalization. As such, the present study aimed to investigate and analyze the clinical characteristics of patients with ICH during hospitalization and variations in time trends between 2015 and 2022. The Chinese Stroke Center Alliance (CSCA) was initiated as a quality improvement program in secondary and tertiary hospitals in China. All physicians undergo a standard training program before data collection. This study was approved by the Ethics Committee of Beijing Tiantan Hospital (No. KY2018-061-02), and the requirement for individual consent was waived due to its retrospective nature. The present study only included data from patients diagnosed with ICH (including both primary [hypertensive, cerebral amyloid angiopathy] and secondary [due to structural vascular lesions, medication, and so on] ICH); as such, those diagnosed with ischemic stroke, transient ischemic attack (TIA), or subarachnoid hemorrhage (SAH) were excluded, along with those diagnosed with an unclassified stroke. Patients were also excluded if key data were incomplete, including age, surgical intervention, and/or geographical region. Study outcomes included in-hospital mortality, death, discharge against medical advice (DAMA), major adverse cardiovascular events (MACEs), and in-hospital complications. MACEs were defined as ischemic stroke, hemorrhagic stroke, TIA, or myocardial infarction. Categorical variables are expressed as numbers (percentage) and numerical variables are expressed as mean ± standard deviation (SD) or median (interquartile range [IQR]). Categorical variables were compared using the chi-squared test or Fisher’s exact test. The Student’s t-test was used for numerical variables with a normal distribution. Nonparametric tests were used for variables with skewed distributions. Considering the extensive sample size, using an absolute standardized difference ≥10 was more appropriate than P <0.05 in demonstrating a clinically significant difference in baseline characteristics. Multiple logistic regression analyses were also performed and included all covariates with a P ≤0.05 in the comparison of baseline characteristics. All statistical analyses were performed using SAS version 9.4 (SAS Institute Inc., Cary, NC, USA). Differences with a two-sided P <0.05 were considered to be statistically significant. Between 2015 and 2022, 1,917,741 patients who experienced stroke or TIA were included in the CSCA program, of whom 1,717,841 experienced ischemic stroke or TIA and 19,625 had SAH. Among the stroke patients who were not classified, 10,553 were excluded. A further 2326 patients with missing age data, 18,658 with missing data regarding surgical intervention, and 13 with missing data regarding geographical region were excluded. Ultimately, the final analysis included data from 148,725 patients diagnosed with ICH. A comparison of the clinical characteristics between nonsurvivors and survivors at discharge is presented in Supplementary Table 1, https://links.lww.com/CM9/C575. A total of 15,377 (10.3%) patients died during hospitalization. Compared with patients who died during hospitalization, those who survived were younger (62.7 ± 12.7 vs. 66.1 ± 13.8 years), exhibited milder neurological impairment status (baseline National Institutes of Health Stroke Scale [NIHSS] score, 8.0 ± 8.5 vs. 22.2 ± 12.3), and longer intervals between onset and admission (17.1 ± 29.9 vs. 7.0 ± 17.8 h). Supplementary Table 2, https://links.lww.com/CM9/C575 summarizes the independent factors related to in-hospital mortality for ICH, including: age (adjusted odds ratio [aOR] 1.017 [95% confidence interval (CI) 1.015–1.019]; P <0.001), NIHSS score (aOR 1.090 [95% CI 1.087–1.092]; P <0.001), onset-to-admission (aOR 0.993 [95% CI 0.992–0.995]; P <0.001), history of ischemic stroke (aOR 1.263 [95% CI 1.178–1.355]; P <0.001), smoking (aOR 0.926 [95% CI 0.867–0.990]; P = 0.024), alcohol consumption (aOR 0.910 [95% CI 0.843–0.981]; P = 0.014), low-density lipoprotein-cholesterol (LDL-c) level (aOR 0.932 [95% CI 0.917–0.948]; P <0.001), fasting blood glucose (FBG) (aOR 1.093 [95% CI 1.083–1.102]; P <0.001), glycated hemoglobin (aOR 0.972 [95% CI 0.957–0.988]; P <0.001); systolic BP (aOR 1.010 [95% CI 1.010–1.011]; P <0.001), dysphagia rehabilitation (aOR 1.334 [95% CI 1.250–1.423]; P <0.001); deep vein thrombosis (DVT) prophylaxis (aOR 1.393 [95% CI 1.320–1.470]; P <0.001); rehabilitation (aOR 0.418 [95% CI 0.395–0.443]; P <0.001); antihypertensive therapy (aOR 0.309 [95% CI 0.29–0.327]; P <0.001), antidiabetic therapy (aOR 0.678 [95% CI 0.612–0.751]; P <0.001), gastrointestinal bleeding (aOR 1.636 [95% CI 1.475–1.815]; P <0.001), decubitus (aOR 1.719 [95% CI 1.352–2.186]; P <0.001), cardiopulmonary arrest (aOR 13.887 [95% CI 11.269–17.114]; P <0.001), hydrocephalus (aOR 2.063 [95% CI 1.808–2.355]; P <0.001), seizure (aOR 1.746 [95% CI 1.469–2.075]; P <0.001), and pneumonia (aOR 2.167 [95% CI 2.053–2.287]; P <0.001). A comparison of clinical characteristics between patients who underwent evacuation for intracranial hematoma (surgical group) and those who did not (medical group) is presented in Supplementary Table 3, https://links.lww.com/CM9/C575. A total of 19,158 patients with ICH (12.9%) underwent intracranial hematoma evacuation during hospitalization. Independent factors associated with undergoing evacuation of intracranial hematoma in ICH are summarized in Supplementary Table 4, https://links.lww.com/CM9/C575, including: age (aOR 0.968 [95% CI 0.967–0.970]; P <0.001); NIHSS score (aOR 1.067 [95% CI 1.065–1.070]; P <0.001; interval from onset to admission (aOR 0.985 [95% CI 0.983–0.986]; P <0.001), FBG (aOR 1.054 [95% CI 1.046–1.062]; P <0.001), systolic BP (aOR 1.004 [95% CI 1.003–1.005]; P <0.001), gastrointestinal bleeding (aOR 1.241 [95% CI 1.128–1.366]; P <0.001), cardiopulmonary arrest (aOR 0.690 [95% CI 0.599–0.795]; P <0.001), hydrocephalus (aOR 3.111 [95% CI 2.777–3.484]; P <0.001), DVT (aOR 1.797 [95% CI 1.569–2.059]; P <0.001), and pneumonia (aOR 2.433 [95% CI 2.321–2.551]; P <0.001). We also summarized the clinical characteristics of patients receiving different surgery types in Supplementary Table 5, https://links.lww.com/CM9/C575. A comparison of the clinical characteristics across different years is presented in Supplementary Table 6 and Supplementary Figure 1, https://links.lww.com/CM9/C575. The proportion of the Eastern region decreased from 2015 to 2022, whereas the proportion of the Middle regions had higher percentages of ICH patients (Pfor trend <0.001). The NIHSS score also increased from 2015 to 2022 (Pfor trend <0.001). The time delay before admission was significantly shortened (Pfor trend <0.001). The comorbidity burden of hypertension and dyslipidemia decreased, whereas that for diabetes mellitus increased. More patients with ICH were taking antiplatelet drugs at the index ICH. The rates of poor outcomes, including in-hospital mortality and DAMA, increased (Pfor trend <0.001). From 2015 to 2022, the rates of ICH among all stroke patients decreased, including all ages and both sexes; however, when comparing NIHSS scores 0–4 v 5–14, those with ICH of NIHSS ≥15 exhibited increased rates [Figure 1].Figure 1: ICH percentages variances from 2015 to 2022 (ICH/all stroke subtypes). (A) Overall. (B) Among different age groups; (C) male vs. female; (D) different NIHSS score. ICH: Intracerebral hemorrhage; NIHSS: National Institutes of Health Stroke Scale.Eastern Asia, particularly China, bears a higher stroke burden than most regions globally.[3] However, data on ICH in China is limited. The CSCA program, aimed at improving stroke care quality, has established a national database covering over 1500 hospitals, representing all provinces and municipalities. This has provided valuable data on stroke, including ICH, across both urban and rural areas. Approximately 8.8% of stroke patients experienced ICH, with a mean age of 63.0 years and 62.6% male. The in-hospital mortality rate for ICH patients in the CSCA program was 10.3%, higher than for ischemic stroke patients.[1] Despite this, mortality at discharge was lower than in the Get With The Guidelines (GWTG)-Stroke program.[4] Factors such as the desire to return home and concerns about medical expenses have led many ICH patients with poor prognoses due to the patient’s refusal to continue inpatient treatment. In addition, ICH patients with mild deficits or favorable prognoses were more likely to be included in the CSCA program, as indicated by lower NIHSS scores compared to the GWTG-Stroke program. From 2015 to 2022, baseline NIHSS scores and in-hospital mortality rates for ICH patients increased. This may reflect improved collaboration with healthcare providers despite poor outcomes. Increased use of antiplatelet medications before ICH also points to advances in vascular risk factor management and ischemic stroke prevention. However, excessive use of antiplatelets raises hemorrhagic risk and worsens outcomes. Over the study period, the proportion of ICH patients with hypertension and dyslipidemia decreased, while use of antihypertensive and antiplatelet medications increased. Tobacco and alcohol consumption also declined, indicating progress in vascular risk management and secondary prevention. Nevertheless, in-hospital complications, such as DVT and pneumonia, rose, likely due to more patients being bedridden or dependent after ICH. The rate of surgical hematoma evacuation also increased, reflecting a shift in ICH etiology from hypertensive to antithrombotic-related causes, contributing to higher NIHSS scores and poorer outcomes.[5] The proportion of ICH patients from eastern China decreased, which may reflect better healthcare access and awareness in these regions. Our study, based on CSCA data, includes over 100,000 ICH patients, a larger sample than other national registries, with data from over 1400 hospitals over 8 years. However, this study had several limitations. First, as a medical quality improvement program for stroke management, the CSCA collects data only on in-hospital outcomes as the study outcomes. Considering the large number of stroke patients (>1,000,000) in the CSCA program, long-term follow-up is impractical for hospital clinicians, and the primary care physician system is not well developed in China. Second, some imaging data, such as hematoma size and peri-hematoma volume, were not collected in the CSCA program due to variability in measurement methods and imaging equipment. In conclusion, our national analysis highlights the clinical characteristics and outcomes of ICH in China, indicating increased in-hospital mortality and rates of DAMA, from 2015 to 2022. Funding This work was supported by grants from the National Natural Science Foundation (Nos. 82401452, 82371302, and 82471489), Beijing Municipal Postdoctoral Work Funding Support (No. 2024-68-167), Beijing Municipal Health Commission’s Excellence Clinical Research Program for Research-Oriented Wards Project (No. BRWEP2024W022040100), Noncommunicable Chronic Diseases-National Science and Technology Major Project (No. 2023ZD0504700), Ministry of Finance of the People’s Republic of China (issued by Finance and Social Security [2015] No. 82; [2016] No. 50; [2017] No. 72; [2018] No. 48; [2019] No. 77; [2020] No. 75; [2021] No. 84, Ministry of Finance), risk factors and assessment techniques for brain aging (No. HX-A-2023037), and Beijing Scholar (No. 097). Conflicts of interest None.