The Long Exercise Test in Primary Periodic Paralysis: a Bayesian Analysis (P2.030)

Objective: To determine optimal methodology and normal values for the long exercise test (LET) in the evaluation of patients with suspected periodic paralysis (PP). Background: The LET involves the measurement of compound muscle action potentials (CMAPs) over time after sustained isometric exercise. The finding of a gradual decrement in CMAP after exercise supports the diagnosis of PP. However, the methodology by which the decrement is measured (intra- or post-exercise CMAP peak to nadir [McManis] vs pre-exercise CMAP baseline to nadir [Fournier], CMAP amplitude vs area) as well as the cut-off values used to define abnormal decrements have varied in the literature. Design/Methods: We reviewed LET data (abductor digiti minimi [ADM] CMAP amplitude and area) from 54 PP patients (31 genetically-definite), 39 healthy volunteers (HV), and 82 disease controls. ROC-curves were constructed and area-under-the-curve (AUC) calculated to compare 1) McManis vs Fournier methodologies, and 2) decrements in ADM CMAP amplitude vs area. Using Bayesian principles, optimal “cut-off” decrements were calculated for various disease pre-test probabilities (pre-TPs). Results: AUC was highest for the McManis method, and equal for CMAP amplitude and area decrements. For pre-TP ≤50%, optimal decrement cutoffs (McManis) were >40% (amplitude) and >50% (area). In the genetically-definite PP versus HV comparison, these cut-offs resulted in sensitivities of 58% (amplitude) and 67% (area) with specificities of 100% for both. The use of either amplitude or area criteria increased the sensitivity to 74% while maintaining 100% specificity. The LET could not be used to exclude PP in patients with periodic weakness even when pre-TP was low. Conclusions: For confirmation of periodic paralysis, our data support the use of the McManis method over the Fournier method to measure CMAP decrement. Bayesian analysis suggests optimal CMAP amplitude and area decrement cut-offs of >40% and >50%, respectively, when pre-test probability is ≤50%. Disclosure: Dr. Simmons has nothing to disclose. Dr. Lanning has nothing to disclose. Dr. Cleland has nothing to disclose. Dr. Puwanant has nothing to disclose. Dr. Twydell has nothing to disclose. Dr. Griggs has received personal compensation for consulting, serving on a scientific advisory board, speaking, or other activities with Serve on DSMB for PTC Therapeutics, Idera Pharma; consultant for Saretpta, Marathon, Strongbridge and Taro Pharma. Dr. Griggs has received personal compensation in an editorial capacity for Correspondence editor for Neurology. Dr. Griggs has received royalty, license fees, or contractual rights payments from Royalties from Marathon and PTC Pharmaceuticals. Dr. Griggs has received research support from Research support from Marathon, PTC. Dr. Tawil has nothing to disclose. Dr. Logigian has nothing to disclose.