BACKGROUNDThe long-time exercise test (ET) is used to diagnose the primary periodic paralyses (PPs). However the reference values of ET are many and various. This study aimed to investigate the reference value of long-time ET in the diagnosis of PPs.

METHODSWe recruited 108 healthy subjects, 68 patients with PPs, and 72 patients with other diseases for the study. The procedure of ET was made on the basis of the McManis' method. Electrical responses were recorded from right abductor digiti minimi (ADM) muscle when stimulation of the ulnar nerve at the wrist. After the compound muscle action potential (CMAP) was monitored, subjects were then asked to contract the muscle as strongly as possible for 5 minutes. CMAPs were recorded for 2 seconds immediately after cessation of exercise, then every 5 minutes for 10 minutes, and finally every 10 minutes for 50 minutes. In general, the CMAP amplitudes will fall below the pre-exercise levels in an hour. The largest decrease was calculated and used as results of ET.

RESULTSThe CMAP amplitude decreases had no significant differences between groups when the healthy adults were grouped according to age, gender, height, weight and test time. Decreases in PPs patients (57.76%) were significantly more than in healthy subjects (15.21%) and other disease patients (18.10%, P < 0.001). Receiver operating characteristic (ROC) curve analysis showed that the best threshold is 35.50%.

CONCLUSIONSIn the long-time exercise test, threshold of 35.50% for the CMAP amplitude decrease was identified for abnormal. The result is not influenced by age, gender, height, weight, and test time. About 7.4% of healthy subjects were abnormal in ET.