Background: 2.8 million Americans sustain a traumatic brain injury (TBI) annually. Mortality or neurosurgical intervention attributed to intracranial hemorrhage in mild TBI occurs in less than 0.3% and 3% of cases respectively. We developed an ED based multidisciplinary observation unit (OU) protocol for patients with mild TBI. We hypothesized that use of this protocol would enable safe discharge home in many of these patients and that the protocol provides sufficient information to predict those who are at risk of worsening. We conducted a retrospective cohort study to examine patients who were admitted to the protocol.
Methods: Our protocol included adults within 24 hours of injury. Patients were excluded if anticoagulated, new neurologic findings, agitated, required neurosurgical intervention, any other severe injury, vital sign abnormalities, other acute comorbid conditions, required inpatient syncope workup, seizures, thrombocytopenia, or pregnant. Relevant data from January 2016 until December 2017 including demographic data, imaging, medications, comorbidities, mechanism of injury, other injuries, length of time in observation, and subsequent disposition of patients admitted to the OU was dual abstracted. The association of predictors of observation failure were analyzed using univariate descriptive statistics.
Results: 173 patients were admitted to the OU over a two-year period. 131 (76%) had a positive head CT. The mean age was 49 (SD 23), 58 (34%) were female, and 135 (78%) were Caucasian. No patients discharged home after the observation protocol required a surgical intervention or ICU admission. 28 subjects (16%) were admitted to the hospital following their OU stay, most commonly for another medical condition (7/28, 25%). One patient who was admitted after an OU stay required a neurosurgical intervention and ICU admission. There were no deaths in either group. Patients admitted were older (mean age: 56 vs 48, p = 0.1) and had a higher proportion of positive head CTs (85% vs. 76%, p=.3); however, these differences were not statistically significant.
Conclusions: A period of observation for a pre-defined cohort of patients with mild TBI allowed successful discharge. Our observation protocol provided a safe, cost-effective triage plan for patients with mild TBI that could allow appropriate patient management with minimal resource utilization.