Machine learning based analysis and detection of trend outliers for electromyographic neuromuscular monitoring.

PURPOSE: Neuromuscular monitoring is frequently plagued by artefacts, which along with the frequent unawareness of the principles of this subtype of monitoring by many clinicians, tends to lead to a cynical attitute by clinicians towards these monitors. As such, the present study aims to derive a feature set and evaluate its discriminative performance for the purpose of Train-of-Four Ratio (TOF-R) outlier analysis during continuous intraoperative EMG-based neuromuscular monitoring. METHODS: Patient data was sourced from two devices: (1) Datex-Ohmeda Electromyography (EMG) E-NMT: a dataset derived from a prospective observational trial including 136 patients (21,891 TOF-R observations), further subdivided in two based on the type of features included; and (2) TetraGraph: a clinical case repository dataset of 388 patients (97,838 TOF-R observations). The two datasets were combined to create a synthetic set, which included shared features across the two. This process led to the training of four distinct models. RESULTS: The models showed an adequate bias/variance balance, suggesting no overfitting or underfitting. Models 1 and 2 consistently outperformed the others, with the former achieving an F1 score of 0.41 (0.31, 0.50) and an average precision score (95% CI) of 0.48 (0.35, 0.60). A random forest model analysis indicated that engineered TOF-R features were proportionally more influential in model performance than basic features. CONCLUSIONS: Engineered TOF-R trend features and the resulting Cost-Sensitive Logistic Regression (CSLR) models provide useful insights and serve as a potential first step towards the automated removal of outliers for neuromuscular monitoring devices. TRIAL REGISTRATION: NCT04518761 (clinicaltrials.gov), registered on 19 August 2020.

The Incidence of Residual Neuromuscular Block in Pediatrics: A Prospective, Pragmatic, Multi-institutional Cohort Study.

Introduction Residual neuromuscular block, defined as a quantitatively measured train-of-four ratio (TOFr) <0.9, is common postoperatively. Using a pragmatic trial design, we hypothesized that qualitative and/or clinical assessment of neuromuscular block would inadequately detect residual block following antagonism with neostigmine or sugammadex. Method After IRB approval and written informed consent, 74 children (aged 2-17 years), undergoing elective surgery and receiving rocuronium, were prospectively enrolled in the study at Children's Hospital Colorado and Children's Healthcare of Atlanta. Routine clinical practice at both institutions consisted of clinical signs and/or qualitative assessment with peripheral nerve stimulators. Children at the Colorado hospital routinely received sugammadex antagonism; whereas children at the Atlanta hospital received neostigmine. Residual neuromuscular block was assessed postoperatively using quantitative electromyography. If TOFr was <0.9, patients received sugammadex until TOFr ≥0.9. Result Qualitative and clinical assessment failed to detect residual block in 29.7% of patients in the neostigmine reversal cohort (adjusted odds ratio (aOR) 29.8, 95% confidence interval (CI): 2.7 to 5,559.5, p-value = 0.002). No residual block was detected in the sugammadex reversal cohort. A correlation between increasing patient weight and incidence of postoperative residual block was observed in the neostigmine cohort (aOR 1.05, 95% CI: 1.02 to 1.10, p-value = 0.002). Conclusion Qualitative and/or clinical assessment of neuromuscular block inadequately detects residual block following neostigmine antagonism.

Corrigendum to "Validation of a convolutional neural network that reliably identifies electromyographic compound motor action potentials following train-of-four stimulation: an algorithm development experimental study" [BJA Open 8 (2023) 100236].

[This corrects the article DOI: 10.1016/j.bjao.2023.100236.].

Validation of a convolutional neural network that reliably identifies electromyographic compound motor action potentials following train-of-four stimulation: an algorithm development experimental study.

