Hi, everybody. My name's Alex Michael. I'm a sixth-year neurosurgery resident at Southern Illinois University Department of Neurosurgery, and I'll be presenting my work on assessing the efficacy and cost of mild traumatic brain injury management and the relevant disclosures. So as an introduction to this topic, traumatic brain injury accounts for more than two and a half million ER visits and hospitalizations in the U.S. annually, and most of these are actually considered mild in nature. Head CT imaging, though, has become ubiquitous nowadays, and the advances in imaging technology have contributed to improved detection of intracranial hemorrhage, or ICH, in this population without any real known clinical significance. For many hospitals, the standard practice is to transfer patients with radiographic evidence of ICH to tertiary care centers for neurosurgical assessment, observation, and repeat head CT to confirm hemorrhage stability. However, there's this growing body of literature that's evaluating clinical features that are predictive of future deterioration in subsequent neurosurgical intervention, and majority of this work is coming out of trauma surgery journals with very little contributions from neurosurgeons. Those clinical features that they've identified include neurological status, the state of their coagulability or coagulopathy, existence of skull fractures and those characteristics, as well as the ICH type and size. So in this study, we defined criteria for mild TBI patients who are at low risk for deterioration in neurosurgical intervention, and we then retrospectively reviewed the mild TBI patients that were presenting to a single level one trauma center to assess the clinical utility and cost effectiveness of repeating these CT head images and hospital transfers in this low risk cohort. So we identified all adult patients that presented in a four-year period, so 2014 to 2017, to a single level one trauma center, and we identified anyone with a international classification of diseases nine or 10 code that were consistent with the TBI. All included patients presented to the emergency department or they were transferred from an outlying hospital, and all patients were then admitted to the trauma service. All included patients received a neurosurgical consultation for a positive head CT finding of ICH and or skull fracture. And prior to the study, no imaging protocol had been instituted at our institution for traumatic brain injury. Briefly here's the ICD-910 codes that were used to identify the patients presenting to our healthcare system. We then determined which patients had the neurosurgical consultation and stratified into low and high risk for deterioration and neurosurgical intervention based on their initial head CT, the clinical examination, and medical history. The patients were categorized as low risk mild TBI if they met all of the following criteria, having a Glasgow coma scale greater than or equal to 13, they had to be without any focal neurological or pupillary deficits, they could not have any recent anticoagulant or antiplatelet use or other coagulopathies, they could not have any displacement of a skull fracture, and then the last couple stipulations for low risk were the intracranial hemorrhage. Subdurals had to be less than eight millimeters. Intraparenchial hemorrhage had to be less than eight millimeters. They must have less than three separate hemorrhages, so two or less. It cannot have scattered subarachnoid hemorrhage, it has to be focal. And then we excluded patients with an epidural hematoma or intraventricular hemorrhage. Only patients in this low risk mild TBI were included for statistical comparisons. All medical records were reviewed by two independent observers for patient demographics and clinical outcome variables including length of stay, ICU admission, number of head CTs, change in neurological status, neurosurgical intervention, transfer from outside hospital, and 30 day readmission related to the TBI. So neurosurgical intervention was also assessed including craniotomy, craniectomy, placement of an ICP monitor, barotraumatic or drain, and any kind of medical therapy such as hyperosmolar therapy that might result from a neurosurgical consultation. Direct health care costs related to the repeat head CT and the cost related to transfer from an outside hospital was also assessed. We found 531 patients based on ICD-9 and 10 codes during the four years of the study period. 119 patients were found to have low risk for deterioration and this was our inclusion group. All patients had at least one head CT scanned upon initial presentation and were diagnosed with various combinations of interstitial hemorrhage and or skull fracture. 31 patients of the included had a single head CT performed and 88 had a repeat head CT and this was approximately 74% of the study group that went on to get a repeat head CT. Out of the patients with a repeat head CT, 71 had only a single repeat head CT and 17 had three or more total head CTs so they had multiple repeat head CTs and this would total 106 total repeat scans performed for all patients. So 106 total repeat scans for 119 patients. Table 3 here lists our patient demographics. In general, most patients were admitted to the hospital and most had a short duration of stay with an average length of 2.9 days. 37 patients or 31% were transferred to our medical center specifically for neurosurgical evaluation. Only three patients developed evolution of initial intracranial hemorrhage on the repeat CT. No patient was symptomatic. No patient included in the study had neurosurgical intervention. Two patients had an in-hospital mortality. This was not related to TBI, this was related to a myocardial infarction. Seven patients had a 30-day readmission. Only two of them were TBI related and these were both seizures. Table 4 lists the intracranial radiographic diagnosis for each patient. The majority of patients had a single subdural hematoma or localized subarachnoid hemorrhage. We also looked into why patients were receiving multiple CT scans. Those reasons are listed in Table 5. Three patients with enlargement on their repeat head CT went on to get a third scan. So these were categorized as ICH expansion. Other reasons that were documented for a repeat scan were altered mental status, headache, and to rule out a delayed ICH expansion after starting anticoagulants or antiplelants. And antiplelants were started for cardiac reasons in all these patients. But most of the time, the real reason for the multiple repeat scan was not documented. We then calculated the total cost of routine repeat head CTs using an average hospital cost of $2,579 per CT and multiplied this by 106 total head CTs obtained. We evaluated 37 patients that were transferred from the outlying hospital to our institution through ambulance. TBI was the primary reason for transport of all patients. The cost for ground transport is $1,000 plus $60 per mile. So the resulting total health care expenditure for both the repeat CTs and the ground transfer combined for over $330,000 worth of health care cost. And this figure doesn't account for indirect costs such as fees for radiographic interpretation, increasing length of stay for hospital admissions with asymptomatic worsening hemorrhage, et cetera. Patients transferred to our facility from an outlying hospital were also found to be 5.7 times more likely to receive repeat head CT than those initially presenting to our hospital. And this meant statistical significance. So in conclusion, routine repeat head CT scans didn't change management of patients identified as having a low risk mild TBI and instead poses a significant financial burden on the health care system without adding much value to patient care. A proposed alternative of serial neurological examination appeared to be safe for the management of select mild TBI patients. The results of this study also brings up an extremely timely topic of the role of telemedicine in health care to potentially avoid unnecessary and costly patient transfer while still being adequately assessed by consultant neurosurgeons. And as always, a larger perspective analysis is warranted to further validate our findings.

neurosurgery resident

mild traumatic brain injury

intracranial hemorrhage

repeat head CT scans

telemedicine