Mechanomyography is still the reference method of neuromuscular monitoring [1,2]. Although many different nerve–muscle units have been described as suitable for neuromuscular monitoring, stimulation of the ulnar nerve and measurement of the adduction force of the thumb is preferred. Accordingly, in studies investigating pharmacodynamic properties of muscle relaxants, this technique has been generally used .
Essential conditions to perform exact measurements are the unidirectional position of the adduction direction of the thumb and the force transducer , as well as a stable resting tension (preload) to ensure isometric contraction . Therefore, forearm and hand must be fixed rigidly during the measurements. Since these conditions cannot always be guaranteed in clinical practice with an acceptable effort, mechanomyography has been restricted to scientific issues.
Beyond these technical problems, the patients' muscle force underlies a wide inter-individual variability. To standardize measurements, pharmacodynamic variables of muscle relaxants drawn from mechanomyography (and from all other neuromuscular monitoring techniques) are presented as twitch responses relative to the control force measured before administration of a muscle relaxant or relative to the first twitch response within a train-of-four stimulation. This normalization is accepted to a general extent. Accordingly, none of the commercially available monitors gives the twitch force in Newton.
So far, it has not been demonstrated that the twitch height of the test muscle does or does not influence the normalized parameters of neuromuscular monitoring. Maybe as a matter of principle in science, the experts of the Copenhagen Consensus Conference advised presenting the absolute force during control measurements at least in study reports during the first phase of drug approvals . In order to provide evidence in this respect with clinical data, we evaluated the influence of the twitch height of the adductor pollicis muscle during control measurements on the pharmacodynamic parameters of mivacurium when measured by mechanomyography.
After obtaining the approval of the Ethics Committee of Rostock University and the patients' written informed consent, we prospectively studied 50 adult patients (ASA I–II) scheduled for elective surgery under general anaesthesia. Based on the preoperative physical examination, difficult intubation was not anticipated in any of the patients. None had a history of malignant hyperthermia or was receiving any medication known to interact with neuromuscular blocking agents. Pregnant patients were excluded.
Premedication with oral 7.5–1 mg midazolam was provided at the discretion of the anaesthetist. After 3 min of pre-oxygenation, anaesthesia was induced with propofol (2 mg kg−1) and alfentanil (20 μg kg−1). After loss of consciousness, the patients' trachea was intubated without administration of neuromuscular blocking agents, and mechanical ventilation with air in oxygen (FiO2 = 0.3) was adjusted to maintain an end-tidal carbon dioxide tension between 4.5 and 5.5 kPa. Anaesthesia was maintained with propofol (6–8 mg kg−1 h−1) and alfentanil (20 μg kg−1 h−1) according to electroencephalographic signs of adequate anaesthesia (standard edge frequency 90% ≤ 12 Hz) using the pEEG-Monitor™ (Dräger, Lübeck, Germany). Patients' heart rate, non-invasive arterial blood pressure and arterial oxygen saturation were monitored continuously (AS/3-Monitor™; Datex-Ohmeda, Helsinki, Finland). Skin temperature was checked at the site of the neuromuscular measurements and maintained at 32.0–35.0°C using heating blankets.
We used equipment for nerve stimulation, force measurement and signal processing designed to separately interpret absolute and relative muscle forces following nerve stimulation. All parts of the equipment are also used in clinical setups or are devices drawn from clinical monitors. In detail, a nerve stimulator (Myotest DBS ™; Biometer, Copenhagen, Denmark) guaranteed square pulses of exactly 200 μs duration up to a resistance of 3.5 kΩ. A myograph 2000™ including the force transducer TD 100 ™ (Biometer, Copenhagen, Denmark) was connected to an analogue/digital converter card (DAS 1601; Measurement Computing Corporation, Middleboro, MA, USA) plugged into a laptop computer (Peacock, Bad Wünnenberg-Haaren, Germany). The converter card (resolution: 12 Bit; linearity: ±1 Bit; conversion failure <0.03%) included a crystal controlled oscillator that provided the time base to equidistantly sample the signals. The force transducer had a resolution of <40 Hz, thus, the sample rate of our workspace with 3000 Hz guaranteed a complete documentation of the force signals. A signal distributor managed information data flow between the hard disk of the laptop, the nerve stimulator and the force transducer. Sensors of the workspace were calibrated, especially the force transducer, with standard weights. The calibration of the complete workspace was performed directly before the study. Data of this unit were collected, calculated and graphically presented using the software package Test Point 1.1. for Windows (Keithley Instruments, Cleveland, OH, USA).
