Myasthenia gravis is an autoimmune disease that compromises neuromuscular transmission by reducing the number of acetylcholine receptors (AChR) at the neuromuscular junction. By reducing these AChRs, the disease has a significant influence on the time course of effect of neuromuscular blocking agents. Compared to ‘healthy’ patients, non-depolarizing neuromuscular blocking agents will produce a prolonged and more intense neuromuscular blockade in myasthenic patients. Reducing the dose does not completely reverse these effects.
In previous studies in myasthenic patients and pigs we demonstrated that the pharmacokinetics of neuromuscular blocking agents do not change to a clinically relevant extent [1,2]. Previously, we published an isolated muscle model, in which we studied the time course of effect of neuromuscular blocking agents in the anterior tibial muscle of rats, the antegrade perfused rat peroneal nerve anterior tibialis muscle model . In this model, it is possible to control pharmacokinetics by perfusing the muscle with donor blood, which contains a known amount of neuromuscular blocking agents.
We previously demonstrated the increased sensitivity and prolonged time course of effect of muscle relaxants in myasthenia gravis, both in humans and in pigs [1,2]. With pharmacokinetic-pharmacodynamic modelling, we found a decreased receptor concentration which we could confirm with our laboratory findings of a decreased AChR concentration in pigs. These observations supported our hypothesis that the decreased receptor concentration results in the altered sensitivity and time course of effect of neuromuscular blocking agents in myasthenic patients. It is the aim of the present study to provide further evidence for this hypothesis that the pharmacodynamic relationship between myasthenia gravis and muscle relaxants is the result of a decreased receptor concentration. To show this, we used two different ways of reducing the AChR concentration at the neuromuscular junction: infusion of α-bungarotoxin (α-BTX) which binds irreversibly to the AChR and passive transfer of antibodies against the AChR.
The preparation of the experimental setup has been extensively described before  and is summarized here. The experiments were conducted following approval of the local Committee for Animal Experiments, University of Groningen, The Netherlands.
Rats were anaesthetized with 60 mg kg−1 pentobarbital sodium intraperitoneally. Anaesthesia was maintained with small increments of pentobarbital sodium, followed by 1-2 mL saline via a catheter inserted in a tail vein. Nadroparine (1000 IE kg−1) (Fraxiparine®; Sanofi-Synthélabo, Praha, Czech Republic), a low molecular weight heparin, was given subcutaneously prior to surgical preparation. The trachea was intubated and artificial ventilation was maintained with room air delivered by an infant ventilator at a frequency of 90 min−1 and a tidal volume of 20 mL kg−1. Rectal temperature was measured continuously during the experiment, and the rats were maintained at a body temperature of 38°C by means of a heating blanket connected to a temperature regulator.
The tibialis anterior muscle was dissected free from the surrounding muscles, and connected to a force transducer. The peroneal nerve was connected to a nerve stimulator. Stainless steel pins were inserted through the proximal and distal end of the tibia. Polythene tubing (Portex Fine Bore Polythene Tubing; SIMS Portex Ltd., Hythe, UK) was placed into the popliteal vein and artery. The rat was then transferred to a thermostatically controlled plexiglas chamber, in which the temperature was maintained at 29°C.
Male rats (body weight approximately 400 g) were anaesthetized with pentobarbital sodium (60 mg kg−1) intraperitoneally. Male rats were chosen because of their higher circulating blood volume. Gelofusine® (B. Braun Melsungen AG, Melsungen, Germany) (10 mL kg−1) and nadroparine (1000 IE kg−1) were given intravenously and 10 min later blood was collected from the abdominal aorta. Nadroparine (100 IE) and acetylsalicylic acid (2 mg) were added to the collected blood before transfer to a cooled reservoir. A magnetic stirrer stirred the blood continuously.
The arterial blood was oxygenated and warmed by a heat exchanger before entering the tibial muscle. The muscle was single pass perfused with a pulsatile flow and left the muscle via the popliteal vein.
The twitch tension of the anterior tibial muscle, elicited by supra-maximal square wave stimuli of 0.2 ms duration and applied to the common peroneal nerve at 0.1 Hz, was recorded by means of a force transducer. Preload was measured continuously and kept constant at approximately 15 g.
