Abstract
Quantitative neuromuscular monitoring (NMM) is widely recommended, yet its clinical adoption remains limited, partly due to usability issues. This study aimed to compare the clinical performance of two electromyography (EMG)-based NMM devices—the Datex-Ohmeda E-NMT module and the Stimpod NMS450X—under real-world operating room conditions. Twenty-four adult patients undergoing elective non-cardiac surgery were monitored simultaneously with both devices, applied contralaterally. Train-of-four (TOF) counts and TOF ratios were recorded at one-minute intervals. Agreement between devices was assessed using Bland–Altman analysis, repeatability coefficients, and the concordance correlation coefficient (CCC). The Datex-Ohmeda recorded an average of 121 TOF-count and 71 TOF-ratio values per patient, compared with 56 and 39, respectively, for the Stimpod. Inter-device correspondence was low, with numerous missing or physiologically implausible values on the Stimpod stream. The repeatability coefficient was 0.81 (95% CI: 0.80–0.83), indicating poor precision. The CCC was 0.662, consistent with weak agreement. In routine clinical use, the Stimpod NMS450X and Datex-Ohmeda E-NMT demonstrated poor concordance and high variability, likelyinfluenced by device-specific artifacts, connectivity limitations and electrode sensitivity.These findings highlight the need for device-specific validation of EMG NMM, both for clinical practice and for research purposes.
Supplementary Information
The online version contains supplementary material available at 10.1007/s10877-026-01411-5.
Keywords: Neuromuscular monitoring, Electromyography, Stimpod, Datex-Ohmeda, Quantitative monitoring
Introduction
Neuromuscular blockers have become a mainstay in all varieties of modern anaesthesia [1]. For anaesthesiologists worldwide it is considered standard of care for the vast majority of cases requiring endotracheal intubation. Deep neuromuscular blockade has been shown to improve operative conditions [2, 3] as well as to decrease postoperative pain [4]. These advantages must be balanced with the possible risk of postoperative residual paralysis [5]. The latter can cause a plethora of complications, ranging from general discomfort to severe postoperative pulmonary complications (POPCs) [6, 7].
Given that neuromuscular blockers display significant pharmacodynamic inter-variability [8], and the fact that subjective clinical assessments are incapable of consistently confirming the absence of postoperative residual neuromuscular blockade, it is broadly recognized that objective neuromuscular monitoring is essential for accurate perioperative neuromuscular monitoring. For practical purposes, electromyographic (EMG) NMM has been recognized as a clinical gold standard. This has been recently recognized in both ASA and ESAIC guidelines [9, 10].
Despite their long-standing availability, neuromuscular monitors (NMM) have a historically well-known anaesthetic practice penetration issue. Reasons for such a suboptimal adoption range from educational shortcomings to usability issues. As such, the goal of the present study is to assess the relative performance of two clinically available EMG neuromuscular monitors (NMM), namely the Datex-Ohmeda E-NMT and the STIMPOD NMX450X in moderate to shallow levels of neuromuscular block.
Materials & methods
The authors conducted an interventional prospective cohort study of adult patients who underwent elective non-cardiac surgery at the Universitair Ziekenhuis Brussel (Brussels, Belgium) between September 13, 2021, and August 30, 2024. Approval was obtained from the Institutional Review Board (IRB) at the Universitair Ziekenhuis Brussel (study ID UZBrussel 2021 − 197; approval date: August 18, 2021). Written informed consent was obtained from all subjects. Data was anonymised without trace-back possibility and handled as per GDPR guidelines.
Population
Inclusion criteria were: patients older than 18 years who underwent elective non-cardiac surgery requiring neuromuscular blockade longer than 30 min. Exclusion criteria were use of different types of neuromuscular blocking agents during the same procedure, known neuromuscular disease/syndromes judged to impair accurate neuromuscular monitoring, known acute or chronic hepatic or neuromuscular disease and inaccessible arms due to surgical positioning.
Randomisation and blinding
The application of the Stimpod neuromuscular monitor was randomised to the dominant or non-dominant hand by means of a computer generated randomisation. Given both devices were used concurrently, the Anaesthesia provider/investigator was not blinded to the intervention.
