Computed tomography for non-traumatic headache in the emergency department and the impact of follow-up testing on altering the initial diagnosis

Abstract

The purpose of this study was twofold: (1) to determine the incidence of positive computed tomography (CT) findings in patients presenting to the emergency department (ED) with non-traumatic headache at our institution and (2) to examine follow-up exams, including lumbar puncture, non-enhanced CT, CT angiogram, CT venogram, and magnetic resonance imaging (MRI), to see how often the use of further testing changes the diagnosis. With IRB approval, 865 patients were identified through ED requisitions for CT head with the indication of headache during the calendar year 2011. Exclusion criteria included head trauma, prior intracranial surgery, focal neurologic symptoms, and known intracranial mass. CT results were divided into three categories: P0, P1, and P2. Negative studies were graded as P0. Positive studies were subdivided into clinically insignificant or P1 and clinically significant or P2. Clinically significant was defined as requiring medical treatment. Subsequently, the electronic medical records and picture archiving and communication system (PACS) were reviewed to determine the incidence of follow-up exams, including lumbar puncture or imaging. The secondary tests were divided into the same P0, P1, and P2 categories. There were 254 positive studies: P1 clinically insignificant (27.1 %, 235/865) and P2 clinically significant (2.2 %, 19/865). Of 257 follow-up exams performed, the majority were lumbar punctures (36.0 %) or CT angiograms (29.5 %). In 19/257 exams or 7.4 %, the additional testing changed the clinically insignificant (P0/P1) diagnosis to a significant (P2) result. At our institution, there was a 2.2 % incidence of significant positive CT findings in patients presenting to the ED with non-traumatic headache. Follow-up testing was variable and resulted in a 7.4 % increase in the severity of diagnosis compared to the initial negative CT scan.

Nontraumatic Headache in Adult Emergency Patients: Prevalence, Etiologies, and Radiological Findings

“…The excellent survival in patients with headache was already shown in Swedish cohorts [35,36]. In detail, the prevalence of headache and the prevalence of sICA is considerably higher as compared to other studies [2,5,37], possible reasons being the higher median age and the lack of selection for certain features in our cohort [38][39][40]. Other studies did not examine emergency patients [37,41,42], which makes comparisons very difficult.…”

“…First, most were previously described, and second, all of these features may be overrepresented in a population "needed to be scanned". In fact, patients with neurological signs, though often subtle and detected by neurologists, and abnormal GCS were included, which was not the case in other studies [38][39][40]50,51]. While the influence of age on sICA seems obvious, the newly described association of high acuity ESI with sICA needs to be highlighted.…”

ACR Appropriateness Criteria® Headache

“…CT Head. The most appropriate initial imaging test in this clinical setting is a noncontrast head CT [31][32][33]. Negative predictive value of CT, when performed with modern scanners within the first 6 hours of symptoms and interpreted by a staff radiologist, might be as high as 99.9% [33,34].…”

Increasing and disparate use of neuroimaging for adults and children with non‐traumatic headaches in the US emergency departments: Opportunities for improvement

“…1,2 In the emergency department (ED) setting, neuroimaging is commonly utilized in patients with headaches to assist with patient triage, differential diagnosis, and to exclude life-threatening conditions, such as stroke, bleed, or neoplasm. [3][4][5] When used inappropriately or excessively, neuroimaging is of low yield 6 and can result in unnecessary radiation exposure, 7 use of sedation, 8 and downstream cascades of care. 9,10 The economic impact is substantial, with annual US health care expenses for neuroimaging in headache patients approaching $1 billion.…”

“…[17][18][19] In the last two decades, overall imaging utilization, regardless of indication, has been on the rise in the United States, 20 including both the adult and pediatric ED settings [21][22][23][24] with up to 30.0% of currently performed medical imaging of dubious clinical necessity. 23,25,26 The scope and focus of prior studies investigating the use of imaging in headache were limited to outpatient settings, 1,27 ED settings before 2007, 23,28 single-institution cohorts, 6 short (1 year) study periods, 6,29 and single imaging modality, 29 or were based on patient-generated data obtained through household surveys. 22 Longitudinal population-level patterns of imaging utilization for headache in US EDs in the last decade remain unknown.…”