After preparation, the muscle was allowed to stabilize, and to regain normal temperature.
The anterior tibial muscle of male Wistar rats (bodyweight 300-400 g) (Harlan Nederland, Horst, The Netherlands) was perfused with a pulsatile flow at a rate of 200 μL min−1 (which corresponds with approximately 200 μL min−1 g muscle−1). A neuromuscular blocking agent, rocuronium or pancuronium in a randomized fashion was infused in a branch of the arterial line at a constant rate of approximately 5 μL min−1 until a stable 90% depression of twitch height was reached. If needed, the rate of infusion of rocuronium or pancuronium was adjusted to reach the required degree of twitch depression. After 2 min of stable block, the infusion was turned off. The concentrations of the infusions were 100 μg mL−1 for rocuronium and 25 μg mL−1 for pancuronium in all experiments. The concentration of neuromuscular blocking agent in blood that arrives in the muscle can be calculated as follows. If the perfusate flow (blood flow into the muscle) is 200 μL min−1, pancuronium (0.025 μg μL−1) is infused at 5 μL min−1 into this perfusate flow of 200 μL min−1. This makes a total perfusate flow of 205 μL min−1 and a pancuronium flow of 0.125 μg min−1, resulting in a pancuronium concentration of 610 μg L−1 or 0.832 μmol. To determine the onset and offset times, 30 min after the twitch height had returned to normal, a second infusion with the previously determined rate for rocuronium or pancuronium was started until a stable 90% block for at least 2 min was obtained. The infusion rate remained unchanged during the determination of onset and offset variables. After the twitch had returned to its control value, an infusion with α-BTX (concentration 25 μg mL−1, flow 5 μL min−1) was started, until there was a slight visually detectable decrease in twitch height. The infusion of α-BTX was then stopped and the twitch height was allowed to stabilize, the new twitch height was set at 100%. After stabilization, an infusion with the same neuromuscular blocking agent was started and the rate was adjusted to obtain a 90% block again. In this group, animals served as their own controls. After finishing the experimental procedure, rats were killed by an overdose of pentobarbital sodium.
Experimental autoimmune myasthenia gravis
Female Lewis rats (bodyweight 200 g) (Harlan Nederland, Horst, The Netherlands) were injected intraperitoneally with 20 pmol AChR binding capacity per 100 g bodyweight of concentrated culture supernatant containing rat anti-AChR monoclonal antibody 35. Monoclonal antibody 35 is a rat immunoglobulin G (IgG) 1 monoclonal antibody directed against the main immunogenic region on the AChR α-subunit . Clinical scoring of a successful induction of passive transfer experimental autoimmune myasthenia gravis was based on the presence of tremor, hunched posture, decreased muscle strength, fatigability and weight loss . Decreased muscle strength was tested by placing the rats on top of their cages and lifting them by the tail. Control rats will hold on to the cage, while myasthenic rats let go very easily. We chose female Lewis rats because rendering these animals myasthenic is a well-established method . Control rats for this group were not immunized and underwent the same experimental procedure.
Muscle preparation was as previously described. The muscle was perfused with a pulsatile flow of 100 μL min−1 (corresponding with approximately 200 μL min−1 g muscle−1). A neuromuscular blocking agent (rocuronium or pancuronium, randomized) was infused in a branch of the arterial line at a constant rate of approximately 2.5 μL min−1 until a stable 90% depression of twitch height was reached. The infusion was then turned off. If needed, the rate of infusion was adjusted so that the twitch depression would reach 90%. To determine the onset and offset times, 30 min after the twitch height had returned to normal a second infusion with the previously determined rate for rocuronium or pancuronium was started until a stable 90% block for at least 2 min was obtained. It should be clear that in this group too the infusion rate remained unchanged during the determination of onset and offset variables. Another 30 min after the twitch height returned to control, an infusion was started with the alternate neuromuscular blocking agent. After the experiment was finished, both anterior tibial muscles were removed and immediately frozen at −80°C, while the rat was still under anaesthesia, to prevent the degradation of the AChR. Muscles were stored at −80°C until they were analysed. After finishing the experimental procedure, rats were killed by an overdose of pentobarbital sodium.
