Advances in Neuromuscular Monitoring Techniques in Anesthesiology: A 2025 Perspective

Introduction

Residual neuromuscular blockade after the use of neuromuscular blocking agents during general anesthesia occurs in 20–40% of patients [1]. Although the fundamentals of neuromuscular monitoring were first applied in practice in 1958, the first guidelines for routine monitoring were published in 2009 at the initiative of the Working Party on Post-Anesthesia Care [2]. Nine years later, another set of guidelines stated that any patient given a non-depolarizing relaxant should be subjected to objective neuromuscular monitoring, as neither subjective methods nor clinical tests are sufficiently sensitive to detect residual neuromuscular blockade and should not be used [3]. Following this, local guidelines were published in various countries [4–10]. Nevertheless, at the end of 2024, the issue of neuromuscular blockade monitoring remained relevant [11]. The latest review on this topic, published in January 2024 [2], has become outdated due to the publication of several significant studies on this subject. Train-of-four (TOF)-Watch, the most popular method for monitoring [12–14], is no longer available [15]. Moreover, in recent months, several important studies have been published, presenting comparisons of various devices placed in different alternative locations. Therefore, this article aims to provide a review and update on current methods for clinical neuromuscular monitoring in anesthesiology.

Barriers to Routine Use of Neuromuscular Monitoring

The TOF technique is the standard method for evaluating the depth of neuromuscular blockade and the recovery of neuromuscular transmission during general anesthesia [16]. This method involves delivering 4 electrical impulses at 0.5-second intervals every 10 seconds via electrodes placed over an accessible peripheral motor nerve [16]. The resulting muscular responses – commonly observed as twitches – are then compared, and the TOF ratio (TOFR) is calculated by dividing the amplitude of the fourth twitch (T4) by that of the first (T1). A TOFR exceeding 0.9 is widely accepted as the threshold above which tracheal extubation can be considered safe, indicating adequate recovery of muscle strength and protection of the upper airway [17]. TOF monitoring reduces the risk of postoperative pulmonary complications and 90-day mortality [18]. In patients with abnormal butyrylcholinestrase activity, it can prevent premature awakening from anesthesia [19]. Despite that, routine use of objective neuromuscular monitoring remains limited.

Polish legal regulations, including the 2016 ordinance from the Minister of Health, recommend the availability of neuromuscular transmission monitoring equipment at every anesthesiology workstation [20]. Nonetheless, many hospitals still fail to adhere to these recommendations, exposing patients to avoidable complications that can be fatal [21]. Similar problems are observed in the other European countries [22–24], in Asia [25], and in the USA [26]. This omission is often attributed to a combination of subjective beliefs that monitoring is unnecessary or time-consuming, and a lack of functional equipment, even when devices are theoretically available [26,27]. Contributing factors may also include widespread burnout among anesthesiologists, which has intensified after COVID-19 and now affects as many as 73% of practitioners [28].

The underutilization of objective neuromuscular monitoring is further compounded by several technical and cultural barriers. Accurate results require proper calibration of the monitoring device prior to the administration of neuromuscular blocking agents. This step, which establishes a baseline and identifies a supramaximal stimulus (ie, a current that activates all muscle fibers in the target nerve by 15–20% above threshold), is essential for eliminating background noise and ensuring reliable data. Calibration can be omitted in newer-generation devices such as TOF-Scan [29], Stimpod NMS 450X [30], or the wireless TOF monitor, which are pre-configured for clinical use and less susceptible to variability [31]. Artifacts caused by patient movement, electrical interference, or improper electrode placement can lead to inaccurate readings, undermining clinician confidence in the results [32]. Moreover, a lack of familiarity with the fundamental principles of neuromuscular physiology and monitoring technology fuels scepticism. Many clinicians are insufficiently trained in interpreting TOF ratios or understanding the implications of residual blockade; however, additional training on neuromuscular monitoring does not change the clinical practice even though post-course assessments indicated improved knowledge among anesthesiologists [33]. In Europe, these limitations are reflected in the finding that approximately 19% of anesthesiologists do not use objective monitoring devices in their practice [34].

