Background

The accuracy and precision of currently available, widely used acceleromyograph and electromyograph neuromuscular blockade monitors have not been well studied. In addition, the normalization of the train-of-four ratio from acceleromyography (train-of-four ratio [T4/T1] divided by the baseline train-of-four ratio) has not been validated in comparison to mechanomyography.

Methods

Enrolled patients had surgery under general anesthesia with a supraglottic airway and without any neuromuscular blocking drugs. Three acceleromyograph monitors, three electromyograph monitors and a mechanomyograph built in our laboratory were tested. Most patients had an electromyograph and the mechanomyograph on one arm and a third monitor on the contralateral arm. Train-of-four ratios were collected every 12-20 seconds for the duration of the anesthetic. At least 1,000 train-of-four ratios were recorded for each device. Gauge repeatability and reproducibility analysis was performed.

Results

Twenty-eight patients were enrolled. In total, 9,498 train-of-four ratio measurements were collected. Since no neuromuscular blocking drugs were used, the expected train-of-four ratio was 1.0. All of the acceleromyograph monitors produced overshoot in the train-of-four ratio (estimated means 1.10-1.13) and substantial variability (gauge standard deviations 0.07-0.18). Normalization of the train-of-four ratio measured by acceleromyography improved the estimated mean for each device (0.97-1.0) but the variability was not improved (gauge standard deviations 0.06-0.17). The electromyograph and the mechanomyograph monitors produced minimal overshoot (estimated means 0.99-1.01) and substantially less variation (gauge standard deviations 0.01-0.02). For electromyography and mechanomyography, 0.3% of all train-of-four ratios were outside of the range of 0.9-1.1. For acceleromyography, 27% to 51% of normalized train-of-four ratios were outside of the range of 0.9-1.1.

Conclusion

Three currently available acceleromyograph monitors produced overshoot and substantial variability that could be clinically significant. Normalization corrected the overshoot in the average results but did not reduce the wide variability. Three electromyograph monitors measured the train-of-four ratio with minimal overshoot and variability, similar to a mechanomyograph.