Arm-to-arm variation when evaluating neuromuscular block: an analysis of the precision and the bias and agreement between arms when using mechanomyography or acceleromyography.

BACKGROUND Studies comparing acceleromyography and mechanomyography indicate that the two methods cannot be used interchangeably. However, it is uncertain to what extent differences in precision between the methods and the naturally occurring arm-to-arm variation have influenced the results of these studies. Accordingly, the purpose of this study was to examine the precision and the arm-to-arm variation, when the same method is used on both of the arms. METHODS In the first part (n=20), mechanomyography was applied bilaterally and in the second part acceleromyography (n=20). Anaesthesia was induced with propofol and opioid, and neuromuscular block with rocuronium 0.6 mg kg(-1). The precision of the two methods and the bias and limits of agreement between the arms were evaluated using train-of-four (TOF) stimulation, without and with referral to the initial baseline value, that is, normalization. RESULTS Both methods were found to be precise (<5% variation) without any difference between the dominant and non-dominant arms. There were no significant biases between the arms, except for the onset time obtained with acceleromyography, which was 10% longer for the dominant arm. However, the individual differences (limits of agreement) were wide (0.20-0.25 at TOF 0.90). Normalization during recovery did not change bias or limits of agreement between the arms. CONCLUSIONS In the research setting, acceleromyography and mechanomyography are both precise methods without difference between the arms. Although there is no mean bias between the arms, both methods show wide individual differences (limits of agreement), which might to a large extend explain the differences often found when two different methods are compared on the contralateral arms. ClinicalTrial.gov identifier: NCT00472121; URL: http://clinicaltrials.gov/ct2/show/study/NCT00472121.