Due to these issues, many clinicians continue to rely on clinical signs, such as the patient’s ability to lift their head or leg for 5 seconds, sustain hand grip, protrude the tongue, or meet minimal ventilatory thresholds like tidal volume or negative inspiratory force. These signs are neither sensitive nor specific and have been repeatedly shown to fail in reliably identifying residual neuromuscular blockade: the 5-second hand grip test had a sensitivity of 24%, specificity of 72%, positive predictive value (PPV) of 52%, and negative predictive value (NPV) of 81%; these values for the 5-second head lift test were 41%, 80%, 48%, and 77%, respectively [25]. Their unreliability is particularly evident in vulnerable patient populations, including the elderly, women (especially during pregnancy) [35–37], and individuals with renal or hepatic dysfunction, malignancy, hypothermia, neuromuscular diseases, or metabolic imbalances [36–42] (Table 1). In such cases, subjective assessments provide false reassurance and can delay the recognition of incomplete recovery. A recent study showed that half of the patients fulfilling the clinical criteria for extubation demonstrated residual neuromuscular blockade [43].

Accurate monitoring is essential not only for detecting residual blockade [44], but also for reducing the risk of potential interactions between relaxants and other numerous drugs used during general anesthesia [45–46]. To overcome these challenges, several objective methods for neuromuscular monitoring have been developed, offering more precise and reproducible assessments. Unlike visual or tactile evaluation, which are highly dependent on clinician experience and particularly prone to misinterpretation at TOF ratio values below 0.6 [21], instrumental methods allow for the quantification of muscle response following standardized electrical stimulation. Technologies such as acceleromyography (AMG), mechanomyography (MMG), electromyography (EMG), kinemyography (KMG), phonomyography (PMG), sonomechanomyography (SMMG), and compressomyography each enable real-time, objective tracking of neuromuscular function by capturing distinct physiological parameters of muscle contraction.

Importantly, these objective methods remain unaffected by the simultaneous administration of intravenous anesthetics, opioids, or volatile agents. Their use supports more precise titration and reversal of neuromuscular blockade, ultimately enabling faster and safer patient recovery. By ensuring the complete return of muscle function prior to extubation, anaesthesiologists can significantly reduce the incidence of postoperative complications such as airway obstruction, hypoxemia, tachypnea, and respiratory insufficiency [47]. The present article provides a concise overview of the currently available methods for objective measurement of neuromuscular blockade.

Mechanomyography

MMG, a historical reference standard [48], quantifies the mechanical response to an applied electrical impulse. Usually, it is based on stimulating the ulnar nerve while a sensor is positioned on the thumb, which is maintained in an abducted position under a preload of 2–3 N, which ensures that the measured force aligns with the axis of the transducer. A crucial aspect of this method is the precise preload application and correct positioning of the thumb. Additionally, calibration of the device is essential, as it establishes a baseline contraction value before and after the administration of hypnotics (since the calibration process itself is painful). Only after this reference point has been determined can muscle relaxants be administered, allowing for accurate measurements. The transducer converts contraction force into an electrical signal, which is then displayed either numerically or in analog form on a monitor.

Despite its precision, MMG has several limitations that make it impractical for routine clinical use. The technique is highly dependent on various factors and requires a time-consuming process, making it unsuitable for everyday practice. Instead, its primary application is in research on muscle relaxants, where it is still considered the benchmark method. However, MMG can only monitor specific muscles, such as the adductor pollicis and flexor digitorum brevis, while it cannot assess critical respiratory muscles like those of the larynx and diaphragm, which are particularly relevant in anesthesia. Furthermore, the discontinuation of the Myograph 2000 device by its manufacturer has contributed to the declining importance of MMG.

Acceleromyography

AMG is the most widely used technique for neuromuscular monitoring, relying on the piezoelectric effect, in which mechanical stress generates electrical charges on a material’s surface [49,50]. This method measures muscle acceleration following nerve stimulation. Since acceleration is the key parameter, AMG is most effective in anatomical regions where nerve-induced movement is easily detectable. The ulnar nerve is typically chosen for this purpose, with electrodes positioned along its pathway on the ventral side of the wrist and an acceleration sensor attached to the thumb. When the nerve is stimulated, the resulting finger movement generates an electric current, with the signal strength proportional to the acceleration. Applying Newton’s Second Law (Force = Mass × Acceleration), the measured acceleration is used to estimate the force exerted by the muscle contraction [51].

Beyond the ulnar nerve, AMG can also assess neuromuscular function by stimulating the posterior tibial nerve or the facial nerve, targeting muscles such as the frontalis, the flexor digitorum brevis, orbicularis oculi, corrugator supercilia, levator labii superiori aleque nasi, zygomaticus major, orbiculus oris, depressor anguli oris, or mentalis, abductor digiti minimi, flexor hallucis brevis, and first dorsal interosseus [2]. While blockade of the orbicularis oculi muscle closely resembles that of the adductor pollicis, the corrugator supercilii muscle has greater resistance to neuromuscular blocking agents (NMBAs). Consequently, its response is more comparable to that of the laryngeal adductor muscles or the diaphragm [17]. AMG is also widely used in children [52].