Anaesthesia
Anaesthesia was performed in a standard fashion and as per local institutional protocol. All patients were intubated and received a variable induction dose of rocuronium. Neuromuscular paralysis strategies were not subject to standardization between patients and were employed at the discretion of the attending anaesthesiologist. All cases used rocuronium as NMBD. Both inhalation Anaesthesia and total intravenous Anaesthesia were used.
Investigational protocol
Study devices
-
(A)
Stimpod NMS450X EMG
The Stimpod NMS450X (Xavant Technology PTY Ltd, South Africa) is a stand-alone monitor allowing for the assessment of the neuromuscular transmission both via accelerometry and electromyography. The latter is based on single-use EMG patches. It provides a pragmatic way to assess perioperative neuromuscular transmission.
-
(B)
Datex-Ohmeda E-NMT Module
The Datex-Ohmeda E-NMT EMG module (GE Healthcare, Finland) is an anaesthesia workstation-dependent module (Datex-Ohmeda) for the monitoring of the neuromuscular transmission. Unlike the Stimpod, the E-NMT module is not dependent on single use EMG strips, but instead uses common silver/silver-chloride electrocardiogram electrodes. It is similarly a convenient monitor that can be used in conjunction with compatible Datex-Ohmeda Anaesthesia workstations.
Neuromuscular measurement protocol
Neuromuscular measurements were started after induction of Anaesthesia and loss of consciousness, but before neuromuscular paralysis. Both devices were started synchronously, and an automatic supramaximal stimulus calibration was done before the first TOF Ratio and Count quantifications. These values were automatically recorded by each device. After the first measurement, TOF Count and TOFratio (if applicable) was automatically performed every minute and continued for 5 min after complete recovery of the neuromuscular blockade or until extubation, whichever came first. If surgery was finished before recovery of the TOFratio up to 90%, antagonization was performed using sugammadex.
Data collection
The data from the Datex-Ohmeda E-NMT Module was registered in the Metavision Anaesthesia workstation software (iMDSoft, Israel) and extracted by the research team under guidance of the IT-department of the hospital.
For the Stimpod NMS450X, the data was extracted from the manufacturer’s own data extractor (NMShow).
The following data were similarly collected and noted for every patient:
Date of surgery.
Timestamp of each neuromuscular measurement (hh: mm: ss).
Patient ID.
Hand to which the Stimpod device was randomised to (dominant or non dominant).
Supra-maximal current (mA) used by the different devices for transcutaneous nerve stimulation.
After completion of every case, the NMShow data and Metavision data were extracted to a Microsoft Excel spreadsheet together with the other patient specific information.
Endpoints
The study’s primary endpoint was the inter-device agreement in TOF ratio and the secondary endpoint the agreement in TOF count.
Sample size
The chosen sample size (n = 30) is comparable to similar research that addresses the same type of data, because in most of the present studies additional observations can be made at a very small cost (measurement interval of 1 min). An actual sample size calculation would require simulation of a mixed model with interdependent sample sizes for both number of time points and number of patients which also involves several parameters that are not yet sufficiently understood. The very high number of repeated measurements expected to be available supported the accurate estimation of the TOF ratio and count evolution at which estimation of variability is conditioned on, the fairly high number of patients captures their heterogeneity which is required for generalization [11].
Statistical analysis
A formal statistical analysis was preceded by a case-by-case analysis and interpretation of the individual patient data. Such an a priori analysis is necessary as due to the pragmatic character of the present study, neuromuscular endpoints will be invariably affected by movement/measurement artefacts and technical issues (disconnection, incomplete data transfer, etc.), among other possible interferences. Based on the aforementioned triage, cases were categorized according to the quality of data. A difference density plot was similarly used to understand the distribution and degree of differences between the two devices. After case categorization, the Stimpod NMS450X and Datex-Ohmeda E-NMT Module measurements were matched in pairs based on time-stamp synchronization. A Bland-Altman plot was subsequently used to plot the inter-method agreement. While an evaluation of the differences between paired scores addresses the agreement between methods, further linear mixed-effects modelling was used to improve those estimates by the estimation of the method specific averages, and the within and between methods’ and patients’ variability.
In analogy to inter-method comparison studies, the repeatability coefficient was also calculated to account for the effect of monitor imprecision. Normally, a low repeatability coefficient indicates a high precision. The repeatability coefficient was quantified as described by Liang et al. [12].
