Secondary Logo

Journal Logo

Original Articles – General

Pharmacodynamic modelling of rocuronium in adolescents with Duchenne muscular dystrophy

Ihmsen, Haralda,*; Viethen, Vanessaa,*; Forst, Juergenb; Schwilden, Helmuta; Schmitt, Hubert Ja; Muenster, Tinoa

Author Information
European Journal of Anaesthesiology: February 2009 - Volume 26 - Issue 2 - p 105-110
doi: 10.1097/EJA.0b013e32831aed11
  • Free



Duchenne muscular dystrophy (DMD) is the most common and progressive muscular dystrophy with an incidence of approximately 1 in 3500 male newborns. It is inherited in an X-linked manner, although up to 30% of the cases represent new mutations in the dystrophine gene located on chromosome Xp21. This results in a quantitative deficit of dystrophine, an important sarcolemmnal structure protein in muscle cells and in the brain. Although this role and the complete function are poorly understood at present, there is strong evidence that dystrophine and its related glycoprotein complex play an important role in stabilizing the skeleton of muscle fibres and in the formation of normal neuromuscular junctions during development and reorganization.

Histological signs in muscle biopsies are the variation of the fibre size, clusters of necrotic fibres with fatty infiltrations, central nuclei in regeneration and a progressive fibrosis that leads to the loss of ambulation at the average age of 10 years [1,2].

Previous studies in patients with DMD showed that onset time was significantly delayed and recovery index was markedly prolonged after application of various nondepolarizing neuromuscular blocking agents (NMBA) such as vecuronium, mivacurium and rocuronium [3–5]. It was further found that a reduction of dose did not speed up the recovery from neuromuscular block (NMB) [6]. The underlying mechanisms of these findings remain unclear. There may be structural changes in the nicotinic acetylcholine receptor or its signal transduction, but alterations in the pharmacokinetics of NMBA in patients with DMD were not excluded at this time.

The objective of the present study was to identify the pharmacokinetic and/or pharmacodynamic origin of the observed alterations in patients with advanced DMD.


Patients and anaesthesia

After approval of the local ethics committee and obtaining written informed consent, 25 male patients, aged between 10 and 18 years, were consecutively enrolled in this study. The 15 patients in the DMD group (ASA physical status III) had advanced stage of the disease and were wheelchair-bound. They were scheduled for corrective orthopaedic surgery of the spinal column by using an implant system (ISOLA-Spinal-Implant-System; Acro-Med, Cleveland, Ohio, USA). The diagnoses of DMD had been verified by muscle biopsy showing the absence of dystrophine. Nine of the patients were partly genotyped and showed a deletion in exons 46–51; six patients were not genetically tested. Ten male age-matched teenagers (control group, ASA I) without any neuromuscular disease served as controls. These patients underwent urological or lower limb surgery that required endotracheal intubation. None of the patients received any medication known to influence neuromuscular function. An echocardiography and a lung function test were performed several days before the planned surgery in all patients with DMD to evaluate cardiopulmonary risk. As patient characteristics, we recorded age, height, weight and BMI.

Patients were visited the day before surgery for a physical examination and review of laboratory test results. Patients were premedicated with 3.75 mg midazolam orally 45 min before anaesthesia. Standard intraoperative monitors were used, including electrocardiography, automatic blood pressure and pulse oximetry. After placing a peripheral intravenous line and preoxygenation with 100% oxygen, anaesthesia was induced with sufentanil 0.2–0.3 μg kg−1 and propofol 3 mg kg−1. Mask ventilation was secured and patients were intubated without any use of NMBA. In the DMD group an intra-arterial and a central venous catheter were placed for continuous pressure recordings. Anaesthesia was maintained with continuous intravenous infusion of propofol 8–12 mg kg−1 and sufentanil was titrated to effect. No volatile drugs were used. Lungs were ventilated with a mixture of oxygen in air; minute volume ventilation was set to obtain end-tidal carbon dioxide levels of 35–40 mmHg. Central body temperature (DMD group, bladder probe; control group, ear probe) was kept between 35.5 and 37°C using a warm forced air device and warmed fluid infusion.