In clinical settings, AMG is straightforward to use but requires careful positioning to ensure the limb returns to its original state between measurements. To minimize gravitational interference, the hand is typically placed in supination with horizontal movement. Wearable thumb sensors have helped reduce errors caused by patient repositioning or incomplete thumb return. The measured values are converted into a TOFR and displayed on the device monitor.

A unique feature of AMG-based monitors, such as TOF-Watch and TOF-Watch SX, is the so-called “reverse fade phenomenon”, in which TOFR values exceeding 1.0 may appear during the period between calibration and muscle relaxant administration. Unlike mechanomyography (MMG) and electromyography (EMG), this is an inherent characteristic of AMG technology, not a measurement error [53].

AMG often registers slightly higher neuromuscular recovery values compared to MMG and EMG, making direct comparisons between these techniques unreliable [54]. A 2013 study by Liang found that AMG tends to overestimate TOFR values by at least 0.15 relative to EMG [55]. Some researchers have proposed “normalizing” AMG-derived recovery values by first comparing them to baseline readings, ensuring that neuromuscular recovery is not deemed sufficient unless the adjusted TOFR reaches at least 0.9 [54,55]. While this level of precision is acceptable in clinical practice, it falls short of the accuracy required for research applications. The bigger clinical problem remains the high variability of this method, which is not improved by normalization [54].

Compressomyography

The TOF-Cuff monitor, a novel neuromuscular monitoring device, first emerged in Europe in 2014, entered the Polish market in 2017, and gained the Food and Drug Administration’s approval for use in the United States in 2019. This system modifies a standard non-invasive blood pressure cuff by incorporating electrodes designed to stimulate peripheral nerves – primarily the ulnar or median nerves in the upper limb, although it can also be applied to the lower extremities [2]. Unlike traditional mechanomyography, which directly measures muscle contraction, this device detects changes in cuff pressure as a proxy for neuromuscular response [2]. A key advantage of the TOF-Cuff monitor is its adaptability to various patient positions, facilitated by an autopilot mode that detects positional changes and autonomously initiates vital function measurements [2]. This feature allows its use during surgical procedures where arm mobility is restricted, such as those requiring the arms to be secured. Additionally, its dual functionality as both a neuromuscular transmission monitor and a blood pressure measurement device streamlines preparation in urgent scenarios, including emergency caesarean sections under general anesthesia [2]. While currently available only for adult patients, a pediatric version is under development. However, widespread adoption of the device has been limited due to its high cost.

Several studies have evaluated the TOF-Cuff monitor’s performance in comparison with other neuromuscular monitoring techniques. A 2020 study [56] examining its use during anesthesia induction in obese patients, relative to the TOF-Scan device, identified significant systemic differences in time to a TOFR of 0. These variations were pronounced both between and within individual subjects, leading researchers to conclude that the 2 systems are not interchangeable. Similarly, a 2018 investigation [57] comparing the TOF-Cuff with AMG found that TOF-Cuff measurements on the upper limb should not be used as a substitute for EMG or AMG readings from the adductor pollicis muscle. On average, recovery to TOFR >0.9 took approximately 25 minutes longer when assessed with EMG or AMG than with the TOF-Cuff, reinforcing the conclusion that EMG and AMG are more reliable for ruling out residual neuromuscular blockade.

A 2020 study [58] assessed TOF-Cuff performance on the lower limb during upper-limb surgeries, reporting strong agreement with TOF-Scan results in the period between atracurium administration and a TOFR of 0. However, a notable failure rate was observed for lower-limb measurements with the TOF-Cuff. Conversely, another 2020 comparison of TOF-Cuff and TOF-Scan indicated that the TOF-Cuff tended to register neuromuscular recovery endpoints earlier than TOF-Scan [59]. A third article was published in 2024 for the same comparison, which showed no significant differences in time to onset; the time from the last dose of mivacurium to TOF ratio >90 was shorter on the leg than on the hand [60].

