Medical Student Case Western Reserve University School of Medicine Cleveland, Ohio, United States
Introduction: There is limited evidence exploring outcomes of very elderly patients undergoing surgical evacuation of subdural hematomas (SDH). This study aims to identify predictors of adverse clinical outcomes following surgical evacuation of SDH in patients ≥ 80 years old.
Methods: A retrospective review of patients ≥ 80 years old undergoing surgical evacuation of SDH (acute, subacute, or chronic) between 2013-2021 at a single Level I trauma center was performed. In-patient mortality was the primary endpoint. Clinical parameters were evaluated using descriptive statistics and simple logistic regression analysis.
Results: A total of 157 patients were identified. Median age was 86 years and 70.1% of patients were male. 119 (75.8%) patients presented with traumatic brain injury. 37 (23.6%) patients had acute SDH, 24 (15.3%) patients had subacute SDH, and 96 (61.1%) patients had chronic SDH. Craniotomy, decompressive craniectomy, and burr hole evacuation were performed in 82.8%, 1.3%, and 15.9% of patients, respectively. 113 (72.0%) patients had post-operative neurological improvement, while 40 (25.5%) patients had at least one complication. At most recent follow-up, 84 (53.5%) patients had favorable outcomes. The median duration of follow-up was 490 days. Overall, in-hospital mortality was 7.6%. We found that pre-admission anticoagulant use (Odds-Ratio [OR]: 4.268; 95%-confidence-interval [CI]: 1.185-15.369; P=0.026) and in-hospital seizure activity (OR: 24.937; 95%-CI: 5.836-106.574; P< 0.001) were predictors of death during admission. Elevated Glasgow Coma Scale (GCS) scores on admission reduced in-hospital mortality (OR: 0.796; 95%-CI: 0.661-0.958; P=0.016).
Conclusion : Previous anticoagulant use and seizures during admission were predictive of in-hospital mortality, while higher GCS scores at admission were associated with reduced risk of death during admission following surgical evacuation. More prospective studies are required to better inform clinical risk stratification and confirm the results of our present study.