In order to facilitate interpretation for clinicians who may be less familiar with these statistical tools, it is important to briefly outline their meaning in this context. Bland–Altman analysis provides a graphical and numerical assessment of agreement by plotting the difference between paired measurements against their mean, thereby allowing the limits of agreement to be estimated and revealing any systematic bias or trend across the measurement range. The repeatability coefficient reflects the expected maximum difference between two measurements made by the same method under similar conditions; a low repeatability coefficient indicates high precision, whereas a high value, as in our study, points to substantial measurement variability. The concordance correlation coefficient combines information on both precision and accuracy, quantifying how closely the observed data fall on the line of identity between two methods; values close to 1 represent strong concordance, whereas values below approximately 0.9 are generally considered to indicate only moderate or poor agreement. Together, these complementary approaches provide a robust framework to evaluate whether two neuromuscular monitors can be considered interchangeable in clinical practice.
Results
Anticipating potential technical difficulties and a high artefact density, a total of 37 patients were sequentially included.
After a first individual case-by-case analysis, it became evident that both neuromuscular monitors had a significant intra- and inter-patient data variation. Data outliers were abundant in both monitors, and the spread in inter-method differences was significant (Fig. 1).
Fig. 1.
Density plot for inter-method differences in TOF-Ratio measurements
As such, a secondary case selection was performed where data was categorised based on data quality. Data points that were physiologically impossible and as such determined to be clinically inaccurate (e.g.: excessively high supra-physiologic TOF ratios; continuous oscillation between low and high TOF Counts; oscillations between low TOF counts and quantifiable TOF ratios) and sets with a lot of missing values were categorised as ‘poor’ data. Cases mildly plagued by erroneous data points were categorised as of ‘moderate’ data quality. Finally a third group was made for ‘satisfactory’ data quality. The aforementioned categorization was made separately for the TOF count and TOF ratio measurements. The sets that scored ‘poor’ on one of the two were dismissed as low quality data and left out the secondary analysis. After the aforementioned case exclusion, a total of 24 patients were considered for further analysis. Figure 2 exemplifies patients in both “Satisfactory” and “Poor” categories.
Fig. 2.
Exemplification of “Satisfactory” (A) and “Poor” (B) data quality patients. Left Panel: TOF-Ratio; Right panel: TOF-Count. Dat -Datex-Ohmeda; Stm - Stimpod
An average of 121 data points were measured per patient for the TOF-Count and 71 for TOF-Ratio in the Datex-Ohmeda monitor group. For the Stimpod, registered values were significantly lower, with a patient average of 56 TOF-Counts and 39 TOF-Ratios. Inter-device correspondence of TOF Count was assessed directly using the heat map (Table 1), which visualizes the full cross-distribution between devices.
Table 1.
Heat map of TOF-Count matrix, graded from green to red
It can be quickly deduced that inter-device correspondence is suboptimal, especially in the mid-range of counts. While the extremes of measurements are somewhat consistently aligned, the counts 1–3 do not imply correspondence. Furthermore, it is clearly appreciated that the Stimpod device frequently does not input a value (NA) while the Datex device does. This happened a total of 1490 times, compared to the reverse only happening 32 times. The lack of correspondence is highlighted in the corresponding heat maps (Table 1; Figs. 3 and 4).
Fig. 3.
Bland-Altman plot for the inter-device difference of TOF-Ratios
Fig. 4.
Concordance Correlation Coefficient per patient. Value 12 signifies total average CCC (0,662)
A Bland Altman analysis was used to assess the synchronized inter-monitor differences (Fig. 3).
From this the limits of agreement (LoA) were derived. The lower Limit of agreement was − 49.02, the upper limit of agreement was 52.46. The 95% confidence interval (CI) limits were [−51.58, 55.01].
This Bland-Altman plot abstracts away from having repeated measures, but if the focus is on how large differences between TOF-Ratios effectively are, this is not practically relevant. Considering the TOF ratio, it is evident that the differences are significant and spread across a large range of average measured TOF-Ratio values.
The LoA were then adjusted for nested structures. The resulting lower and upper LoA were − 47.13 and 50.57 respectively with a 95% CI [−51.80, 55.24]. The SD of differences was 24.92 (Table 2).