A 2017 study comparing TOF-Cuff with MMG and invasive blood pressure monitoring confirmed its effectiveness in tracking neuromuscular blockade and non-invasive blood pressure (NIBP), but it could not be considered a direct substitute for MMG [61]. Additionally, research conducted by Markle et al in 2020 [62] highlighted technical challenges associated with the device. Among the 56 patients enrolled, technical malfunctions occurred at least once in 11 individuals using the TOF-Cuff, compared to 9 cases with the TOF-Scan. Both observations were confirmed in 2024 by Radkowski et al [60]. While systolic blood pressure was comparable between TOF-Cuff placed on the lower leg and the standard method for blood pressure measurement on the arm (Spearman correlation was significant in 76.9% of patients and insignificant in 19.2%), diastolic blood pressure was significant only in 50%; in 24% of cases, TOFR=0 was not reached. The same authors also compared the TOF-Cuff placed on the arm with TOF-Scan placed on the eyelid [63] and TOF-Scan placed on the palm [64]. While recognizing the need to have an alternative method and location of measurement, as well as the differences between TOF-Cuff and TOF-Scan, the authors proposed a correction factor to convert the results of one method to the other. However, this calculation needs to be confirmed in a larger sample and validated in other patient populations.

In 2024, Guo et al presented a prototype of another CMG device, which utilizes a pressurized catheter balloon to detect thumb twitch responses during the TOF test [1]. After constructing an analytical model, performing numerical simulations, and mechanical finger testing, they conducted a pilot clinical study involving human participants. The results indicated no significant correlation between the device’s performance and individual subject characteristics, such as hand size, giving hope for a new alternative in the future.

Kinemyography

KMG is an alternative method for monitoring neuromuscular transmission, utilizing the piezoelectric effect, much like AMG. However, instead of measuring acceleration, KMG relies on the deformation of a mechanosensor embedded with piezoelectric material. This sensor is positioned between the thumb and index finger, capturing mechanical changes during muscle contraction. Upon stimulating the ulnar nerve, the resulting contraction of the adductor pollicis muscle distorts the sensor, generating an electrical signal that is recorded by the neuromuscular transmission module. The best-known device employing this method is the GE Datex-Ohmeda M-NMT Module. Despite the shared underlying principle, KMG and AMG yield differing results. A 2016 study by Salminen et al [65] comparing KMG (M-NMT Mechanosensor) with EMG found that KMG consistently overestimated neuromuscular transmission values relative to EMG. In clinical trials, kinemyography is used rarely; there are only 2 papers published in 2024 whose authors used this method. One of them [66] showed that kinemyography overestimates TOFR, whereas acceleromyography underestimates that; the other compared sugammadex with neostigmine during monitoring with one method, indicating significant differences in the time to reach a TOF ratio 90% and extubation time but not in the time to first flatus, defecation, or postoperative nausea and vomiting [67]. However, in some cases (eg, in Charcot-Marie-Tooth syndrome patients) this method may be more reliable than EMG [68].

Phonomyography

PMG is a more recent approach to neuromuscular monitoring, with initial studies conducted in 2004 and further investigations using animal models appearing in 2006. Unlike other techniques, PMG relies on detecting the low-frequency sounds produced by muscle contractions. A specialized microphone placed on the skin’s surface captures these sounds, which are directly proportional to the force of contraction. Once the acoustic signal is processed, the data are displayed in a graphical format on a monitor. Although PMG correlates well with other neuromuscular monitoring methods such as MMG, EMG, and AMG [69], its use remains confined to research settings, and it has yet to be integrated into clinical practice. Additionally, this method is not restricted to the adductor pollicis muscle, allowing for evaluation of neuromuscular function across various muscle groups.

Several important considerations should be kept in mind when using PMG for perioperative neuromuscular monitoring. In contrast to voluntary muscle contractions, evoked muscle responses exhibit a unique characteristic: a measurable delay between electrical stimulation and the resulting acoustic signal [69]. Research has shown that while this delay is consistent for a given muscle, it varies between different muscles [69]. For example, stimulation of the median nerve produces a sound signal approximately 3.6±1.0 milliseconds afterward, whereas stimulation of the ulnar nerve results in a delay of about 3.9±1.1 milliseconds [69]. Moreover, the amount of subcutaneous fat can influence the clarity and transmission quality of muscle-generated sounds [69].

The first preliminary clinical study revealed good concordance between a prototype and AMG [70], but the study had several inherent limitations. It was initially conceived as a small-scale observational study to preliminarily assess the feasibility of using the PMG prototype for neuromuscular monitoring and refining its technical aspects. As a result, the limited sample size precludes a comprehensive evaluation of the device’s efficacy. The study focused exclusively on rocuronium as the NMB. Moreover, although the study explored the prototype’s neuromuscular monitoring capabilities at varying depths of muscle relaxation, it did not assess its performance across diverse patient populations, such as those who are critically ill, obese, elderly, or pediatric. Finally, the study encountered technical challenges due to early-stage development of the PMG prototype and the aging TOF-Watch SX, which is no longer in production. These issues resulted in frequent operational disruptions, further impacting the study’s execution.

