LETTERS TO THE EDITOR: Letters & Announcements
1. We do indeed understand that the technical details of the different principles on which different neuromuscular monitors are based could sometimes be rather confusing. Allow me to start by first clarifying what the M-NMT is not. The M-NMT is indeed not an acceleromyograph, since it does not measure acceleration by any means. Furthermore, Dr. Hemmerling used, in his title and in his letter, the same terms the manufacturer used to commercially describe the M-NMT as a “kinomyograph,” “KMG,” and “mechanosensory.” Such terms we have completely refrained from using in our article, as they could obviously imply that the M-NMT is some sort of a force displacement mechanomyograph. Anesthesiologists who are familiar with different neuromuscular monitors would immediately realize that it is not, since it applies no preload on the thumb nor does it quantify any force displacement. As detailed in our article on page 592, the M-NMT uses a piezo strip to generate an electric signal proportional to the degree to which the piezo strip is bent. So when we describe the motion as a “bending motion,” it is just that in its simplest form. To clarify matters further, if one would simply take the M-NMT sensor in one’s hand and quickly move the sensor (as a whole), no signal will be generated whatsoever. When the sensor strip is bent or folded, only then a signal will be generated. This clearly demonstrates that the sensor is neither an acceleromyograph nor a force displacement sensor. This is a completely different application and one of the numerous versatile uses of piezo sensors in medicine and other fields of science.
Dr. Hemmerling cites, in references 2 and 3, that when the M-NMT is compared to mechanomyography, it is “more sensitive” and “could be used interchangeably,” descriptions which are both in contrast to the title he chose for his letter. Reference 2 is a handpicked selected case on the home page of the manufacturer obviously intended for demonstration purposes. This hardly constitutes a scientific study with an appropriate sample size. Using terms such as “more sensitive” lacks the scientific bases of quantifying how “sensitivity” is being measured. As for reference 3, until the peer-reviewed publication of a full study, we feel that it is always hard to render a judgment based only on an abstract.
2. In reference 4, Dr Hemmerling mentions the ParaGraph® (Vital Signs, NJ), another commercially available neuromuscular monitor which similarly quantifies the neuromuscular block based on the same “bending motion” of a piezo strip. In that regard, the two monitors are similar. Alhough based on the same principle, the piezo sensor of the ParaGraph is attached to the hand by its adhesive pad, while the M-NMT uses, for the piezo strip attachment, the malleable plastic semicircular rings shown in the photo on page 592. This limits the “drift” in recovery from neuromuscular block, as we explained on page 595.
3. Dr. Hemmerling is clearly misquoting our study when he states that “the data could be used interchangeably.” As a matter of fact, we stated in the recovery phase, and again in the conclusion on page 595, the exact opposite. We stated that “the limits of agreement are unacceptably wide to allow the values given by the two monitors to be used interchangeably for individual patients,” as monitors based on different phenomena would not be expected to yield identical results.
The widely acceptable Bland and Altman statistical analysis (1) that we used in our study considers both monitors equally liable to experimental error, rather than assuming that one of the monitors is the correct true “standard,” thus rendered “reliable,” while the other is not. Our study showed that M-NMT could be used to indicate the onset time, times to repeat dose administration, and full recovery, and to that effect the M-NMT could be clinically reliable in a daily anesthesia setting. On the other hand, we objectively demonstrated the differences between the two monitors. We strongly feel that how one should regard the significance of these differences should be left entirely to the discretion of each anesthesiologist, who might choose, for instance, to take a parameter difference of 0.3 min into consideration rather than have a bulky mechanomyograph in the operating room.
Ashraf A. Dahaba
1. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986; 1: 307–10.