We monitored neuromuscular transmission mechanomyographically by measuring the force of the adductor pollicis muscle following stimulation of the ulnar nerve. These measurements and the setup were thoroughly adapted to the guidelines for good clinical research practice in pharmacodynamic studies of neuromuscular blocking agents . In detail, the forearm was rigidly immobilized, the force transducer was connected unidirectionally to the thumb's adduction, and the thumb was abducted with a preload of 2.0–2.9 N, which yielded maximal twitch response. Best position of the stimulating electrodes was assumed when an initial threshold of stimulation with single twitches (1 Hz) was achieved below 20 mA. The supra-maximal stimulus (15% above the stimulus necessary for the maximal twitch response) was established by single twitches (1 Hz), the baseline mechanomyographic response was stabilized over a period of 20 min again with single twitches (1 Hz). Thereafter, the position and direction of the force transducer were optimized in order to achieve highest values, and the transducer was once again calibrated. After 10 min of stable twitch response, the individual measured force of the adductor pollicis muscle was determined and served as ‘baseline twitch height’ (T0). After these measurements, the stimulation mode was switched to train-of-four stimuli every 12 s and a single bolus of mivacurium (75 μg kg−1) was administered.
The clinical pharmacology of mivacurium was described with several parameters. The force of the first twitch response of each train related to the control force (T1/T0) and the ratio of the height of the fourth related to the first response (T4/T1) were calculated. Maximum neuromuscular blockade is defined by the minimum T1/T0 following injection of mivacurium (1 − T1/T0), the onset time by the time from injection of mivacurium to the maximum neuromuscular blockade, the maximum onset speed by the highest individual difference between two consecutive T1/T0 values per minute, duration of action by the time from injection of mivacurium to T1/T0 > 0.9 or by the time from injection of mivacurium to T4/T1 > 0.8, and the speed of recovery by the recovery index, i.e., the time between T1/T0 = 0.25 and T1/T0 = 0.75.
To investigate the effect of the baseline contraction force of the adductor pollicis muscle on pharmacodynamic variables obtained from measurements at the same muscle, patients were divided into two groups: the data of patients whose thumb adduction force was below the median value of all patients were collected in a ‘low force’ group and, accordingly, the data of all other patients were collected in a ‘high force’ group.
The correlation between the contraction force of the adductor pollicis at baseline and the onset time was calculated by a linear least-squares regression. Consecutive values of the neuromuscular block (1 − T1/T0) and the values of the onset speed of the neuromuscular block (ΔT1/T0/Δt) following injection of mivacurium were subjected to a two-way repeated-measurement analysis of variance (ANOVA) with the within-groups factor, time, the between-groups factor, force and their interaction term (time × force). If this interaction term was significant (P < 0.05), different onset properties between the groups were assumed. The patients' characteristics and the pharmacodynamic variables were compared by two-sided t-tests. The distribution of gender was compared by χ2-test. All variables are presented as mean ± standard deviation (SD). Statistical analyses were performed using SPSS 11.0 for Windows (SPSS Inc., Chicago, IL, USA).
The onset time of mivacurium was inversely related to the contraction force of the adductor pollicis muscle before administration of mivacurium (onset time = (410 − 19s N−1) × force; r = 0.763; P < 0.0001). Variables are grouped according to a high or a low baseline contraction force of the adductor pollicis muscle. The groups of patients were comparable with regard to age, height, weight and sex distribution (Table 1). According to the Good Clinical Practice guidelines, two patients were withdrawn from the study because the initial recorded preload decreased by about 50% during measurements. All other patients' initial load did not drift during measurements (Table 2). Neither both documented preloads nor stimulation current to achieve maximum twitch response nor skin temperature during measurements differed between the groups (Table 2). Onset of neuromuscular block following 75 μg kg−1 mivacurium was faster in patients with a high baseline twitch height of the adductor pollicis muscle when compared to patients with low baseline twitch height of the adductor pollicis muscle (Table 3; Fig. 1). Recovery and duration properties did not differ between the groups.