Sonomechanomyography

Sonomechanomyography (SMMG) is a novel method, defined in 2020 by Ling et al [71], who proposed an innovative monitoring technique based on ultra-fast ultrasound imaging to create a two-dimensional representation of transient muscle motion. Ultra-fast ultrasound used in this method provides high frame rates, allowing for precise detection of muscle motion onset. However, this method is sensitive to a subject’s movement that can introduce artifacts, affecting data quality [71].

In December 2024, Zhu et al [72] applied this method to assess the TOFR in 20 healthy adults by analyzing SMMG of the adductor pollicis muscle alongside AMG of the thumb. The findings revealed no statistically significant differences in TOFR measurements between the left and right hands for both AMG and SMMG, as indicated by P values exceeding 0.05. Furthermore, SMMG demonstrated superior accuracy, with a mean relative error of only 0.6%, compared to 1.4% for AMG. Bland–Altman analysis confirmed strong agreement between the 2 methods, with all differences falling within the limits of agreement and a mean bias of just 0.02. Notably, SMMG offered the advantage of not requiring additional calibration prior to testing. Overall, these results suggest that SMMG has potential as a novel quantitative neuromuscular monitoring technique. However, further research is necessary to validate its clinical utility by evaluating its performance in anesthetized patients, both intraoperatively and during recovery.

Future Directions

In 2024, the first study describing the application of machine learning to support the analysis of the TOF ratio was published [73]. Patient data from the Datex-Ohmeda Electromyography (136 patients and a total of 21,891 TOFR measurements) and The TetraGraph system (388 patients and 97 838 TOFR measurements) were merged into a synthetic dataset, which was used to develop and train 4 distinct predictive models. Although an average precision score (95% confidence interval) of 0.48 (0.35, 0.60) indicates that further model development is needed, this result is the first step toward automated removal of outliers in neuromuscular monitoring devices.

An interesting alternative is monitoring diaphragmatic thickness fraction using ultrasound, as proposed by Huang et al [74]. The authors postulated that using diaphragmatic ultrasound to monitor patients during postoperative emergence from anesthesia could be an effective strategy to prevent residual neuromuscular block (RNMB), providing a more comfortable alternative to AMG. They conducted a prospective clinical trial enrolling patients undergoing radical thyroid cancer surgery. Participants were randomly allocated to 1 of 3 study groups: (1) a combined ultrasonography and AMG group (USG+AMG), (2) an AMG-only group, and (3) a standard clinical care group (UCP). The primary endpoints included the frequency of RNMB and hypoxemia following tracheal extubation. Among 127 enrolled patients, the occurrence of RNMB and hypoxemia was significantly higher in the UCP group compared to the US+AMG and AMG groups at 15 and 30 minutes after extubation. Additionally, the area under the receiver operating characteristic (ROC) curve and the decision curve analysis demonstrated that the recovery rate of diaphragmatic thickness fraction (DTF) provided superior predictive value compared to absolute DTF measurements. This indicated that the approach is at least as effective as continuous AMG throughout the perioperative period, offering a viable and potentially more comfortable alternative for patient management.

As demonstrated in this review, neither equipment availability nor specific clinical situations pose a barrier to the routine use of quantitative neuromuscular monitoring. Therefore, initiatives aimed at implementing scientific society guidelines into clinical practice are necessary. A successful example of such an initiative is the study by Weigel et al, who demonstrated that implementation of a multifaceted intervention – including equipment trials, educational videos, quantitative monitors in all anesthetizing locations, electronic clinical decision support with real-time alerts, and ongoing professional practice metrics – was an effective strategy that resulted in shorter lengths of stay in the post-anesthesia care unit, a lower rate of pulmonary complications, and reduced hospital stay durations [75]. These practices should be adapted to local conditions and the challenges faced by anesthesiologists.

Conclusions

This review strongly supports that objective neuromuscular monitoring during general anesthesia should be a mandatory standard of care. Current standards in anesthesia dictate that the restoration of neuromuscular transmission – quantified by a TOFR exceeding 0.9 – is essential to ensure safe extubation. This level of recovery can only be accurately assessed using objective instrumental methods designed to quantify muscle strength. Subjective assessments lack precision and are incapable of detecting residual neuromuscular blockade beyond a TOFR of 0.6, making them unsuitable for modern anesthesiology practice.