Accuracy of Conversion Disorder Diagnosis via Telestroke Network Consultation: A Retrospective Cohort Study Article

Full Text via DOI: 10.31486/toj.25.0043 Web of Science: 001583160000006

Cited authors

  • Poongkunran M, Ulep RD, Chokhawala H, Mohammed AE, Vidal G, Iwuchukwu IO, McGrade H, Chehebar D, Zweifler RM

Abstract

  • Background: Conversion disorder is a common mimic of ischemic stroke (IS) and transient ischemic attack (TIA). Data comparing the demographic and clinical features of patients with conversion disorder and IS/TIA are scarce, as are data evaluating the accuracy of conversion disorder diagnosis via telestroke consultation. Methods: We retrospectively analyzed consecutive patients evaluated through the Ochsner Health TeleStroke program from April 2015 through July 2016. Cases were classified into 2 categories: IS/TIA or conversion disorder. Patients with other stroke mimics and uncertain diagnoses were excluded. Both initial diagnosis and final diagnosis (following review of all records) were determined. The baseline characteristics of patients with conversion disorder were compared with the baseline characteristics of patients with IS/TIA. Initial diagnosis was compared with final diagnosis, and the diagnostic accuracy of telestroke consultations for conversion disorder was calculated. Results: We evaluated 885 patients with complete data. Of the 254 (28.7%) cases of stroke mimics, 50 cases (19.7%) were conversion disorder, representing 6% of all suspected strokes. We analyzed 616 patients with a final diagnosis of IS/TIA or conversion disorder. Compared to patients with IS/TIA, patients with conversion disorder were more likely to be female (P=0.0006) and younger (P<0.0001); were less likely to have diabetes mellitus (P=0.0252), hypertension (P=0.0004), or atrial fibrillation (P<0.0001); were less likely to receive tissue plasminogen activator (P<0.0001); and had a shorter median consultation duration (P=0.0175). The sensitivity, specificity, positive predictive value, and negative predictive value of conversion disorder diagnosis were 0.820, 0.998, 0.976, and 0.989, respectively. The adjusted area under the curve (95% CI) was 0.92 (0.87, 0.97). Conclusion: We found a conversion disorder rate of 19.7% of stroke mimics in the Ochsner Health TeleStroke Network, with a positive predictive value of 0.976 of discriminating conversion disorder from IS/TIA. In our study, patients with conversion disorder were more likely to be female and younger and to have fewer vascular risk factors compared with IS/TIA patients.

Authors

Publication date

  • 2025

International Standard Serial Number (ISSN)

  • 1524-5012

Number of pages

  • 6

Start page

  • 181

End page

  • 186

Volume

  • 25

Issue

  • 3