Background: International guidelines recommend quantitative neuromuscular monitoring when administering neuromuscular blocking agents. The train-of-four count is important for determining the depth of block and appropriate reversal agents and doses. However, identifying valid compound motor action potentials (cMAPs) during surgery can be challenging because of low-amplitude signals and an inability to observe motor responses. A convolutional neural network (CNN) to classify cMAPs as valid or not might improve the accuracy of such determinations. Methods: We modified a high-accuracy CNN originally developed to identify handwritten numbers. For training, we used digitised electromyograph waveforms (TetraGraph) from a previous study of 29 patients and tuned the model parameters using leave-one-out cross-validation. External validation used a dataset of 19 patients from another study with the same neuromuscular block monitor but with different patient, surgical, and protocol characteristics. All patients underwent ulnar nerve stimulation at the wrist and the surface electromyogram was recorded from the adductor pollicis muscle. Results: The tuned CNN performed highly on the validation dataset, with an accuracy of 0.9997 (99% confidence interval 0.9994-0.9999) and F1 score=0.9998. Performance was equally good for classifying the four individual responses in the train-of-four sequence. The calibration plot showed excellent agreement between the predicted probabilities and the actual prevalence of valid cMAPs. Ten-fold cross-validation using all data showed similar high performance. Conclusions: The CNN distinguished valid cMAPs from artifacts after ulnar nerve stimulation at the wrist with >99.5% accuracy. Incorporation of such a process within quantitative electromyographic neuromuscular block monitors is feasible.

Good clinical research practice (GCRP) in pharmacodynamic studies of neuromuscular blocking agents III: The 2023 Geneva revision.

The set of guidelines for good clinical research practice in pharmacodynamic studies of neuromuscular blocking agents was developed following an international consensus conference in Copenhagen in 1996 (Viby-Mogensen et al., Acta Anaesthesiol Scand 1996, 40, 59-74); the guidelines were later revised and updated following the second consensus conference in Stockholm in 2005 (Fuchs-Buder et al., Acta Anaesthesiol Scand 2007, 51, 789-808). In view of new devices and further development of monitoring technologies that emerged since then, (e.g., electromyography, three-dimensional acceleromyography, kinemyography) as well as novel compounds (e.g., sugammadex) a review and update of these recommendations became necessary. The intent of these revised guidelines is to continue to help clinical researchers to conduct high-quality work and advance the field by enhancing the standards, consistency, and comparability of clinical studies. There is growing awareness of the importance of consensus-based reporting standards in clinical trials and observational studies. Such global initiatives are necessary in order to minimize heterogeneous and inadequate data reporting and to improve clarity and comparability between different studies and study cohorts. Variations in definitions of endpoints or outcome variables can introduce confusion and difficulties in interpretation of data, but more importantly, it may preclude building of an adequate body of evidence to achieve reliable conclusions and recommendations. Clinical research in neuromuscular pharmacology and physiology is no exception.

Comparison of visual and electromyographic assessments with train-of-four stimulation of the ulnar nerve: a prospective cohort study.

PURPOSE: The use of a peripheral nerve stimulator to assess the level of neuromuscular blockade tasks the anesthesia clinician with subjectively assessing the response to neurostimulation. In contrast, objective neuromuscular monitors provide quantitative information. The purpose of this study was to compare subjective evaluations from a peripheral nerve stimulator with objective measurements of neurostimulation responses from a quantitative monitor. METHODS: Patients were enrolled preoperatively, and intraoperative neuromuscular blockade management was at the discretion of the anesthesiologist. Electromyography electrodes were placed over the dominant or nondominant arm in a randomized fashion. Following onset of nondepolarizing neuromuscular blockade, the ulnar nerve was stimulated, the response was measured with electromyography, and anesthesia clinicians, who were blinded to the objective measurements, subjectively (visually) evaluated the response to neurostimulation. RESULTS: Fifty patients were enrolled and 666 neurostimulations were performed at 333 different time points. Anesthesia clinicians subjectively overestimated the response of the adductor pollicis muscle following neurostimulation of the ulnar nerve 155/333 (47%) of the time when compared with objective electromyographic measurements. When subjective evaluations and objective measurements differed to any degree, subjective evaluations were higher than objective measurements 155/166 (92%) of the time (95% CI, 87 to 95; P < 0.001), representing significant evidence that subjective evaluation overestimates the response to train-of-four stimulation. CONCLUSIONS: Subjective observations of a "twitch" do not consistently correspond to objective measurements of neuromuscular blockade with electromyography. Subjective evaluation overestimates the response to neurostimulation and may be unreliable for determining the depth of block or confirming adequate recovery.