Determination of AChR concentration
The concentration of AChR in anterior tibial muscles of experimental autoimmune myasthenic rats was determined by radio-immunoassay as described previously with minor modifications [7,8]. Briefly, frozen muscle was homogenized at 4°C and AChR was extracted with 2% Triton X-100 (Sigma, Brunschwig Chemie b.v., Amsterdam, The Netherlands). An aliquot of 250 μL of each extract was labelled with 2 × 10−9 M 125I-α-BTX (specific activity 74 TBq mmol−1, Amersham, Little Chalfont, UK), incubated overnight with excess rat anti-AChR IgG (Department of Psychiatry and Neuropsychology, University of Maastricht, Maastricht, The Netherlands) and precipitated by goat anti-rat antibodies. AChR concentration is expressed as loss of receptor concentration compared to the healthy controls.
Results are shown as median (range) unless otherwise stated. The U-test was used to detect statistical differences between different treatment groups. The Wilcoxon signed ranks sum test was used to detect statistical differences for the rats that served as their own controls. Intervals 75-25% during onset and intervals 25-75% during offset were obtained from the mechanomyographic recordings. The interval 75-25% was defined as the time elapsed between 75% twitch height and 25% twitch height during the onset of the neuromuscular block; the interval 25-75% was defined as the time elapsed between 25% twitch height and 75% twitch height during the offset of the neuromuscular block. We will refer to these intervals by ‘onset’ and ‘offset’ in the subsequent sections.
The median concentration of rocuronium in the infused blood to obtain a 90% neuromuscular block was 5.01 μM (4.78-6.31 μM) in control rats vs. 2.82 μM (2.58-3.61 μM) following the α-BTX infusion. For pancuronium the median concentration was 1.07 μM (0.91-1.15 μM) in control rats vs. 0.60 μM (0.50-0.70 μM) following the α-BTX infusion. The effects of the decreased AChR concentration caused by the α-BTX infusion on onset and offset of neuromuscular block are summarized in Table 1.
The onset time was not significantly prolonged when comparing the α-BTX groups to the control groups (both rocuronium and pancuronium). Offset times were significantly prolonged when comparing both the rocuronium and pancuronium α-BTX groups to the rocuronium and pancuronium control groups (P = 0.012 and 0.028, respectively).
Experimental autoimmune myasthenia gravis
The median concentration of rocuronium in the infused blood to obtain a 90% neuromuscular block was 5.09 μM (5.09-6.31 μM) in control rats vs. 4.64 μM (2.77-5.01 μM) in the autoimmune rats. For pancuronium the median concentration was 1.31 μM (1.25-1.77 μM) in control rats vs. 1.10 μM (0.78-1.15 μM) in the autoimmune rats. Onset and offset are summarized in Table 2.
The onset time was not significantly prolonged when comparing the control groups to the autoimmune groups (both rocuronium and pancuronium). Offset times were significantly prolonged when comparing both the rocuronium and pancuronium control groups to the rocuronium and pancuronium autoimmune groups (P = 0.024 for both groups).
The average AChR loss, measured in the tibialis anterior muscle of autoimmune rats compared to control muscles was 40% (±11%).
The results of this study in the rat model with a reduced AChR concentration indicate that the altered time course of effect of the increased sensitivity to neuromuscular blocking agents observed in myasthenia gravis is the result of the decreased receptor concentration. Furthermore, this rat model is a suitable model to study the time course of effects of neuromuscular blocking agents in a ‘myasthenic’ state.
The results from this study indicate that we did study a myasthenic-like situation: there was a prolonged offset time in all treated groups compared to controls. This prolonged offset time was significant for both rocuronium and pancuronium. We decreased the number of AChRs in two different ways. There was an acute model that used an α-BTX infusion, which decreases the number of available receptors for ACh binding at the neuromuscular junction by binding irreversibly to the AChRs. The other model consisted of the passive transfer of antibodies against the AChR, thus reducing the number of available binding sites for ACh by inducing an immune reaction with internalization of the receptors and destruction of the postsynaptic membrane  and by simply blocking the receptor . In human studies [1, 10-12] and in pigs , this prolonged offset time is also seen.
The potency was increased in the ‘myasthenic’ animals, as indicated by the lower concentration of neuromuscular blocking agent needed to obtain 90% block in ‘myasthenic’ animals, compared to control animals. This reflects the higher sensitivity of myasthenic animals to neuromuscular blocking agents, due to the decrease of functional AChRs. This phenomenon is also seen in humans [1,13,14].