Neuromuscular monitoring

Neuromuscular transmission was monitored by acceleromyography (AMG) using train-of-four (TOF) Watch SX equipment (Organon Ireland Ltd, a part of Schering-Plough Corporation, Dublin, Ireland) within the guidelines of the Copenhagen Consensus Conference [7]. For surgery, patients with DMD were set into a prone position, and the controls mostly stayed in a supine position. Following final positioning, the right forearm was prepared for AMG monitoring. The hand and the forearm were immobilized in a splint allowing free mobility of the thumb. Skin temperature was monitored and maintained above 32°C throughout the study. The monitoring arm was kept free from arterial and i.v. indwelling catheters, and from the blood pressure cuff.

The ulnar nerve was stimulated at the wrist via surface electrodes by supramaximal square wave impulses of 0.2 ms duration in a TOF sequence (four consecutive impulses with 2 Hz). These stimuli were delivered every 15 s throughout the investigation. Response of the adductor pollicis muscle was quantified using an AMG-probe fixed to the volar surface of the distal phalanx of the thumb. The TOF monitor was connected to a portable PC for online data recording and processing (TOF Watch SX monitor; Organon Ireland Ltd, a part of Schering-Plough Corporation, Dublin, Ireland). After calibration and signal stabilization of the control response, 0.3 mg kg−1 of rocuronium was administered intravenously over 5 s. The time from administration of rocuronium to complete clinical recovery was monitored. The following times of NMB were measured: maximal depression of the first twitch; time between rocuronium administration and the first change of TOF response (lag time); time from injection to the first of three consecutive twitches with the same or increasing amplitude (onset time) [7]; time from injection to maximum effect (Tmax); time between rocuronium administration and recovery of first twitch of the TOF response to 10, 25 and 90% (T10, T25, T90); time between 25 and 75% recovery of first twitch (recovery index); and time between 25% recovery of first twitch and recovery of TOF ratio to 90% (recovery time).

Blood sampling and drug analysis

In five patients with DMD, blood samples of 2.8 ml each were collected from an arterial catheter. One baseline sample was drawn before injection of rocuronium. Arterial samples were taken at 2, 4, 7, 10, 15, 20, 30, 45, 60, 120 and 240 min after bolus administration. The blood samples were collected in heparinized tubes and were immediately centrifuged at 4°C for 10 min to separate the blood plasma. Afterwards 1 ml of plasma was incubated with 200 μl of 1 mol l−1 NaH2PO4 and stored at −70°C.

Analysis of rocuronium plasma samples was performed using high-performance liquid chromatography (HPLC) and electrochemical detection. As internal standard, 100 μl of 70 μg ml−1 Org 7402 (N.V. Organon, a part of Schering-Plough Corporation, The Netherlands) was added to 1 ml plasma. The samples were prepared by solid-phase extraction with an Oasis HLB column (Waters, Milford, Massachusetts, USA). After elution with 1 ml methanol/citric acid (90/10 v/v), the extract was evaporated at 40°C and reconstituted with 100 μl methanol. An aliquot of 20 μl of the eluted extract was automatically injected into the analytical column (C18 Gravity 125/4, 3 μm; Machery-Nagel, Düren, Germany). The eluent (pH 5.4) consisted of 920 ml water, 225 ml acetonitrile, 25 ml methanol and 1 g sodium acetate. Electrochemical detection (Coulochem II; ESA Laboratories Inc., Chelmsford, Massachusetts, USA) was carried out at 850 mV. The intra-assay and interassay coefficients of variation were 5.9 and 2.7%, respectively. The lower limit of quantification for rocuronium was 2.5 ng ml−1.

Pharmacokinetic/pharmacodynamic modelling

For the DMD group, we performed a two-step nonlinear regression analysis with the software NONMEM (GloboMax LLC, Hanover, Maryland, USA). The measured rocuronium plasma concentrations of five patients in the DMD group were analysed with a two-compartment and three-compartment model and proportional residual error. Goodness of fit was assessed by plots of the residuals and by the NONMEM objective function. The three-compartment model was accepted as significantly better than the two-compartment model if the difference in the objective function between the two fits was at least 9.2 (P < 0.01, Chi-squared test with two degrees of freedom).