In this study, the onset of a neuromuscular block by mivacurium was related to the baseline twitch height of the adductor pollicis muscle. Patients who responded to supra-maximal stimulation of the ulnar nerve during baseline measurements with a higher force presented a faster onset speed of neuromuscular block. This faster onset speed manifested itself in a steeper increase of the block per time showing higher and earlier maximum values in the onset speed of the twitch response.
The differences in the onset character significantly affected the commonly accepted onset variable of a neuromuscular blockade, the onset time to the maximum block. Since the onset of mivacurium was statistically significantly prolonged by more than 1 min in thumbs with lower baseline twitch heights, the delay must be rated clinically relevant if it is representative for other muscles in a respective patient. In this study protocol, however, we focused on an exact monitoring of the time course blocking the adductor pollicis muscle, which is the most-referenced one. Thus, we cannot decide if the thumb's delay would have correlated, e.g., with worse intubating conditions .
Many factors affect neuromuscular measurements. Increasing the frequency of nerve stimulation decreases onset time . Similarly, train-of-four stimulation results in shorter onset times and deeper relaxation than single twitch stimulation. Increasing periods of control stimulation from 1 to 20 min are associated with decreasing onset times and prolonged duration of action . Therefore, we stabilized twitch response for exactly 20 min before the baseline twitch response was recorded. During this period, twitch height increased (staircase phenomenon) , but it was stable in each patient for the last 10 min. Likewise, further possible confounding factors, e.g., preload, skin temperature and stimulation current were thoroughly controlled in this study (Table 2).
The statistically and methodologically stable observation presented is, therefore, of principle interest, reflecting either different responses to neuromuscular blocking agents with respect to patients' muscle power or a problem of the mechanomyographic measuring technique. The latter is of major interest since mechanomyography, which is still the gold standard of neuromuscular monitoring, would produce biased results, at least, if test persons with different forceful (or athletic) thumbs were recruited for a study. Since in this investigation the differences in thumb force could not be predicted by biometric parameters (height, body weight or body mass index), we confirm the postulate given by the Copenhagen Consensus Conference to present the absolute forces during baseline measurements in scientific manuscripts .
Besides these considerations, some speculations about the reasons for a less steep and a slower decrease in neuromuscular blockade following an ED95 of mivacurium in less powerful thumbs should be addressed. Differences between onset times and sensitivity in different muscles of one individual have consistently been reported for all muscle relaxants and also for mivacurium [9,10]. From these investigations it has been deduced that bigger muscle fibres in type or size are more sensitive to muscle relaxants [11-13]. In our investigation, patients with a higher force of their thumbs showed both a shorter onset time and also a deeper neuromuscular blockade in the respective muscle. Since there is no reason to doubt that larger muscles are in many cases more forceful than smaller ones, the higher twitch height may be causal for the increased sensitivity to mivacurium.
Another explanation might be different pharmacokinetic conditions, e.g., better muscle perfusion in the stronger and probably larger adductor muscles . Neither the systemic haemodynamic conditions nor the skin temperature as an indicator for a regional circulatory insufficiency and additional impact on neuromuscular function and monitoring , differed in these two groups of ASA I–II patients. We, therefore, tend to reject this view. Nonetheless, we cannot exclude that perfusion of the adductor pollicis, when evaluated by invasive techniques (e.g., Laser Doppler flowmetry), might be different in our groups.
Importantly, the differences in the onset of the mivacurium-induced neuromuscular blockade were not paralleled during its recovery. This is important to note since we do not aim at discouraging anaesthetists from using neuromuscular monitoring with further methodological discussions [16-18].
Although a direct relationship between the adductor pollicis twitch height and general muscle power, leading to differences in the sensitivity to neuromuscular blocking agents, cannot be drawn from our study, our data demonstrate the scientific importance and clinical relevance of the consideration of twitch-height/force during pharmacodynamic studies to neuromuscular blocking agents. Furthermore this approach is highly desirable to meet the requirements of the Copenhagen Consensus Conference .
