BACKGROUND: Many studies have indicated that acceleromyography and mechanomyography cannot be used interchangeably. To improve the agreement between the two methods, it has been suggested to use a preload and to refer all train-of-four (TOF) ratios to the control TOF (normalization) when using acceleromyography. The first purpose of this study was to test whether a preload applied to acceleromyography would increase the precision and the agreement with mechanomyography. The second purpose was to evaluate whether normalization would improve the agreement with mechanomyography. METHODS: Sixty patients were randomized to acceleromyography with or without a preload (Hand Adapter; Organon, Oss, the Netherlands). On the contralateral arm, mechanomyography was used. Anesthesia was induced with propofol and an opioid, and neuromuscular block with 0.6 mg/kg rocuronium. The precision and the bias and limits of agreement (with or without normalization) between the methods were evaluated using TOF stimulation. RESULTS: Preload improved the precision of acceleromyography by 21%, but it also increased the mean control TOF ratio from 1.07 to 1.13. Normalization of the acceleromyographic TOF ratios diminished the bias to mechanomyography during recovery (e.g., from 0.15 to 0.05 at TOF 0.90); when the mechanomyographic TOF values were normalized as well, the bias was eliminated. However, normalization did not exclude wide individual differences between acceleromyography and mechanomyography (+/- 0.10-0.20 at TOF 0.90). CONCLUSION: Preload increases the precision of acceleromyography, and normalization of the TOF ratios decreases bias in relation to mechanomyography. When both acceleromyography and mechanomyography are normalized, there is no significant bias between the two methods.