Split-hand phenomenon quantified by the motor unit number index for distinguishing cervical spondylotic amyotrophy from amyotrophic lateral sclerosis

Abstract

To investigate and compare split-hand phenomenon quantified by motor unit number index (MUNIX) between patients with cervical spondylotic amyotrophy (CSA) and those with amyotrophic lateral sclerosis (ALS). MUNIX was performed on abductor pollicis brevis (APB), abductor digiti minimi (ADM) and first dorsal interosseous (FDI) in 46 CSA patients, 39 ALS patients and 41 healthy subjects. Split-hand measurements including split-hand index (SHI=ABP×FDI/ADM), ratio of APB to ADM (AA), ratio of FDI to ADM (FA) were measured by compound muscle action potential (CMAP) and MUNIX. There was a significant difference in both AA and SHI measured by two different methods between ALS and CSA patients (P<0.05). Receiver operating characteristic (ROC) curve and logistic regression analysis demonstrated good differential diagnostic accuracy for AA, SHI and their combination between ALS and CSA. A larger area under the curve (AUC) was observed in these measurements calculated by MUNIX than those measured by CMAP (AA: 0.885 vs. 0.700, SHI: 0.865 vs. 0.703, Combination: 0.925 vs. 0.750; P<0.05). Sub-group analysis of ROC curves revealed an AUC of 0.893 for AAMUNIX, 0.801 for SHIMUNIX and 0.896 for their combination in differentiating "clinically possible" ALS (Awaji-Shima criteria) from CSA (P<0.05). Both AA and SHI measured by two different methods are useful in distinguishing ALS from CSA, and those quantified by MUNIX may be a better differential diagnostic marker to provide an accurate and noninvasive additional test for distinguishing CSA from ALS, even in their early stages.