OBJECTIVE: To assess the real-world performance of a live (LCBA) versus a fixed (FCBA) cell-based assay for the detection of serum antibodies directed against myelin oligodendrocyte glycoprotein (MOG-IgG) and aquaporin-4 (AQP4-IgG). METHODS: This was a retrospective study of patients evaluated at a single tertiary academic referral center, with serum testing performed clinically for AQP4-IgG and/or MOG-IgG by FCBA and LCBA on the same day. Additionally, frozen banked sera from the same day for patients tested only by one assay were retrieved and tested by the other assay. FCBA was performed by the Johns Hopkins Immunology Laboratory using Euroimmun kits with detection by indirect immunofluorescence (FCBA-IF), whereas LCBA was performed by the Mayo Clinic Neuroimmunology Laboratory with detection by flow cytometry (LCBA-FACS). RESULTS: Of 594 specimens with paired MOG-IgG testing, 500 were negative by both assays, 33 were positive by both assays, 56 were positive exclusively by LCBA-FACS, and 5 were only positive by FCBA-IF. Overall, MOG-IgG LCBA-FACS exhibited 95.1% sensitivity and 97.7% specificity, whereas MOG-IgG FCBA-IF had 45.7% sensitivity and 99.8% specificity. Of 577 specimens with paired AQP4-IgG testing, 503 were negative by both assays, 51 were positive by both assays, 21 were positive exclusively by LCBA-FACS, and 2 were only positive by FCBA-IF. Overall, AQP4-IgG LCBA-FACS exhibited 97.3% sensitivity and 100% specificity, whereas AQP4-IgG FCBA-IF had 71.6% sensitivity and 100% specificity. INTERPRETATION: LCBA-FACS for both MOG-IgG and AQP4-IgG had markedly better sensitivity than FCBA-IF, with similar specificity. The use of FCBA-IF may result in underrecognition of both MOG antibody-associated disease (MOGAD) and AQP4-IgG seropositive neuromyelitis optica spectrum disorder (NMOSD).