To clarify the relationship between nerve conduction study (NCS) and prognosis in patients with amyotrophic lateral sclerosis (ALS).We included 190 patients with sporadic ALS. We used onset age, sex, onset site (bulbar vs. spinal), revised El Escorial criteria category (definite vs. others), and the King's clinical systems, and the Milano-Torino (MiToS) functional staging systems, and decline rates of revised ALS functional rating scale (ALSFRS-R) as known prognostic factors. An NCS was performed on the median, ulnar, tibial, and sural nerves. The endpoint was death or the introduction of tracheostomy positive-pressure ventilation. Multivariate analysis for each NCS variable, known prognostic factors was performed using Cox stepwise proportional hazards analysis. Univariate analysis was performed for NCS variables that showed a significant association with prognosis in multivariate analysis. Survival was analyzed with a Kaplan-Meier curve and log-rank test.The Cox model identified the compound muscle action potential (CMAP) and sensory nerve action potential (SNAP) amplitudes of the median nerve as prognostic factors. In the log-rank test, patients with higher median nerve CMAP amplitude had a significantly better prognosis than those with lower amplitude, regardless of age. And prognosis was better in the group with lower median nerve SNAP amplitude only in patients younger than the 25th percentile (~ 57 years).CMAP and SNAP amplitudes of the median nerve are considered to be independent prognostic factors of sporadic ALS.