Evaluation of the proposed anti-N-methyl-d-aspartate receptor encephalitis clinical diagnostic criteria in psychiatric patients.

Abstract

The gold standard for diagnosing anti-NMDAR encephalitis is demonstration of the antibody in CSF. Clinical diagnostic criteria have been proposed for when this is not available in a timely manner which is evaluated, in this study, for a psychiatric population. This study retrospectively assessed the proposed criteria in patients presenting to psychiatric services for the first time with known anti-NMDAR antibody status. Antibody-positive cases were derived from the literature (conception to December 2019) and a state-wide (Queensland, Australia) cohort. Antibody-negative cases were derived from a service-wide (Metro South, Queensland, Australia) cohort of psychiatric cases which underwent antibody testing for routine organic screening. Sensitivity and specificity were calculated at 1week following admission and the point of discharge. The proposed criteria were applied to 641 cases (500 antibody-positive and 141 antibody-negative), demonstrating a sensitivity which increased from around 19% after 1week to 49% by the point of discharge. Specificity was 100% at both time points. The mean average time to become positive using the proposed criteria was 19.5days compared to 34.9days for return of antibody testing. High specificity of the proposed criteria, seen in this study, suggests that cases which are positive can be considered for expedited commencement of treatment. However, if clinical suspicion is high despite criteria being negative, it is essential to test CSF for anti-NMDAR antibody.