The intervals 75-25% (i.e. time from 75 to 25% twitch height during onset of neuromuscular block) in this rat model were not different between controls and myasthenic rats. This can be understood because there are less receptors at the neuromuscular junction to be occupied, but at the same time, the administered amount of neuromuscular blocking agent is lower in myasthenic patients. In myasthenic patients, the onset time of neuromuscular blocking agents varies from a shorter onset time to a normal onset time [15-17]. In this model we did not expect to find significant differences in onset time, since there is no clear evidence of such in the literature.
The intervals 25-75% in the myasthenic animals are markedly prolonged compared to the healthy controls. This prolonged recovery is also seen in myasthenic patients for rocuronium [10,18] and other neuromuscular blocking agents like mivacurium [11,12] and vecuronium [15,19,20].
Both models had a similar change in time course of effect and potency of neuromuscular blocking agents as observed in myasthenic patients, indicating that the decrease in AChR concentration is the mechanism that causes the observed changes in myasthenic animals and man.
The prolonged offset time of muscle relaxants in case of a decreased AChR concentration can easily be understood if one takes a closer look at the neuromuscular junction. In a myasthenic situation, a relatively smaller fraction of the AChRs is occupied by the neuromuscular blocking agent. Due to the decrease of the safety margin in myasthenic rats, fewer relaxant molecules need to bind to the AChR in order to provide a 50% neuromuscular block, so less molecules will be present in the effect compartment at 50% block. In a simplified way, one could state that in order to return to the control twitch height, one out of x relaxant molecules need to be cleared from the effect compartment, and x has a smaller value in myasthenic rats than in normal rats, so a relatively larger number of molecules has to be cleared from the effect compartment in myasthenic rats, which accounts for the observed prolonged time course of action (see Fig. 1).
We chose to use both pancuronium and rocuronium for the experiments of this study, because we have already studied them in this model . Data on onset and offset of pancuronium and rocuronium in this model are known, and using the same relaxants permitted us to make a comparison between the situation with a decreased number of AChR and the normal/control animals. We found that offset times of both pancuronium and rocuronium were the same, in control animals as well as ‘myasthenic’ animals. This confirms our earlier statement that potency is no major determinant of offset time .
This model is a unique model to study the effects of neuromuscular blocking agents for several reasons.
Firstly, it allows the effects to be investigated on a single muscle. When studying animals (or humans) in vivo, it is impossible to stimulate (or to measure the effect on) only one muscle, there will always be some recruitment and it is not always clear what the influence of repetitive stimulation on this recruitment will be. When measuring one muscle in this model, the so called ‘staircase phenomena’ is not observed. It is also unknown how myasthenia gravis will affect different muscles.
Secondly, the influence of pharmacokinetics is excluded and the concentration of neuromuscular blocking agents in the blood perfusing the muscle can be controlled. By isolating the muscle from the rest of the body, the researchers control the flow of blood that supplies the muscle exactly. There is no (re)distribution or elimination.
Thirdly, the experimental environment can be kept constant. The rat is placed in a thermostatically controlled chamber and body temperature is kept at a constant temperature. There are neither sudden changes in temperature nor in humidity in this chamber.
Finally, the experimental conditions can be varied widely. It is possible to study the behaviour of muscle relaxants at different flow rates (which would be almost impossible to obtain in in vivo experiments), at extremes of temperatures, and with different amounts of oxygen in blood.
We want to stress that the two models described in this paper do not produce myasthenia gravis in rats resulting in a situation that is fully comparable to that in myasthenic patients. Therefore the model may not be suitable to study all aspects of myasthenia gravis. However, both models result in a reduction of the concentration of AChRs in the neuromuscular junction, similar to the situation in myasthenic patients. This allowed us to study the influence of the concentration of AChRs in the neuromuscular junction on the potency and time course of action of neuromuscular blocking agents. The present study confirms the results from our earlier studies in man and in pigs and gives further support to our hypothesis that the increased sensitivity to and prolonged duration of action of neuromuscular blocking agents in myasthenic patients can be explained by a reduction of the AChR concentration in the neuromuscular junction.
