Comparative study of the conservative and surgical treatment of spontaneous cerebellar hemorrhage

Background and purpose The mortality rate of spontaneous cerebellar hemorrhage (SCH) is extremely high. Currently, only surgical treatment (ST) and conservative treatment (CT) methods are available; however, the indications for the treatment of SCH are not yet clear. In this study, we compared the outcomes of conservative and surgical treatment methods for patients with SCH and a hematoma volume of >10 mL. Methods We retrospectively included patients with SCH who were treated in the Emergency Department of the Affiliated Hospital of Guizhou Medical University, the Neurosurgery Department of the Affiliated Jinyang Hospital of Guizhou Medical University, and the Neurosurgery Department of the Second Affiliated Hospital of Guizhou Medical University from April 2014 to January 2024. Patients were divided into CT group and ST group using a 1:2 stratified matching method based on hematoma volume and diameter. We collected baseline clinical characteristics of patients, including age, blood pressure, imaging data, complications, and prognosis, and conducted univariate analysis. After excluding factors with collinearity effects through collinearity diagnosis, we used a binary logistic regression model to analyze the independent correlation between good and poor prognosis. Results Based on the inclusion criteria, 98 patients with SCH were screened, comprising 41 patients in the CT group and 57 patients in the ST group. Univariate analysis showed that the ST group had a higher proportion of patients with good prognosis at 1 and 3 months [41(71.9%) vs. 19(46.3%), p = 0.010], [34(59.6%) vs. 14(34.1%), p = 0.013], and a lower mortality rate than the CT group [10(17.9%) vs. 15(39.5%), p = 0.020]. SCH is further divided into a 1-month good prognosis group and a poor prognosis group, and a 3-month good prognosis group and a poor prognosis group. After excluding factors through collinearity diagnosis, the results of multivariate binary logistic regression analysis showed that surgical treatment had better 1- and 3-month prognosis than conservative treatment in SCH patients (OR: 4.898, 95% CI: 1.559–15.388, p = 0.007, OR: 3.965, 95% CI: 1.429–11.004, p = 0.008). Conclusion When the bleeding volume of SCH patients is greater than 10 mL, surgery is an independent predictor of good short-term prognosis.