Incidence and outcomes of delayed intracranial hemorrhage: a population-based cohort study

Hyojeong Kwon,Youn-Jung Kim,Jae-Ho Lee,Seonok Kim,Ye-Jee Kim,Won Young Kim
DOI: https://doi.org/10.1038/s41598-024-70553-w
2024-08-22
Abstract:Head trauma is a common reason for emergency department (ED) visits. Delayed intracranial hemorrhage (ICH) in patients with minor head trauma is a major concern, but controversies exist regarding the incidence of delayed ICH and discharge planning at the ED. This study aimed to determine the incidence of delayed ICH in adults who developed ICH after a negative initial brain computed tomography (CT) at the ED and investigate the clinical outcomes for delayed ICH. This nationwide population cohort study used data from the National Health Insurance Service of Korea from 2013 to 2019. Adult patients who presented to an ED due to trauma and were discharged after a negative brain CT examination were selected. The main outcomes were the incidence of ICH within 14 days after a negative brain CT at initial ED visit and the clinical outcomes of patients with and without delayed ICH. The study patients were followed up to 1 year after the initial ED discharge. Cox proportional hazard regression analysis was used to estimate the hazard ratio for all-cause 1-year mortality of delayed ICH. During the 7-year study period, we identified 626,695 adult patients aged 20 years or older who underwent brain CT at the ED due to minor head trauma, and 2666 (0.4%) were diagnosed with delayed ICH within 14 days after the first visit. Approximately two-thirds of patients (64.3%) were diagnosed with delayed ICH within 3 days, and 84.5% were diagnosed within 7 days. Among the patients with delayed ICH, 71 (2.7%) underwent neurosurgical intervention. After adjustment for age, sex, Charlson Comorbidity Index, and insurance type, delayed ICH (adjusted hazard ratio, 2.15; 95% confidence interval, 1.86-2.48; p < 0.001) was significantly associated with 1-year mortality. The incidence of delayed ICH was 0.4% in the general population, with the majority diagnosed within 7 days. These findings suggest that patient discharge education for close observation for a week may be a feasible strategy for the general population.
What problem does this paper attempt to address?