Outcomes and Prognostic Factors After Hematoma Evacuation With or Without Decompressive Craniectomy for Adult Supratentorial Spontaneous Intracerebral Hemorrhage: A Single-Center Retrospective Cohort Study.

Abstract

The role of decompressive craniectomy (DC) during surgical treatment of supratentorial spontaneous intracerebral hemorrhage (ICH) remains controversial. Because DC is usually selected for patients with more severe mass effect, brain swelling, or concern for intracranial hypertension, its association with postoperative outcome must be interpreted in the context of baseline disease severity. This study evaluated clinical outcomes after hematoma evacuation with or without DC and identified factors associated with postoperative prognosis. We conducted a single-center retrospective cohort study including consecutive adult patients with supratentorial spontaneous ICH who underwent surgical treatment between January 2023 and December 2024. Patients were categorized into hematoma evacuation alone (HE group) and hematoma evacuation plus decompressive craniectomy (HE+DC group). The primary outcome was unfavorable functional outcome at 90 days (modified Rankin Scale [mRS] 4-6). Secondary outcomes included mortality, perioperative complications, and recovery indicators. Multivariable logistic regression was performed to adjust for prespecified confounders. Propensity score matching (PSM) was used as a sensitivity analysis to reduce treatment-selection bias. Prespecified subgroup analyses with formal interaction testing were conducted to assess whether the association between surgical strategy and outcome differed across clinically relevant subgroups. A total of 221 patients were included (HE group: n=132; HE+DC group: n=89). Patients in the HE+DC group presented with more severe baseline neurological status, including lower GCS scores, higher NIHSS scores, larger hematoma volumes, and higher rates of intraventricular hemorrhage (all P<0.001). In unadjusted analysis, the HE+DC group had a higher rate of 90-day unfavorable outcome compared with the HE group (36% vs 13%, P<0.001) and higher 30-day mortality (18% vs 6%, P=0.005). However, after multivariable adjustment, surgical strategy was not independently associated with 90-day unfavorable outcome (OR 1.727, 95% CI 0.759-3.946, P=0.191). Independent predictors of poor outcome included older age and larger hematoma volume, whereas higher admission GCS score was protective. Subgroup analysis demonstrated no significant interactions except for hemorrhage location (P=0.025). In the PSM cohort (n=128), no significant differences were observed in functional outcome or mortality between groups, consistent with the adjusted analysis. In this real-world cohort of surgically treated supratentorial ICH, HE+DC was not independently associated with 90-day unfavorable functional outcome after adjustment for baseline disease severity. The poorer crude outcomes observed in the HE+DC group should be interpreted in the context of greater initial disease severity and confounding by indication.