The data of the first twitch response were analysed with a sigmoid Emax model and a constant residual error:

where E is the predicted twitch response, CE is the effect site concentration, EC50 is the concentration that produces half-maximum effect and γ is a measure of curve steepness. The effect site concentration CE was calculated from the plasma concentration CP by dCE/dt = ke0(CPCE) with the rate constant ke0 as a measure for the hysteresis, that is, the delay between the plasma concentration and the effect. The equilibration half-time was calculated as T½ke0 = ln(2)/ke0. CP was calculated using the population estimates of the pharmacokinetic parameters obtained in the pharmacokinetic modelling of the patients with DMD. The dose for half-maximum effect ED50 was calculated as ED50 = D ⋅ EC50/CEmax from the half-maximum concentration EC50 and the maximum effect site concentration CEmax achieved with the administered dose of D = 0.3 mg kg−1.


As the pharmacokinetic/pharmacodynamic parameters were obtained by population analysis, they were tested for significant differences between the control group and the DMD group using the standard errors of the estimates. Two parameters were considered significantly different if their 95% confidence intervals, calculated as estimate ± 1.96 SE, did not overlap. Individually estimated variables such as onset time or recovery index were tested for statistical significance using the t-test or the Mann–Whitney U-test, if not normally distributed.


We studied 25 patients, 15 with DMD and 10 without neuromuscular disease. Patients did not differ with respect to age, height, weight or BMI (Table 1). Pharmacodynamic data (time course of neuromuscular response) were obtained from all patients. Plasma levels of rocuronium were determined in five out of the 15 patients with DMD.

Table 1
Table 1:
Patients' data


Compared with the control group, the onset of the neuromuscular blockade was significantly delayed and the recovery was significantly prolonged in patients with DMD, whereas the maximum effect was comparable (Table 2). It is worth noting that, particularly in the DMD group, there was a large difference between onset time (as defined by the Copenhagen Consensus Conference [7]) and time to maximum effect.

Table 2
Table 2:
Pharmacodynamics of neuromuscular blockade

Pharmacokinetic/pharmacodynamic modelling

Figure 1 shows the measured and predicted concentrations of rocuronium in the five patients with DMD with arterial blood samples. The time course could be adequately described by a two-compartment model; a three-compartment model was only slightly better with an improvement of 6.8 in the objective function (P = 0.04) and large standard errors for k13, k31 and the terminal half-life T1/2γ (Table 3). The three-compartment model was superior only for the late concentrations at 120 and 240 min, whereas the two-compartment model was more adequate at 20 and 30 min. As the pharmacodynamic data did not exceed 120 min, we used the two-compartment model for further pharmacodynamic analysis. The pharmacokinetic/pharmacodynamic model with the pharmacokinetic parameters obtained from the five patients with DMD adequately described the measured NMB in all patients with DMD (Fig. 2 and Table 4).

Fig. 1
Fig. 1
Table 3
Table 3:
Results of the pharmacokinetic analysis of rocuronium in five patients with Duchenne muscular dystrophy
Fig. 2
Fig. 2
Table 4
Table 4:
Results of the pharmacodynamic analysis



The present study confirmed our previous findings that the administration of rocuronium in patients with DMD led to a significantly prolonged onset and duration of NMB than controls [5,6]. The large difference between the onset of NMB and the time to maximum effect in children with DMD is remarkable. The onset time is defined by the Copenhagen Consensus Conference as the time point when the first twitch response is not becoming smaller in three consecutive recordings [7]. In cases of a slow onset and a flat downslope of the twitch response, the signal noise may complicate the determination of onset time by this method. Therefore, we determined the time to maximum effect to demonstrate that in DMD the onset is considerably prolonged compared with healthy individuals.

Pharmacokinetic/pharmacodynamic modelling

The pharmacokinetics of rocuronium in adolescents with DMD could be adequately described by a two-compartment model. A three-compartment model yielded a slightly better fit but with large standard errors for the terminal half-life and the transfer rate constants of the second peripheral compartment. A more precise determination of the three-compartment model would have required a longer time of arterial sampling, which was not considered because of ethical reasons. Although fluid shifts during surgery may generally influence pharmacokinetics, there were no relevant shifts within the first 2 h during which most of our samples were taken.