RéSUMé: OBJECTIF: Lors de l'utilisation d'un stimulateur nerveux périphérique pour évaluer le niveau de bloc neuromusculaire, le clinicien en anesthésie doit évaluer de manière subjective la réponse à la neurostimulation. Les moniteurs neuromusculaires objectifs fournissent quant à eux des informations quantitatives. L'objectif de cette étude était de comparer les évaluations subjectives d'un stimulateur nerveux périphérique avec des mesures objectives des réponses de neurostimulation provenant d'un moniteur quantitatif. MéTHODE: Les patients ont été recrutés avant leur opération, et la gestion du bloc neuromusculaire peropératoire était à la discrétion de l'anesthésiologiste. Des électrodes d'électromyographie ont été placées sur le bras dominant ou non dominant de manière aléatoire. Après l'amorce du bloc neuromusculaire non dépolarisant, le nerf cubital a été stimulé, la réponse a été mesurée par électromyographie, et les cliniciens en anesthésie, qui n'avaient pas accès aux mesures objectives, ont évalué subjectivement (visuellement) la réponse à la neurostimulation. RéSULTATS: Cinquante patients ont été recrutés et 666 neurostimulations ont été réalisées à 333 moments différents. Les cliniciens en anesthésie ont subjectivement surestimé la réponse du muscle adducteur du pouce après neurostimulation du nerf cubital 155/333 (47 %) fois par rapport aux mesures électromyographiques objectives. Lorsque les évaluations subjectives et les mesures objectives différaient à quelque degré que ce soit, les évaluations subjectives étaient plus élevées que les mesures objectives 155/166 (92 %) du temps (IC 95 %, 87 à 95; P < 0,001), ce qui représente une preuve significative que l'évaluation subjective surestime la réponse à la stimulation par train-de-quatre. CONCLUSION: Les observations subjectives d'une « contraction ¼ ne correspondent pas systématiquement aux mesures objectives d'un bloc neuromusculaire telles qu'obtenues par électromyographie. L'évaluation subjective surestime la réponse à la neurostimulation et pourrait ne pas être fiable pour déterminer la profondeur du bloc ou confirmer une récupération adéquate.

A survey on the availability, usage and perception of neuromuscular monitors in Europe.

PURPOSE: Neuromuscular blocking agents (NMBAs) are routinely administered to patients in a multiplicity of anesthetic settings. Absence of postoperative residual neuromuscular block is widely considered an anesthetic patient safety mandate. Despite the increasing availability of a wider range of commercial quantitative neuromuscular monitors, the availability and use of neuromuscular monitoring devices is deemed to be suboptimal even in countries with above-average health system ratings. The present study aims to more precisely characterize the perceived availability, cost sensitivity and usability of neuromuscular monitoring devices within European anesthesia departments. METHODS: A pre-registered internet-based survey assessing the availability, cost sensitivity and usability of neuromuscular monitoring devices was distributed as e-mail newsletter by the European Society of Anaesthesiology and Intensive Care (ESAIC) to all of its active full members. The survey was available online for a total of 120 days. RESULTS: Having targeted a total of 7472 ESAIC members, the survey was completed by a total of 692 anesthesiologists (9.3%) distributed across 37 different European countries. Quantitative monitors were reported to be proportionally more available than qualitative ones (87.6% vs. 62.6%, respectively), as well as in greater monitor-per-operating room ratios. Most anesthesiologists (60.5%) expressed moderate confidence in quantitative monitors, with artifactual recordings and inaccurate measurements being the most frequently encountered issues (25.9%). The commercial pricing of quantitative devices was considered more representative of a device's true value, when compared to qualitative instruments (average cost of €4.500 and €1.000 per device, respectively). CONCLUSION: The availability of quantitative NMM in European operating theaters has increased in comparison with that reported in previous decades, potentially indicating increasing monitoring rates. European anesthesiologists express moderate confidence in quantitative neuromuscular monitors, along with a sentiment of adequate pricing when compared to their qualitative counterparts. Trust in quantitative monitors is marked by caution and awareness for artifactual recordings, with a consequent expectation that developments focusing on accuracy, reliability and ergonomics of neuromuscular monitors be prioritized.

