Variability in head computed tomography use for minor head injury after ground-level falls in the emergency department: A subanalysis of EPI-TC study

Abstract

Objective The aim of this study was to assess the variation in the use of head computed tomography (CT) scan in patients attending EDs with ground-level fall-related minor head injury. Secondary objectives were: i) to measure the prescription rate of appropriate head CT scan, ii) to identify patients’ and EDs characteristics associated with head CT scan prescription iii) to explore potential correlation between head CT scan use and traumatic intracranial hemorrhage (ICH) yield rate in this population. Materials and methods This research was a planned sub-analysis of a cross-sectional prospective multicentric study performed in 63 EDs in France (EPI-TC study). Patients sustaining ground-level fall-related with minor head injury were included in this sub-analysis. The main outcome was head CT scan performed during ED stay. Variations in the use of head CT scan were described depending on each ED and French region. Multiple fixed effects mixed logistic regression model was performed to identify factors associated with head CT scan. Results A total of 631 patients admitted for head injury after ground-level fall were included. Median age was 79 [63–88] years. A head CT scan was performed in 409 patients (64.8%, CI95% 61.0–68.5); 19.6% (CI95% 15.8–23.7) of them were appropriated; and 29 (7.1%, CI95% 4.8–10.0) showed a traumatic ICH. At regional level, head CT scan prescription rate ranged from 45.5% (CI95%: 24.4–67.8) to 84.6% (CI95% 54.5–98.1). Head CT scan use was not correlated with the yield rate of traumatic ICH. In multivariable analysis, preinjury antiplatelets (OR 29.2, CI95%: 12.2–69.9), anticoagulants (OR 69.9, CI95%: 20.0–243.9), syncope (OR 6.9, CI95%: 2.0–24.2), post-trauma amnesia (3.2, CI95%: 1.0–10.5) and post-trauma loss of consciousness (OR 5.6, CI95%: 2.0–15.9) were associated with head CT scan use. Conclusions Head CT scan use in patient presenting to EDs with head injuries after ground-level falls is highly variable. High rate of head CT scan use is not correlated with high traumatic intracranial hemorrhage yield rate. The use of a clinical decision rule dedicated to this population would be suitable for harmonizing our practices.