Table 2.
Limits of agreement in different subsets
It is also possible to split up the analysis for different subsets of data, which does not allow for any comparison between them, but only focuses on the estimation for the new subset of data. The subsets were defined as rates below 40, between 40 and 90, between 90 and 95 and above 95.
In order to further quantify inter-method differences, a Linear Mixed-Model (fit by REML algorithm) was used with a polynomial order of 8. Although overfitting, the aim was to capture the wildly erratic measurements observed between the devices, simply to focus on the resulting remaining variance due to different methods.
The variance of the method by item interaction is estimated as almost 7 (the same because estimating only 2 variables), the residual variance is about 19 for the Stimpod device and 79% of that (15) for Datex-Ohmeda. This resulted in confidence bounds for the absolute difference of TOF-Ratios of approximately 53,5%.
Note that the difference between the two methods is not significant, which may seem contradictory given the presented results, but is simply due to the fact that on average the difference is very small because big differences between methods exist in both ways, and because the residual variance (noise) is substantial.
The repeatability coefficient was significantly high (0.81) (95% CI: 0.80–0.83), suggesting very poor repeatability between the devices.
Finally a concordance correlation coefficient was calculated to be 0.662. Patient specific coefficients were also calculated to evaluate individual (patient-specific) concordance.
Discussion
While aiming primarily to compare clinically available EMG neuromuscular monitors, the present study was significantly affected by device-related usability issues. Most encountered hindrances pertained to poor data quality and consistency, which enhanced the statistical uncertainty of the presented results. Nevertheless, important clinical insight was gained into the devices themselves, which directly relates to uncontrolled daily practice settings.
Technical considerations
Although a priori one would expect to have symmetric datasets from both devices (i.e. every measurement from one device would be paired to a synchronous measurement by the second device), there was an evident discrepancy between the amount of data points recorded by both devices. In general, and despite having been set to equal measurement frequencies (1 min), the Stimpod device recorded significantly less data points than its counterpart. Firstly, it became apparent that the Stimpod monitor did not in fact measure every 60 s, but every 80 s. While in daily clinical practice this might not be of utmost relevance, in a controlled study setting this introduces unnecessary data skewing as within a same time interval this yields less data points. Moreover, this significantly increased the cumbersomeness of the process of calibrating/syncing both devices. Moreover, it became evident that the bespoke data management software for the Stimpod device (NMShow, v1.0.0) did not capture every datapoint measured by the device. In fact, it occurred frequently that the software would display the same value repeated over two minutes, despite effectively measuring more often. On the other hand, the Datex-Ohmeda E-NMT device recorded data without any issues (although captured by a third party software).
Furthermore, it occurred frequently that the Stimpod abruptly disconnected from the NMShow software during electromagnetic interference (EMI) episodes from monopolar electrocauterization use. Sometimes it disconnected without apparent reason. Reactivating or replugging the device or computer did not reset the issue, and a new recording session had to be initiated. Not only impractical, this led to obvious data collection losses. Additionally, the Stimpod did not allow for the extraction/recording of the first twitch height (T1%). Although not of direct relevance for the present study, this theoretically limits the device’s possibilities in terms of study of depolarizing agents, as well of pharmacokinetic/pharmacodynamic neuromuscular block onset studies [13].
Some other practical aspects of the Stimpod are worth mentioning as these also influenced the device’s usability. In fact, the hub for the disposable strip connection is quite bulky, incurring a significant risk of inducing pressure injuries when the arms are tucked against the patient’s body (which occurs frequently when clinicians opt to use EMG instead of the more movement sensitive acceleromyography). Secondly, the device as tested in the present study still used disposable batteries, which significantly hurt the financial rationale of using the device. The newly commercialized device iterations now incorporate an integrated rechargeable battery.
Finally, it stands out that the use of the Stimpod for research purposes is not ideal from a set-up point of view. Since the device lacks a substantial internal memory, data collection is conditional on the simultaneous connection of the device to a laptop that processes the data in live fashion. This leads to an overly complicated cable setup, where a data splitter (purchased apart) is required to allow for the simultaneous connection of the data and clinical measurement cables. Devices with internal memory (e.g. Senzime Tetragraph, or Blink Device company TwitchView) relieve the investigator’s burden by automatically storing data and allowing for a post-hoc data transfer.