In conclusion, we have studied the time course of effects of neuromuscular blocking agents in our rat model. Both models of AChR reduction were equally suitable to alter the time course of effect and potency of neuromuscular blocking agents that is observed in humans in case of myasthenia gravis.
Support was provided solely from institutional and/or departmental sources. The authors wish to thank Barbie Machiels, PhD, from the Department of Psychiatry and Neuropsychology, Section Basal Neurosciences, University of Maastricht, The Netherlands, for laboratory assistance.
1. De Haes A, Proost JH, Kuks JB, van den Tol JC, Wierda JMKH. Pharmacokinetic/pharmacodynamic modeling of rocuronium
in myasthenic patients is improved by taking into account the number of unbound acetylcholine receptors. Anesth Analg
2. De Haes A, Proost JH, De Baets MH, Stassen MHW, Houwertjes MC, Wierda JMKH. Pharmacokinetic-pharmacodynamic modelling in case of a decreased number of acetylcholine receptors. A study in myasthenic pigs. Anesthesiology
3. De Haes A, Houwertjes MC, Proost JH, Wierda JMKH. An isolated, antegrade, perfused, peroneal nerve anterior tibialis muscle model in the rat
: a novel model developed to study the factors governing the time course of action of neuromuscular blocking agents. Anesthesiology
4. Tzartos S, Hochschwender S, Vasquez P, Lindstrom J. Passive transfer of experimental autoimmune myasthenia gravis by monoclonal antibodies to the main immunogenic region of the acetylcholine receptor. J Neuroimmunol
5. Hoedemaekers AC, Verschuuren JJ, Spaans F et al.
Age-related susceptibility to experimental autoimmune myasthenia gravis: immunological and electrophysiological aspects. Muscle Nerve
6. Hoedemaekers AC, Graus Y, van Breda Vriesman P, de Baets M. Age- and sex-related resistance to chronic experimental autoimmune myasthenia gravis (EAMG) in Brown Norway rats. Clin Exp Immunol
7. Lindstrom JM, Lennon VA, Seybold ME, Whittingham S. Experimental autoimmune myasthenia gravis and myasthenia gravis: biochemical and immunochemical aspects. Ann NY Acad Sci
8. Verschuuren JJ, Graus YM, Theunissen RO et al.
Role of acetylcholine receptor antibody complexes in muscle in experimental autoimmune myasthenia gravis. J Neuro-immunol
9. Jahn K, Franke C, Bufler J. Mechanism of block of nicotinic acetylcholine receptor channels by purified IgG from seropositive patients with myasthenia gravis. Neurology
10. Baraka A, Haroun-Bizri S, Kawas N, Hajjar AM, Kawkabani N. Rocuronium
in the myasthenic patient [Letter]. Anaesthesia
11. Paterson IG, Hood JR, Russell SH, Weston MD, Hirsch NP. Mivacurium in the myasthenic patient. Br J Anaesth
12. Seigne RD, Scott RP. Mivacurium chloride and myasthenia gravis. Br J Anaesth
13. Baraka A. Anaesthesia and myasthenia gravis. Can J Anaesth
14. Buzello W, Noeldge G, Krieg N, Brobmann GF. Vecuronium for muscle relaxation in patients with myasthenia gravis. Anesthesiology
15. Baraka A, Taha S, Yazbeck V, Rizkallah P. Vecuronium block in the myasthenic patient. Influence of anticholinesterase therapy. Anaesthesia
16. Baraka A. Onset
of neuromuscular block in myasthenic patients. Br J Anaesth
17. Huang YG, Luo AL, Luo LK. Evaluation of atracurium in myasthenic patients undergoing thymectomy. Proc Chin Acad Med Sci Peking Union Med Coll
18. Sanfilippo M, Fierro G, Cavalletti MV, Biancari F, Vilardi V. Rocuronium
in two myasthenic patients undergoing thymectomy. Acta Anaesthesiol Scand
19. Hunter JM, Bell CF, Florence AM, Jones RS, Utting JE. Vecuronium in the myasthenic patient. Anaesthesia
20. Nilsson E, Meretoja OA. Vecuronium dose-response and maintenance requirements in patients with myasthenia gravis. Anesthesiology