Spontaneous intracranial hemorrhage (ICH) is defined as a condition resulting from the rupture of small intracranial blood vessels due to non-traumatic causes (Qureshi et al., 2009; Wilkinson et al., 2018). It ranks as the second most common type of stroke; however, its mortality and disability rates exceed those of ischemic stroke. According to the Global Burden of Disease Study, the incidence of ICH in China is approximately 45/100,000 (GBD 2019 Stroke Collaborators, 2021). Reports indicate that by 2030, the incidence of cerebrovascular disease events in China will increase by approximately 50% compared with 2010 (Tu et al., 2023), thereby imposing a substantial burden on society and families.

Pharmacological and surgical interventions both serve as effective components of comprehensive treatment for ICH (Wu et al., 2020; Zhou et al., 2011). However, for patients with severe ICH, surgery may represent a more favorable option. Clinical studies have demonstrated that rapid removal of the intracranial hematoma can significantly enhance the prognosis in patients with severe ICH (Wu et al., 2020; Zhou et al., 2011; Shi et al., 2019). However, the occurrence of postoperative major adverse cerebrovascular events (post-MACEs) can markedly diminish the surgical benefits for patients with ICH (Shen et al., 2021; Wu et al., 2017; Ren et al., 2018). Therefore, effectively reducing the occurrence of post-MACEs constitutes a critical strategy to improve surgical outcomes.

It has been reported that approximately 14% of ICH patients are comorbid with diabetes mellitus (DM) (Tu et al., 2023). Pathological studies have indicated that small vessel injury represents a critical mechanism by which DM affects multiple organs throughout the body (Antar et al., 2023). Preclinical investigations into ICH have further demonstrated that DM can influence intracranial edema and bleeding volume by altering cerebral vascularization (Antar et al., 2023; Bahadar and Shah, 2021; Gómez-de Frutos et al., 2024). Based on these findings, we hypothesized that DM may contribute to an increased incidence of post-MACEs in ICH patients via its impact on small intracranial blood vessels. Indeed, an increasing number of clinical studies have shown that DM is associated with poorer prognoses in ICH patients (Antar et al., 2023; Liebkind et al., 2018; Boulanger et al., 2016; Cavallari et al., 2021). However, to date, there remains a paucity of clinical evidence regarding the specific effects of DM on post-MACEs in the ICH population. Therefore, to evaluate the potential influence of DM on post-MACEs in ICH patients, we designed this multicenter cohort study.

This study is a multicenter retrospective study. As this study is a retrospective analysis and poses no risk to participants, the Ethics Committee of Peking University Third Hospital waived the requirement for informed consent.

To evaluate the impact of DM on post-MACEs in patients with ICH, we conducted this retrospective cohort study. As demonstrated here, DM was identified not only as a risk factor for overall post-MACEs, but also as an independent risk factor for ischemic post-MACEs and hemorrhagic post-MACEs, respectively. Additionally, our analysis confirmed that DM is an independent risk factor for 30-days mortality in ICH patients.

The correlation between DM and ischemic post-MACEs has consistently drawn the attention of clinical researchers. Spiotta et al. (2011) reported that the patients with DM appeared to be more susceptible to post-MACEs during a retrospective analysis of interventional treatment for wide-neck aneurysms. Similarly, clinical studies on unruptured middle cerebral artery aneurysm clipping surgery have also observed this association (Lee et al., 2021). In 2021, a study on acute coronary syndrome revealed that type 2 DM increases the risk of MACEs, regardless of insulin use (Cavallari et al., 2021). Another retrospective study by Pantelis Stavrinou et al., which reviewed 218 carotid endarterectomy procedures, found that patients with DM had approximately 2.3 times higher risk (95% CI: 1.13–10.09) of experiencing ischemic post-MACEs within 30-days after surgery compared to nDM patients (Stavrinou et al., 2016). These findings align with our results, where we identified that the risk of ischemic post-MACEs in DM patients was approximately 2.14 times higher (95% CI: 1.090–4.197) than in nDM patients. Although the aforementioned studies (Cavallari et al., 2021; Spiotta et al., 2011; Lee et al., 2021; Stavrinou et al., 2016) primarily focused on cardiovascular and cerebrovascular populations rather than ICH patients, pathological evidence suggests that ICH is often caused by intracranial small vessel disease (Qureshi et al., 2009). Thus, the conclusions remain consistent.

The exact effect of DM on hemorrhagic post-MACEs remains controversial. Indeed, clinical evidence regarding the impact of DM on hemorrhagic post-MACEs in ICH patients is extremely limited. Cavallari et al. (2021) evaluated the influence of DM on hemorrhagic post-MACEs in patients with acute coronary syndrome. Their results indicated that only DM treated with insulin was a risk factor for hemorrhagic post-MACEs (Cavallari et al., 2021). This may be attributed to the elevated blood glucose levels observed in DM patients receiving insulin therapy, as compared to those not undergoing such treatment (Antar et al., 2023; Cavallari et al., 2021). Conversely, Hao et al. (2019) conducted a meta-analysis of 42 studies involving 10,001 cases of mechanical thrombectomy. The findings revealed that DM is a risk factor for all forms of post-MACEs (Hao et al., 2019). Similarly, Zhang et al. (2022) analyzed patients undergoing mechanical thrombectomy and found that hyperglycemia increased the occurrence of post-MACEs. These results are consistent with our study findings. Moreover, similar conclusions have also been observed in populations without cardiovascular or cerebrovascular diseases. For instance, it has been reported that DM increases postoperative bleeding complications after renal puncture (Hasegawa et al., 2022) and vitreoretinal surgery (Lauermann et al., 2021). Interestingly, these areas are also susceptible to DM (Antar et al., 2023; Demir et al., 2021; Yu et al., 2024; Cole and Florez, 2020). Therefore, it is reasonable to speculate that the influence of DM on hemorrhagic post-MACEs may be associated with small vessel injury caused by DM.