For the control group, a pharmacokinetic analysis was not reported because of the absence of arterial blood samples. In these patients, it would have been unethical to place an arterial catheter. As this study focuses on the pharmacokinetics and pharmacodynamics of rocuronium in patients with DMD, we decided to draw frequent arterial blood samples only in some patients of the DMD group and to abandon blood samples in the control group. Venous sampling would have been feasible, but it was emphasized in earlier studies that arteriovenous differences may lead to misspecifications in the pharmacodynamic model [8]. Particularly in the nonsteady-state situation after a bolus, the hysteresis may be underestimated if venous samples are used because venous concentrations usually show a lag time compared with arterial concentration. In order to provide a model-based comparison of the two groups, different approaches were tested. Pharmacodynamic fits of the control group data were performed using the pharmacokinetic parameters obtained in the analysis of the DMD group, the pharmacokinetic parameters reported by Wierda et al.[9] for children aged 2–8 years and the pharmacokinetic parameters reported by Wierda et al.[10] for adults. The model with the smallest value of the NONMEM objective function was considered the best. These approaches revealed that the model using the pharmacokinetic parameters obtained from the DMD group also yielded the best results in the control group, followed by the fit with the pharmacokinetic parameters for adults reported by Wierda et al.[10], whereas the fit with the pharmacokinetic parameters reported by Wierda et al.[9] for children was less appropriate (Table 4 and Fig. 3). The delayed onset and prolonged recovery in the DMD group is reflected in the significantly smaller equilibration rate constant ke0. In addition, the half-maximum effect site concentration EC50 was found to be significantly reduced in the DMD group, whereas the dose for half-maximum effect ED50 and the steepness γ of the concentration–effect relation did not differ. This explains why the maximum effect was comparable in both groups. Although the estimates of ke0 and EC50 in the control group were dependent on the pharmacokinetic parameter set used, T1/2ke0 was in any case significantly prolonged in the DMD group, whereas EC50 was significantly smaller in patients with DMD, irrespective of the assumed pharmacokinetic model.

Fig. 3
Fig. 3

When comparing the results of this study with the data for rocuronium in the literature [9–19], the pharmacokinetics of rocuronium in the patients with DMD were more similar to the reported pharmacokinetics of rocuronium in healthy adults than to the reported pharmacokinetics in children (Table 5), although the small number of patients in the present study may limit the significance of these results. The estimated pharmacodynamic parameters for the control group were also in fair agreement with the data for adults reported in the literature. These findings may be explained by the mean age of 13 years in our study population, whereas the children in the cited studies were between 2 and 10 years. The results of the pharmacokinetic analysis in the DMD group together with the finding that the time course of NMB in the control group could be well described using the pharmacokinetic parameters of the DMD group give at least some indication that only pharmacodynamics but not pharmacokinetics of rocuronium are altered in patients with DMD. The physiological mechanisms for the pharmacodynamic alterations in patients with DMD still remain unclear. Possible explanations are structural changes in the nicotinic acetylcholine receptor or its signal transduction.

Table 5
Table 5:
Pharmacokinetics and pharmacodynamics of rocuronium in subjects without Duchenne muscular dystrophy as reported in the literature

In conclusion, the pharmacodynamics but not the pharmacokinetics of rocuronium seemed to be altered in adolescent patients with DMD. The applied pharmacokinetic/pharmacodynamic analysis also showed that there were two alterations of rocuronium pharmacodynamics in patients with DMD: prolonged equilibration between the central and effect site compartment, and decreased EC50, whereas the ED50 did not differ.


The authors thank N.V. Organon, a part of Schering-Plough Corporation, for providing Org 7402 for HPLC analysis of plasma levels of rocuronium.

The authors thank Rainer Knoll for his support in HPLC analysis of rocuronium.

The authors thank all members of the Department of Anaesthesia at the Waldkrankenhaus St. Marien for technical assistance and discussions.