Clinical considerations
While the Datex-Ohmeda E-NMT module has been commercialized for a long time and has been extensively studied in several independent studies, the same cannot be said for the Stimpod. It has been put forward as the market follow up for the TOF Watch SX once the latter ceased to be commercialized. However, the clinical performance of the Stimpod seems to be significantly different from its predecessor. In fact, in the present study we observed wide inter-device dispersion for TOF ratio with clinically relevant disagreement and instability, alongside frequent missing or non-physiologic readings, especially on the Stimpod data stream. The Bland–Altman limits of agreement for TOF ratio were broad (approximately − 49% to + 52%), and the repeatability coefficient was high (0.81), all indicating poor interchangeability of the two devices in routine practice.
Nevertheless, it should be also considered that the performance characteristics of the Datex-Ohmeda E-NMT are not uniformly aligned with those of research-grade EMG devices. Bussey et al. demonstrated that the GE/Datex E-NMT tends to overestimate TOF ratio and TOF count when compared with the TwitchView EMG, likely due to its susceptibility to noise and artifact misinterpretation [14].
Two recent validation studies suggested that the Stimpod EMG performance approaches that of mechanomyography (MMG), the reference standard. In a same-arm comparison, Wedemeyer et al. reported a small bias (0.028) and narrow limits of agreement (− 0.18 to 0.24) for TOF ratio, with moderate agreement for TOF count (weighted κ = 0.44) and substantial agreement for post-tetanic count (weighted κ = 0.80); the authors concluded that Stimpod EMG results were comparable to MMG [15].
Extending this line of work, a multi-device study without neuromuscular blockers compared three acceleromyographs and three EMG monitors against MMG. Acceleromyographs showed overshoot (mean TOF ratios ranging from 1.10 to 1.13) and higher variability, while the three EMG monitors clustered tightly around unity (means 0.99–1.01) with much lower variability, essentially mirroring MMG [16].
Conversely, and particularly relevant to the present study, Bussey et al. found that the GE E-NMT EMG systematically overestimated both TOF ratio and count compared with a validated EMG (TwitchView), frequently reporting TOF count of 4 when none were present, consistent with artifact misinterpretation. This known positive bias of the GE E-NMT may help explain part of the non-trivial differences we observed when it was paired with the Stimpod in dailypractice.
Moreover, recent clinical data also emphasize the importance of proper electrode placement and strict calibration procedures when using EMG monitors. Scheffenbichler et al. showed that failure to recalibrate after repositioning or changes in impedance leads to systematic drift in TOF ratio, while Ebert et al. highlighted the influence of electrode site preparation and the use of high-quality single-use electrodes [17, 18]. These factors may partially explain discrepancies between devices in real-world practice and may also contribute to the variability observed in the present study.
The discrepancy between the present study’s findings and those reported in controlled validation studies—such as the work by Wedemeyer and colleagues and the perspectives highlighted in Bowdle’s recent editorial—can be explained by several interacting factors [14, 15, 19]. Controlled evaluations typically occur under tightly standardized conditions, including meticulous electrode preparation, controlled limb positioning, absence of electrocautery, and stable impedance, whereas real-world operating rooms introduce continuous variability. Differences in electrode technology also play a role: the Stimpod uses single-use EMG strips that are prone to partial detachment, while the Datex-Ohmeda E-NMT relies on generic ECG electrodes, whose gel composition and hydration vary and may affect signal quality. Furthermore, emerging evidence emphasizes the importance of frequent recalibration of EMG monitors after patient repositioning or impedance changes—an aspect that is challenging to maintain consistently in routine practice. Beyond these physiologic and procedural variables, the above-mentioned device-specific technical limitations (Sect. 4.1—Technical considerations) contributed to the divergence observed in our cohort. Collectively, these factors likely explain why devices that perform reliably under bench-like conditions may demonstrate substantial variability and reduced concordance in pragmatic clinical use.
