Routine 24-hour head CT scan for oral anticoagulated patients with mild traumatic brain injury after a normal baseline brain CT scan: a systematic review and meta-analysis

Abstract

Abstract Background Traumatic delayed intracranial hemorrhage (dICH) is a potential but rare complication after minor traumatic brain injury (mTBI), especially in anticoagulated patients, but there is no strong evidence on the best management after a first negative brain CT. Aim of this systematic review is to estimate the need of a routine 24-hour head CT scan for all anticoagulated patients with mTBI after a normal baseline brain CT scan as recommended by current guidelines Methods PROSPERO protocol registration number is CRD42022316048. A systematic search on MEDLINE and EMBASE was performed. Weighted mean proportion and 95% confidence intervals (CIs) were calculated (Stuart-Ord), using a fixed-effect model (inverse variance). Risk of bias assessment of each study was defined based on study design, population sampling and lost to follow-up. Statistical heterogeneity was explored using the I² statistic. Results Twenty-five observational cohort studies, encompassing 8780 patients, were included. Oral anticoagulated patients whose baseline brain CT was negative are 8568. Of these, 92 patients developed dICH within 24 hours (1.3% [IC 95% 0.79%-1.97%]; I2 79.9% - random effects model), 6 patients died of hemorrhage (0.1% [IC 95% 0.05%-0.20%; I2 0% - fixed effects model]) and 5 underwent neurosurgery (0.1% [IC 95% 0.04%-0,19%; I2 0% - fixed effects model]). Symptomatic dICH were 29.8% [95% CI 15.78%-46.18%]; I2 2.3% - fixed effects model). In the subgroup of anticoagulated patients rescanned at 24 hours after a normal baseline brain CT scan (5674 patients), 1.8% (95% CI 1.14%-2.77%; I2 75.4% - random effects model) developed dICH. The following secondary outcomes were estimated: symptomatic dICH between 48 hours and 7 days (2 patients; 0.14% [95% CI 0.01%-0.45%]; I2 32.2% - random effects model) and between 8 days and 30 days (6 patients; 0.4% [95% CI 0.15%-0.78%]; I2 29.1% - fixed effects model), dICH within 24 hours in patients on both antiplatelet and anticoagulant drugs (2 patients; 1.3% [95% CI 0.23%-3.34%]; I2 0% - fixed effects model dICH within 24 hours according to anticoagulant therapy administered: vitamin K antagonist (VKA) (50 patients; 1.7% [95% CI 0.84%-3.09%]; I2 79.6% - random effects model) versus direct oral anticoagulants (DOACs) (30 patients, 1.6% [IC 95% 1.12%-2.25%]; I2 7,5% - fixed effects model). Conclusions A low incidence of major bleeding complications in oral anticoagulated patients with mTBI after a normal baseline brain CT scan suggests that routine 24-hour head CT scan should not be routinely recommended for all anticoagulated patients. A prospective management study using a single baseline CT scan and clinical follow-up is warranted to definitely assess which oral anticoagulated patients with mTBI need a routine 24-hour head CT scan.