In our study, we also found that DM significantly impacts the 30-days mortality of ICH patients. Similarly, Boulanger et al. (2016) conducted a meta-analysis of 18 cohort studies involving 813 DM patients and 3,714 nDM patients. The results indicated that DM was associated with a higher 30-days mortality (1.52, 95%CI 1.28–1.81) (Boulanger et al., 2016). Subsequently, a retrospective cohort study conducted in Scotland between 2004 and 2013 among individuals aged 40 to 89 years further confirmed that both type 1 DM and type 2 DM increased the in-hospital mortality rate of ICH patients (Boulanger et al., 2017). Forti et al. (2020) categorized ICH patients into Pre-DM, DM, and nDM groups and investigated their 30-days mortality. The results showed that both Pre-DM and DM significantly elevated the 30-days mortality. However, Liebkind et al. (2018) did not observe a significant impact of DM on 30-days mortality but instead noted that DM significantly increased the one-year mortality of ICH patients. Similarly, no independent association between DM and 30-days mortality was observed in studies by Nilsson et al. (2002) and Lee et al. (2010). This discrepancy may be related to stress-induced hyperglycemia in ICH patients (Liang et al., 2024). As reported in a multicenter prospective study, stress-induced hyperglycemia was a stronger predictor of 30-days and 90-days mortality in ICH patients, compared with DM (Chen et al., 2023). Hypoglycemic treatment for DM has been shown to improve the prognosis of ICH patients (Chiu et al., 2025; Proietti et al., 2023), supporting this hypothesis. Therefore, researchers recommend implementing effective blood glucose management during the perioperative period of ICH to reduce the occurrence of post-MACEs and mortality.

With the aging of society, the impact of gender and age on disease prognosis has increasingly drawn the attention of clinical researchers. A study indicates that female with DM face a 27% higher relative risk of stroke compared to male, and they also have a higher mortality rate at age of 40 (Kautzky-Willer et al., 2016). These findings suggest that gender and age differences may amplify the effects of DM on study outcomes. However, in our multivariate analysis, gender and age did not demonstrate a statistically significant association with post-MACEs and 30-days mortality. This implies that in the studied population, gender and age did not significantly increase the occurrence rates of post-MACEs and 30-days mortality. In addition, the severity of DM seems to also have an impact on the research results. It is reported that among patients with acute coronary syndrome, the odds of developing bleeding complications in DM patients receiving insulin treatment is 2.3 times that of nDM patients (Cavallari et al., 2021); however, no statistical difference was observed between DM patients not receiving insulin treatment and nDM patients (Cavallari et al., 2021). This might be related to the fact that DM patients receiving insulin treatment have higher blood sugar levels. Manosroi et al. (2023) in a prospective study analyzed the relationship between glycated hemoglobin (HbA1c) variations and the risk of cardiovascular events in the pre-DM and DM patient populations; the results show that the highest quartile of HbA1c variability showed a statistically significant association with cardiovascular events (Manosroi et al., 2023). These studies all suggest that the blood glucose level and stability of DM can have an impact on post-MACEs (Cavallari et al., 2021; Manosroi et al., 2023). However, due to some reasons, we did not conduct further stratified studies on the blood glucose levels, HbA1c levels and insulin use. Therefore, more clinical evidence and deeper research are still needed for further exploration.

To explore the potential mechanisms underlying the effects of DM on post-MACEs, Terenzi et al. (2019) developed a high-throughput flow cytometry-based assay to quantify circulating proangiogenic and proinflammatory cell content in the peripheral blood of individuals with type 2 DM. Consistently, they detected an increased inflammatory cell burden and decreased pro-vascular progenitor content in individuals with type 2 DM, which may lead to an increase in microvascular fragility and endothelial dysfunction (Terenzi et al., 2019). Subsequently, Gómez-de Frutos et al. (2024) confirmed through preclinical studies that DM can increase cerebral edema by affecting cerebral vascularization in mice. In addition, some researchers has also confirmed that neuroinflammation, oxidative emergency stress and increased glial cell activation levels are associated with cognitive dysfunction in ICH mice models (Antar et al., 2023; Bahadar and Shah, 2021; Bahader et al., 2021). Nevertheless, the precise mechanism by which DM influences post-MACEs in ICH patients remains unclear, and further extensive studies are warranted.

As a multicenter retrospective study, our research mitigated some biases inherent in single-center data. Additionally, we assessed the impact of DM on post-MACEs within the ICH population and further stratified the analysis by the type of post-MACEs, providing valuable clinical insights for practice. However, this study has certain limitations. First, as a retrospective rather than prospective study, it has some inherent biases, such as different clinical and procedural experience among different centers, different surgical details, and lack of long-term follow-up data. This makes it challenging to adequately control for case heterogeneity during study implementation. Second, our sample size was relatively small, which may reduce the reliability of the study results. Third, we did not systematically document the pre-ICH treatment methods for DM, which may have influenced the study results. To address these shortcomings, we plan to conduct a prospective study to comprehensively evaluate the impact of DM on post-MACEs in ICH patients.

In conclusion, this study demonstrates that DM serves as a risk factor for ischemic post-MACEs, hemorrhagic post-MACEs, and 30-days mortality in patients with ICH.