Implications for research
Good Clinical Research Practice guidelines recommend EMG as an acceptable alternative to mechanomyography (MMG) for pharmacodynamic research [13]. However, the present findings highlight that this recommendation should not be interpreted as blanket endorsement of all EMG devices. Instead, each EMG monitor must be individually validated for signal fidelity, noise rejection, electrode specificity, sampling stability, susceptibility to electrocautery and EMI, internal memory and data export reliability. Devices such as TwitchView and Tetragraph have undergone device-specific validation, whereas the Stimpod has limited clinical-context validation. Our study suggests that the Stimpod NMS450X, in its current implementation, may not be suitable for research applications requiring precise tracking of onset, offset, or pharmacodynamic modelling. Future research should include head-to-head comparisons of all major EMG devices in identical clinical conditions, as well as validation against MMG under controlled and real-world (electrocautery, movement, repositioning) situations.
With all facts taken together, while the published literature suggests that in controlled comparisons the Stimpod EMG can be accurate and precise relative to MMG under ideal conditions, our results highlight that in routine operating rooms, both the Stimpod and the Datex-Ohmeda E-NMT may struggle with artifacts, connectivity, and synchronization, thus limiting interchangeability and increasing clinical uncertainty.
Conclusion
The present study highlights a significant discrepancy in the clinical concordance of the electromyographic neuromuscular monitors Stimpod NMX450X and Datex-Ohmeda E-NMT. Both devices show a clinically significant intra-device variability/instability, as well as a clinically meaningful susceptibility to artifacts, precluding the establishment of clinically meaningful and quantifiable differences for statistical reporting.
Electronic Supplementary Material
Below is the link to the electronic supplementary material.
Author contributions
M.Z., U.J and H.C wrote the main manuscript text and made all figures and tables.W.C. performed all statistical analyses.M.Z., U.J, H.C. and J.D. all assisted in data collection.All authors reviewed the manuscript.
Data availability
No datasets were generated or analysed during the current study.
Declarations
Competing interests
The authors declare no competing interests.
Footnotes
Manish Zambre and Ulrich Janssens have shared first author.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Meistelman C, Donati F. Do we really need Sugammadex as an antagonist of muscle relaxants in anaesthesia? Curr Opin Anaesthesiol. 2016;29(4):462–7. 10.1097/aco.0000000000000359. [DOI] [PubMed] [Google Scholar]
- 2.Blobner M, Frick CG, Stäuble RB, Feussner H, Schaller SJ, Unterbuchner C, et al. Neuromuscular Blockade improves surgical conditions (NISCO). Surg Endosc. 2014;29(3):627–36. 10.1007/s00464-014-3711-7. [DOI] [PubMed] [Google Scholar]
- 3.Madsen MV, Staehr-Rye AK, Gätke MR, Claudius C. Neuromuscular Blockade for optimising surgical conditions during abdominal and gynaecological surgery: a systematic review. Acta Anaesthesiolog Scand. 2014;59(1):1–16. 10.1111/aas.12419. [DOI] [PubMed] [Google Scholar]
- 4.Moira HD, van Bruintjes, Helden AE, Dahan A, Scheffer GJ, van Laarhoven CJHM, et al. Deep neuromuscular block to optimize surgical space conditions during laparoscopic surgery: a systematic review and meta-analysis. Br J Anaesth. 2017;118(6):834–42. 10.1093/bja/aex116. [DOI] [PubMed] [Google Scholar]
- 5.Tsai CC, Chung HS, Chen PL, Yu CM, Chen MS, Hong CL. Postoperative residual curarization: clinical observation in the Post-Anaesthesia care unit. Chang Gung Med J. 2008;31(4):364–72. http://cgmj.cgu.edu.tw/3104/310405.pdf. [PubMed] [Google Scholar]
- 6.Kirmeier E, Eriksson L, Lewald H. Post-anaesthesia pulmonary complications after use of muscle relaxants (POPULAR): a multicentre, prospective observational study. Lancet Respir Med. 2019;7(2):129–40. 10.1016/s2213-2600(18)30294-7. [DOI] [PubMed] [Google Scholar]