Quantitative Neuromuscular Monitoring and Postoperative Outcomes: A Narrative Review.

Over the past five decades, quantitative neuromuscular monitoring devices have been used to examine the incidence of postoperative residual neuromuscular block in international clinical practices, and to determine their role in reducing the risk of residual neuromuscular block and associated adverse clinical outcomes. Several clinical trials and a recent meta-analysis have documented that the intraoperative application of quantitative monitoring significantly reduces the risk of residual neuromuscular blockade in the operating room and postanesthesia care unit. In addition, emerging data show that quantitative monitoring minimizes the risk of adverse clinical events, such as unplanned postoperative reintubations, hypoxemia, and postoperative episodes of airway obstruction associated with incomplete neuromuscular recovery, and may improve postoperative respiratory outcomes. Several international anesthesia societies have recommended that quantitative monitoring be performed whenever a neuromuscular blocking agent is administered. Therefore, a comprehensive review of the literature was performed to determine the potential benefits of quantitative monitoring in the perioperative setting.

Ipsilateral and Simultaneous Comparison of Responses from Acceleromyography- and Electromyography-based Neuromuscular Monitors.

BACKGROUND: The paucity of easy-to-use, reliable objective neuromuscular monitors is an obstacle to universal adoption of routine neuromuscular monitoring. Electromyography (EMG) has been proposed as the optimal neuromuscular monitoring technology since it addresses several acceleromyography limitations. This clinical study compared simultaneous neuromuscular responses recorded from induction of neuromuscular block until recovery using the acceleromyography-based TOF-Watch SX and EMG-based TetraGraph. METHODS: Fifty consenting patients participated. The acceleromyography and EMG devices analyzed simultaneous contractions (acceleromyography) and muscle action potentials (EMG) from the adductor pollicis muscle by synchronization via fiber optic cable link. Bland-Altman analysis described the agreement between devices during distinct phases of neuromuscular block. The primary endpoint was agreement of acceleromyography- and EMG-derived normalized train-of-four ratios greater than or equal to 80%. Secondary endpoints were agreement in the recovery train-of-four ratio range less than 80% and agreement of baseline train-of-four ratios between the devices. RESULTS: Acceleromyography showed normalized train-of-four ratio greater than or equal to 80% earlier than EMG. When acceleromyography showed train-of-four ratio greater than or equal to 80% (n = 2,929), the bias was 1.3 toward acceleromyography (limits of agreement, -14.0 to 16.6). When EMG showed train-of-four ratio greater than or equal to 80% (n = 2,284), the bias was -0.5 toward EMG (-14.7 to 13.6). In the acceleromyography range train-of-four ratio less than 80% (n = 2,802), the bias was 2.1 (-16.1 to 20.2), and in the EMG range train-of-four ratio less than 80% (n = 3,447), it was 2.6 (-14.4 to 19.6). Baseline train-of-four ratios were higher and more variable with acceleromyography than with EMG. CONCLUSIONS: Bias was lower than in previous studies. Limits of agreement were wider than expected because acceleromyography readings varied more than EMG both at baseline and during recovery. The EMG-based monitor had higher precision and greater repeatability than acceleromyography. This difference between monitors was even greater when EMG data were compared to raw (nonnormalized) acceleromyography measurements. The EMG monitor is a better indicator of adequate recovery from neuromuscular block and readiness for safe tracheal extubation than the acceleromyography monitor.