1 Blake DJ, Kroger S. The neurobiology of Duchenne muscular dystrophy: learning lessons from muscle? Trends Neurosci 2000; 23:92–99.
2 Emery AE. Population frequencies of inherited neuromuscular diseases: a world survey. Neuromuscul Disord 1991; 1:19–29.
3 Ririe DG, Shapiro F, Sethna NF. The response of patients with Duchenne's muscular dystrophy to neuromuscular blockade with vecuronium. Anesthesiology 1998; 88:351–354.
4 Schmidt J, Muenster T, Wick S, et al. Onset and duration of mivacurium-induced neuromuscular block in patients with Duchenne muscular dystrophy. Br J Anaesth 2005; 95:769–772.
5 Wick S, Muenster T, Schmidt J, et al. Onset and duration of rocuronium-induced neuromuscular blockade in patients with Duchenne muscular dystrophy. Anesthesiology 2005; 102:915–919.
6 Muenster T, Schmidt J, Wick S, et al. Rocuronium 0.3 mg × kg−1 (ED95) induces a normal peak effect but an altered time course of neuromuscular block in patients with Duchenne's muscular dystrophy. Paediatr Anaesth 2006; 16:840–845.
7 Viby-Mogensen J, Engbaek J, Eriksson LI, et al. Good clinical research practice (GCRP) in pharmacodynamic studies of neuromuscular blocking agents. Acta Anaesthesiol Scand 1996; 40:59–74.
8 Tuk B, Danhof M, Mandema JW. The impact of arteriovenous concentration differences on pharmacodynamic parameter estimates. J Pharmacokinet Biopharm 1997; 25:39–62.
9 Wierda JM, Meretoja OA, Taivainen T, Proost JH. Pharmacokinetics and pharmacokinetic: dynamic modelling of rocuronium in infants and children. Br J Anaesth 1997; 78:690–695.
10 Wierda JM, Kleef UW, Lambalk LM, et al. The pharmacodynamics and pharmacokinetics of Org 9426, a new nondepolarizing neuromuscular blocking agent, in patients anaesthetized with nitrous oxide, halothane and fentanyl. Can J Anaesth 1991; 38:430–435.
11 Bock M, Klippel K, Nitsche B, et al. Rocuronium potency and recovery characteristics during steady-state desflurane, sevoflurane, isoflurane or propofol anaesthesia. Br J Anaesth 2000; 84:43–47.
12 Dragne A, Varin F, Plaud B, Donati F. Rocuronium pharmacokinetic–pharmacodynamic relationship under stable propofol or isoflurane anesthesia. Can J Anaesth 2002; 49:353–360.
13 Kopman AF, Chin WA, Moe J. Dose–response relationship of rocuronium: a comparison of electromyographic vs. acceleromyographic-derived values. Acta Anaesthesiol Scand 2005; 49:323–327.
14 Kumar N, Mirakhur RK, Symington MJ, McCarthy GJ. Potency and time course of action of rocuronium during desflurane and isoflurane anaesthesia. Br J Anaesth 1996; 77:488–491.
15 Saldien V, Vermeyen KM, Wuyts FL. Target-controlled infusion of rocuronium in infants, children, and adults: a comparison of the pharmacokinetic and pharmacodynamic relationship. Anesth Analg 2003; 97:44–49.
16 Taivainen T, Meretoja OA, Erkola O, et al. Rocuronium in infants, children and adults during balanced anaesthesia. Paediatr Anaesth 1996; 6:271–275.
17 Woloszczuk-Gebicka B, Wyska E, Grabowski T, et al. Pharmacokinetic–pharmacodynamic relationship of rocuronium under stable nitrous oxide-fentanyl or nitrous oxide-sevoflurane anesthesia in children. Paediatr Anaesth 2006; 16:761–768.
18 Woolf RL, Crawford MW, Choo SM. Dose–response of rocuronium bromide in children anesthetized with propofol: a comparison with succinylcholine. Anesthesiology 1997; 87:1368–1372.
19 Wulf H, Ledowski T, Linstedt U, et al. Neuromuscular blocking effects of rocuronium during desflurane, isoflurane, and sevoflurane anaesthesia. Can J Anaesth 1998; 45:526–532.

Duchenne muscular dystrophy; neuromuscular block; neuromuscular disease; pharmacodynamics; pharmacokinetics; pharmacology; rocuronium

© 2009 European Society of Anaesthesiology