- 7.Carvalho H, Verdonck M, Cools W, Geerts L, Forget P, Poelaert J. Forty years of neuromuscular monitoring and postoperative residual curarisation: a meta-analysis and evaluation of confidence in network meta-analysis. Br J Anaesth. 2020;125(4):466–82. 10.1016/j.bja.2020.05.063. [DOI] [PubMed] [Google Scholar]
- 8.Carvalho H, Michaël Verdonck, Eleveld DJ, David Moreno Ramírez, DʼHaese J, Panagiotis Flamée, et al. Neuromuscular end-point predictive capability of published Rocuronium pharmacokinetic/pharmacodynamic models: an observational trial. J Clin Anaesth. 2023;1:90:111225–5. 10.1016/j.jclinane.2023.111225. [DOI] [PubMed] [Google Scholar]
- 9.Fuchs-Buder T, Romero CS, Lewald H, Lamperti M, Afshari A, Hristovska AM, et al. Peri-operative management of neuromuscular blockade: A guideline from the European society of anaesthesiology and intensive care. Europ J Anaesthesiolog | EJA. 2023. 10.1097/eja.0000000000001769. 10.1097/EJA.0000000000001769. [DOI] [PubMed] [Google Scholar]
- 10.Thilen SR, Weigel WA, Todd MM, Dutton RP, Lien CA, Grant SA et al. (2023) American Society of Anesthesiologists Practice Guidelines for Monitoring and Antagonism of Neuromuscular Blockade: A Report by the American Society of Anesthesiologists Task Force on Neuromuscular Blockade. Anesthesiolog. 1;138(1):13–41. 10.1097/aln.0000000000004379 [DOI] [PubMed]
- 11.Carstensen B, Repeatability. Reproducibility and coefficient of variation. Comparing clinical measurement methods: A practical guide. 1st ed. Copenhagen: Wiley; 2010. pp. 107–14. 10.1002/9780470683019.ch9. [Google Scholar]
- 12.Liang SS, Stewart PA, Phillips S. An ipsilateral comparison of acceleromyography and electromyography during recovery from nondepolarizing neuromuscular block under general anaesthesia in humans. Acta Anaesthesiolog Scand. 2013;117(2):373–9. 10.1213/ane.0b013e3182937fc4. [DOI] [PubMed] [Google Scholar]
- 13.Fuchs-Buder T, Brull SJ, Fagerlund MJ, Renew JR, Cammu G, Murphy GS, et al. Good clinical research practice (GCRP) in pharmacodynamic studies of neuromuscular blocking agents III: the 2023 Geneva revision. Acta Anaesthesiolog Scand. 2023;67(8):994–1017. 10.1111/aas.14279. [DOI] [PubMed] [Google Scholar]
- 14.Bussey L, Jelacic S, Togashi K, Hulvershorn J, Bowdle A. Train-of-four monitoring with the TwitchView monitor electromyograph compared to the GE NMT electromyograph and manual palpation. J Clin Monit Comput. 2021;35:1477–83. 10.1007/s10877-020-00615-7. [DOI] [PubMed] [Google Scholar]
- 15.Wedemeyer Z, Jelacic S, Michaelsen K, Silliman W, Togashi K, Bowdle A. Comparative performance of stimpod electromyography with mechanomyography for quantitative neuromuscular Blockade monitoring. J Clin Monit Comput. 2023;38:205–12. 10.1007/s10877-023-01087-1. [DOI] [PubMed] [Google Scholar]
- 16.Wedemeyer Z, Michaelsen KE, Jelacic S, et al. Accuracy and precision of three Acceleromyographs, three Electromyographs, and a mechanomyograph measuring the Train-of-Four ratio in the absence of neuromuscular blocking drugs. Anesthesiolog. 2024;141(2):262–71. 10.1097/aln.0000000000005051. [DOI] [PubMed] [Google Scholar]
- 17.Scheffenbichler FT, Ulm B, Borgstedt L, et al. Precision of electromyography according to the calibration approach of neuromuscular monitoring: a randomised prospective agreement study. J Clin Monit Comput. 2025;39:1047–56. 10.1007/s10877-025-01304-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Ebert MT, Szpernal J, Vogt JA, et al. Improving quantitative neuromuscular monitoring: an education initiative on stimulating electrode placement. J Clin Monit Comput. 2025;39:169–74. 10.1007/s10877-024-01227-1. [DOI] [PubMed] [Google Scholar]
- 19.Bowdle A, Jelacic S, Michaelsen K. Let’s say goodbye to acceleromyographic twitch monitoring. Anesthesiolog. 2025;142(6):1002–4. 10.1097/aln.0000000000005439. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
No datasets were generated or analysed during the current study.