Mild Traumatic Brain Injury and the Risk for Traumatic Intracranial Hemorrhage

Overview

Management of traumatic brain injuries causes significant efforts on emergency departments (ED) and overall health care. Patients on antithrombotic treatment with even minor trauma to the head, although without significant clinical findings, represent special challenges because the risk of traumatic intracranial hemorrhage (tICH) with these agents. The aim of this study was to compare the prevalence of tICH in patients on various pre-injury antithrombotic treatment exposed to minor Traumatic Brain injuries (mTBI) in Sundsvall with untreated patients. Secondary aim was to explore different risk factors for tICH. Data from medical records and radiology registry with mTBI in Sundsvall hospital between 2018-2020 in Sundsvall identified 2044 patients. Demographic data, pre-injury medications with antithrombotic treatment, state of consciousness at admission and the results of CT-scans of brain was investigated.

Inclussion criteria: 1. low energy traumas to the head, 2. pre-injury antithrombotic treatment, 3. \> 18 years, 4. reaction level scale (RLS) \< 3. Since data collection started at the hospital's radiology registry, we could easily retrieve all patients with low energy traumas to the head who underwent CT scans, were we also included patients without any pre-injury treatment for comparison. Exclusion criteria: 1. high energy trauma, 2. spontaneous hemorrhage, 3. concealed journals. This is a retrospective observational study investigating intracranial hemorrhage in patients with mTBI in Sundsvall hospital. The city is located on the east coast of northern Sweden and is the 20th largest municipality which serves a population of about 100 000 people. The studied period was between Jan 1st 2018 till May 30th 2020 and data collection was done by two persons following a predetermined protocol. Hospital's radiology registry and patient journals were used to find relevant participants. CT-scan referrals from surgical branch of the emergency department and acute surgical department, suspecting traumatic brain hemorrhage, were read. Mechanism of injury, pre-injury antithrombotic treatment and patient arrival status was subsequently retrieved from patient medical journals using social security number identified in the earlier step. mTBI is defined by Scandinavian guidelines as a GCS score of 14, or 15 if the patient also has either suspected/confirmed loss of consciousness, repeated vomiting, posttraumatic seizures, focal neurological deficits, signs of skull fracture, shunt-treated hydrocephalus, therapeutic anticoagulation or both \> 65 and antiplatelet medication \[3\]. GCS is not widely used in Sundsvall ED. Instead, Reaction level scale was chosen because of its more regular use. For comparison between the two scales, GCS 15 and RLS 1, and GCS 3 and RLS 8 are fully comparable. GCS 8 and RLS 4 are roughly comparable and are considered as an indication for intubation. If neither scale was utilized the patient was included if consciousness given in the patients´ journal was unaltered. High energy trauma was defined as any traffic accident or fall \> 3 meters, or other unstable patients who underwent immediate trauma care by a full trauma team. These patients often present themselves with multiple organ injury which was not within the scope of this study. Patients \< 18 years of age and patients with concealed journals were excluded because of legal and practical reasons. Demographic variables collected was age and gender. Severity of head trauma, treatment with anticoagulation therapy (DOAC/VKA/LMWH/other) and/or antiplatelet treatment (ASA/clopidogrel/ticagrelor/other) were noted. To make definitions clear, antithrombotic treatment is used as an overall term to include both anticoagulation therapy and antiplatelet treatments. tICH was categorized as subdural, subarachnoid, epidural, and other. Statistical analysis was executed using IBM® SPSS® Statistics version 27 for Microsoft Windows®. Demographic data were presented as means and standard deviations (SD), or as counts and percentage where appropriate. Comparisons between different groups' patient characteristics were performed with Pearson's χ2 test (for categorical variables), or independent-samples Student's t test or Mann-Whitney-Wilcoxon U-test (for continuous variables). P-values \< 0.05 were considered significant. Significant parameters in group comparisons were inserted in a multivariate logistic regression model. The sample size was estimated by the study supervisors to match comparable studies in the field, and to assure an adequate number of patients to examine the risk of dICH which we discuss in a separate paper. Since dICH is a much rarer condition, we found that patients included would be sufficient for this question at issue. Hence, power calculation prior to study start was done.