Introduction: The long exercise test, or McManis test, is an electrophysiological study utilised in the diagnosis of periodic paralysis (PP), reported to be highly specific and sensitive.
Objectives: A service evaluation was carried out to demonstrate sensitivity and specificity of the McManis test in an unselected cohort of patients referred for consideration of a skeletal muscle channelopathy. Findings were correlated with clinical phenotype and genotype.
Methods: A retrospective analysis of 335 patients was performed, who had a McManis test at the National Hospital for Neurology and Neurosurgery (2009 to 2017).
Results: Analysis was focussed on 83/335 patients with strong clinical suspicion of PP and/or a positive McManis. Among 45/83 patients with genetically confirmed PP, 28 had a positive McManis, resulting in a sensitivity of 62.2%. In the group of genetically confirmed PP with negative McManis (17/45) 76% presented clinically with progressive weakness or rare episodes of weakness. This group also had a higher proportion of females (35% vs 18%). Some patients (3/17) with negative McManis despite genetically confirmed PP could be re-classified when considering the maximum CMAP amplitude obtained in the short exercise test performed in the same hand preceding the McManis. 10/24 patients with genetically unconfirmed PP but a positive McManis had a typical PP phenotype. 16/83 patients had a repeat McManis with only 4 showing a contradictory result upon test repetition. An overall intra-individual test-to-test mean variance of 13% was identified.
Conclusion: A lower sensitivity of the McManis test is seen when comparing with previous studies, which showed sensitivity ranging from 71% to 84% (McManis et al., 1986, Kuntzer et al., 2000, Fournier et al., 2004). The phenotypic heterogeneity and higher proportion of females might be contributing factors for a negative McManis in genetically confirmed PP. In addition, unresolved cases of PP with a positive McManis and exclusion of common PP mutations might require further genetic testing with new PP candidate genes, such as RYR1. Our data also suggests that the short exercise test should not precede the McManis test in the same hand in cases of clinical suspicion of PP. Repetition of the McManis test had little additional value. The intra-individual test-to-test mean variance of 13% may be of use for future clinical trials using the McManis as an outcome measure.