Clinically Significant Neuroimaging Findings Among Pediatric Patients Presenting to the Emergency Department With Symptoms of Psychosis: A Multicenter Retrospective Study.

Abstract

Background: The clinical utility of diagnostic neuroimaging for pediatric patients presenting to the emergency department (ED) for psychosis remains unclear. We sought to estimate the prevalence of and characteristics associated with clinically significant neuroimaging findings among pediatric patients presenting to the ED with symptoms of psychosis who had neuroimaging performed.

Methods: This retrospective cross-sectional study included visits by patients 5 to < 18 years old presenting with symptoms of psychosis to 28 EDs affiliated with the Pediatric Emergency Medicine Collaborative Research Committee from 2016 to 2019 and had neuroimaging performed. We estimated the rate of clinically significant neuroimaging findings, defined as those resulting in further testing, treatment, or medical admission, overall and by imaging modality. Multivariable logistic regression models examined presenting features associated with clinically significant findings.

Results: Clinically significant neuroimaging findings were identified in 5.4% (95% CI 4.2%, 6.9%) of 1118 ED visits (54% male, median [IQR] 14 [11-16] years old). Clinically significant findings occurred in 4.9% (34/699) of head computed tomography scans and 7.5% (45/604) of brain magnetic resonance imaging studies (p = 0.07). In a model that imputed missing data, no presenting features were associated with clinically significant neuroimaging findings. In a model that treated missing documentation as absence of the clinical feature, the adjusted odds of clinically significant neuroimaging findings were lower among ED visits by patients with suspected alcohol or substance use (aOR 0.38, 95% CI 0.16, 0.87).

Conclusion: Among pediatric patients presenting to the ED with symptoms of psychosis who had neuroimaging obtained, approximately 1 in 20 had clinically significant findings. Suspected alcohol or substance use was associated with lower odds of clinically significant neuroimaging findings, although this finding was not consistent across modeling approaches. Prospective studies are needed to definitively evaluate the utility of neuroimaging among children and adolescents presenting to the ED with symptoms of psychosis.