Three different short-interval intracortical inhibition methods in early diagnosis of amyotrophic lateral sclerosis.

To compare the utility of conventional amplitude measurements of short-interval intracortical inhibition (A-SICI) with two threshold-tracking (T-SICI) methods, as aids to early diagnosis of amyotrophic lateral sclerosis (ALS). The new parallel threshold-tracking method (T-SICIp) was compared with the previously used serial tracking method (T-SICIs). 112 consecutive patients referred with the suspicion of ALS and 40 healthy controls were prospectively included. Based on clinical follow-up, patients were divided into 67 patients with motor neuron disease (MND) comprising progressive muscular atrophy (PMA) as well as ALS, and 45 patient controls. SICI was recorded from first dorsal interosseus muscle using the three different protocols. MND patients had significantly reduced T-SICIp, T-SICIs and A-SICI, compared with healthy controls and patient controls, while healthy and patient controls were similar. Paradoxically, T-SICIp was least affected in MND patients with the most upper motor neuron (UMN) signs (Spearman  = 0.537,  < 0.0001) whereas there was no correlation for T-SICIs or A-SICI. T-SICIp also provided the best discrimination between patient controls and MND as determined by the receiver operating characteristic (ROC) curves. For patients with no UMN signs, area under ROC curve for 2-3ms inter-stimulus intervals was 0.931 for T-SICIp, 0.771 for T-SICIs and 0.786 for A-SICI. SICI is a sensitive measure for detection of cortical involvement in ALS patients. T-SICIp has higher sensitivity and specificity than T-SICIs and A-SICI, particularly in patients without any upper motor neuron signs.