A new analytical method to diagnose congenital myasthenia with stimulated single-fiber electromyography

Stimulated single-fiber electromyography (SSF-EMG) is useful to assess neuromuscular junction (NMJ) abnormalities in children. Conventionally mean consecutive difference (MCD) analysis measures the jitter for each muscle-fiber potential. We present a new algorithm that analyzes the entire SSF-EMG waveform. Cross-correlational coefficients (between 0-1.0) are calculated for consecutive pairs of 100 SSF-EMG waveforms obtained at each needle position in orbicularis oculi, and averaged. A lower normal limit (0.722, mean -3 SD) was established from 123 SSF-EMG samples in 10 adult control subjects, and applied to SSF-EMG data from 23 children referred for a suspected myasthenic syndrome. Results were compared with MCD analysis and related to the final clinical diagnosis. Our results showed that compared with conventional MCD measurement, the new algorithm had better specificity (87% vs. 53%) but similar sensitivity (88% for both). These findings indicate that the cross-correlational method is a useful predictor of NMJ dysfunction in children.