The M-NMT Mechanosensor Cannot be Considered as a Reliable Clinical Neuromuscular Monitor in Daily Anesthesia Practice

To the Editor: We read with interest the article by Dahaba et al. (1), which evaluated the clinical use of the M-NMT (Datex-Ohmeda, Helsinki, Finland) neuromuscular monitor. However, we have several technical comments and believe that their conclusion that the M-NMT could be a reliable clinical monitor in daily anesthesia practice is not supported by their findings. 1. Information about the physical basis of the M-NMT sensor is incomplete. Dahaba et al. describe it as a “piezoelectric motion sensor measuring the signal generated from the bending and deformation of a piezoelectric sensor wafer strip.” What does that mean? Does it measure movement, and, if so, the extent of the movement, the amplitude of the evoked signal being proportional to the actual extent of the movement of the thumb, or does it simply measure acceleration like acceleromyography? Paloheimo (2) at the Datex-Ohmeda home page describes it as “kinemyography (KMG)” and compares it in one patient with mechanomyography. In this patient, KMG measurements are more sensitive than mechanomyographic measurements, which is in contrast to Dahaba’s findings, the M-NMT monitor being less sensitive than mechanomyography during recovery. Recently, Schwaerzler et al. (3) presented an abstract at the 7th International Neuromuscular meeting, which showed in 20 patients that TOF-ratios of more than 0.7 during recovery of neuromuscular blockade after rocuronium derived using the Datex-Ohmeda™ Mechanosensor and mechanomyography of the thumb can be used interchangeably. The term “mechanosensor” is confusing because it gives the impression that, like mechanomyography, the force is measured. The question remains: what does it actually measure? 2. In a previous study (4), Dahaba et al. presented another piezoelectric train-of-four neuromuscular monitor, the illustration of which is not very different from the current M-NMT mechanosensor. Do the authors know in which way these two devices are related? And do they have any explanation why in the former study, TOF ratio of 0.7 or greater derived using both methods could not be used interchangeably in contrast to the findings in the current study? 3. The most important problem we have with the M-NMT sensor is the fact that only TOF ratio of 0.7 or higher can be used interchangeably with mechanomyography. Although this is better than previously reported for acceleromyography (5), the rest of the pharmacodynamic comparisons with mechanomyography are disappointing. A mean difference of −0.3 min and limits of agreement of 0.4 and −0.2 min for onset time with a mean onset time of 1.5 min can matter clinically and is not superior to other methods such as acceleromyography or electromyography. Furthermore, it would be interesting to determine the agreement with neuromuscular blocking agents of longer onset time, such as cisatracurium. Even more disappointing is the fact that there were important differences between the M-NMT sensor and mechanomyography for the recovery times, such as DUR10or DUR 25, which are important for maintaining surgical neuromuscular blockade and determination of the time for “rescue” reversal. This finding contradicts the conclusion that the M-NMT sensor could be a reliable clinical monitor in daily anesthetic practice. The mere fact that it is integrated into the anesthesia monitor is not enough to qualify it as reliable or even useful. We agree that the M-NMT sensor can be used to determine a TOF ratio of more than 0.7 during recovery of neuromuscular block. However, the device cannot be used to determine accurately neuromuscular blockade for surgery (a neuromuscular blockade of 10–25 % of control twitch height) or early recovery of neuromuscular blockade and can therefore not be considered as a reliable clinical monitor in daily practice. Thomas M. Hemmerling, MD, DEAA François Donati, PhD, MD, FRCPC

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.