Early indicators of relapses vs pseudorelapses in neuromyelitis optica spectrum disorder

OBJECTIVE: The purpose of this study was to review cases of neuromyelitis optica spectrum disorder (NMOSD) relapses and pseudorelapses to identify early features that differentiate between them at onset of symptoms. METHODS: This was a retrospective analysis of 74 hospitalizations of patients with NMOSD who were admitted to the Johns Hopkins Hospital for workup and treatment of a presumed relapse. Standard workup included MRI and blood and urine testing for metabolic and infectious etiologies. The gold standard for a relapse was defined as new or worsening symptoms and a change in neurologic examination correlating with a new or enhancing MRI lesion. A pseudorelapse was a clinical exacerbation with similar symptoms and signs but the MRI was negative, and workup identified an alternative cause for the symptoms that, when treated, resulted in the improvement of neurologic symptoms. Factors considered to be early predictors of relapses vs pseudorelapses were analyzed using the Fisher test. RESULTS: Among 74 NMOSD hospitalizations for presumed relapse, 57 were confirmed relapses while 17 had a negative MRI and an identifiable cause of pseudorelapse. The most common causes of pseudorelapse were infection, pain, and dysautonomia. The only early predictor that reliably differentiated relapse from pseudorelapse among this NMOSD patient population was vision loss (p = 0.039). Race, sex, presentations of weakness, numbness, and bowel/bladder dysfunction, white blood cell count, and urinary tract infection were not different among patients with relapses vs pseudorelapses. CONCLUSIONS: Vision loss in NMOSD is strongly suggestive of a true relapse vs a pseudorelapse. Pseudorelapses localized to the spinal cord in patients with previous myelitis presented similarly to true relapses and could only be ruled out by a negative MRI.