Finally our study underlines the ‘gold standard’ – position of mechanomyography since electromyographic technique does not measure force by definition  and one may speculate if onset times obtained by this technique are not sensitive to baseline muscle force and, therefore, result in more stable values.
This work was supported by the German Research Foundation (FR 1606/1-1).
1. Katz RL. A nerve stimulator for the continuous monitoring of muscle relaxant action. Anesthesiology
1965; 26: 832–833.
2. Viby-Mogensen J. Clinical assessment of neuromuscular transmission. Br J Anaesth
1982; 54: 209–223.
3. Viby Mogensen J, Engbaek J, Eriksson L et al
. Good clinical research practice (GCRP) in pharmacodynamic studies of neuromuscular blocking agents. Acta Anaesthesiol Scand
1996; 40: 59–74.
4. Ali H, Savarese JJ. Monitoring of neuromuscular function. Anesthesiology
1976; 45: 216–249.
5. Donlon Jr JV, Savarese JJ, Ali HH. Cumulative dose– response curves for gallamine: effect of altered resting thumb tension and mode of stimulation. Anesth Analg
1979; 58: 377–381.
6. Meretoja OA, Taivainen T, Brandom BW, Wirtavuori K. Frequency of train-of-four stimulation influences neuromuscular response. Br J Anaesth
1994; 72: 686–687.
7. McCoy EP, Mirakhur RK, Connolly FM, Loan PB. The influence of the duration of control stimulation on the onset and recovery of neuromuscular block. Anesth Analg
1995; 80: 364–367.
8. Lee GC, Iyengar S, Szenohradszky J et al
. Improving the design of muscle relaxant studies. Stabilization period and tetanic recruitment. Anesthesiology
1997; 86: 48–54.
9. Le Corre F, Plaud B, Benhamou E, Debaene B. Visual estimation of onset time at the orbicularis oculi after five muscle relaxants: application to clinical monitoring of tracheal intubation. Anesth Analg
1999; 89: 1305–1310.
10. Plaud B, Debaene B, Lequeau F, Meistelman C, Donati F. Mivacurium neuromuscular block at the adductor muscles of the larynx and adductor pollicis in humans. Anesthesiology
1996; 85: 77–81.
11. Ibebunjo C, Srikant CB, Donati F. Properties of fibres, endplates and acetylcholine receptors in the diaphragm, masseter, laryngeal, abdominal and limb muscles in the goat. Can J Anaesth
1996; 43: 475–484.
12. Ibebunjo C, Srikant CB, Donati F. Duration of succinylcholine and vecuronium blockade but not potency correlates with the ratio of endplate size to fibre size in seven muscles in the goat. Can J Anaesth
1996; 43: 485–494.
13. Ibebunjo C, Hall LW. Muscle fibre diameter and sensitivity to neuromuscular blocking drugs. Br J Anaesth
1993; 71: 732–733.
14. Goat VA, Yeung ML, Blakeney C, Feldman SA. The effect of blood flow upon the activity of gallamine triethiodide. Br J Anaesth
1976; 48: 69–73.
15. Eriksson LI, Lennmarken C, Jensen E, Viby Mogensen J. Twitch tension and train-of-four ratio during prolonged neuromuscular monitoring at different peripheral temperatures. Acta Anaesthesiol Scand
1991; 35: 247–252.
16. Fuchs-Buder T, Hofmockel R, Geldner G, Diefenbach C, Ulm K, Blobner M. Einsatz des neuromuskulären Monitorings in Deutschland. Der Anaesthesist
2003; 52: 522–526.
17. Blobner M, Söhnel A, van de Roemer A et al
. Anwendung von Muskelrelaxanzien in Deutschland – eine Umfrage an deutschen Anästhesieeinrichtungen. Der Anaesthesist
2003; 52: 427–434.
18. Gatke MR, Viby-Mogensen J, Rosenstock C, Jensen FS, Skovgaard LT. Postoperative muscle paralysis after rocuronium: less residual block when acceleromyography is used. Acta Anaesthesiol Scand
2002; 46: 207–213.
19. Peterson CR, Wise CS. Electromyographic method for objective measurement of muscle relaxant drugs. Arch Phys Med Rehabil
1961; 42: 566–572.