Acta Anaesthesiol Scand 2002; 46: 684–691 Copyright C Acta Anaesthesiol Scand 2002Printed in Denmark. All rights reserved

ACTA ANAESTHESIOLOGICA SCANDINAVICA

0001-5172

Pharmacokinetics and pharmacodynamics ofmivacurium in young adult and elderly patients

D. ØSTERGAARD1, J. VIBY-MOGENSEN2, N. A. PEDERSEN1, H. HOLM1 and L. T. SKOVGAARD3

1Department of Anaesthesia, Gentofte University Hospital, 2Department of Anaesthesia, Rigshospitalet, 3Department of Biostatistics, University ofCopenhagen, Denmark

Background: Mivacurium is hydrolyzed by plasma cholin-esterase, and is therefore less dependent on liver metabolismand renal elimination than other neuromuscular blocking drugs.This might favor the use of mivacurium in elderly patients. Thepurpose of this study was to compare the pharmacodynamicsand the pharmacokinetics of the three isomers of mivacuriumand their metabolites in young adult and elderly patients.Methods: Sixty-four patients were included in a dose–re-sponse study, in which 32 young adults and 32 elderly patientsreceived one of four doses of mivacurium. An additional bolusdose of mivacurium to a total of 0.1mg/kg was given followedby a continuous infusion adjusted to maintain a 91–99% neuro-muscular block. The times to maximum block and differentlevels of recovery were measured using mechanomyographyand train-of-four (TOF) nerve stimulation. Thirty-two patientswere randomly selected for the pharmacokinetic study. Venoussamples were taken for determination of the three mivacuriumisomers and the metabolites.Results: The estimated ED95 were 0.053 and 0.061mg/kg inyoung adults and elderly patients, respectively (NS). The me-dian infusion rate did not differ, but duration to a TOF ratio of

SEVERAL papers have described the pharmacodyn-amics of mivacurium following bolus doses and

continuous infusions. Most of these studies were per-formed in young adults (1–5), with only a few inelderly patients (6–8). The pharmacokinetics of miva-curium has been described in young healthy adults(9, 10), whereas pharmacokinetic data of elderly pa-tients are sparse (11). As mivacurium is rapidly hy-drolyzed by plasma cholinesterase (pChe) (12) it isless dependent on liver metabolism and renal elimin-ation than some other non-depolarizing neuromuscu-lar blocking drugs. This might favor the use of mivac-urium in elderly patients who may have a variabledecrease in their glomerular filtration rate. On theother hand, lean body mass and total body water de-crease with age, both of which may affect the pharma-cokinetics of mivacurium. So far only one pharmaco-kinetic study of mivacurium has been performed inelderly patients, which indicated a reduced clearance

684

0.7 was significantly longer in elderly patients than in youngadult patients (21.0 vs. 16.5min). No statistically significant dif-ference between the age groups in clearance and eliminationhalf-life of the isomers was seen. The half-lives of the metabo-lites were significantly prolonged in the elderly patients.Conclusion: There were no significant differences in the po-tency or infusion requirements between the adult and elderlypatients, but the rate of recovery was significantly, though onlymoderately prolonged, in the elderly patients. No significant dif-ference in clearance was seen but the elimination half-lives ofthe metabolites was longer in the elderly patients.

Received 9 August 2001, accepted for publication 8 February 2002

Key words: age factors; butyrylcholinesterase; cholinesterase;dose–response curves; enzymes; metabolites; mivacurium;neuromuscular relaxants; pharmacodynamics; pharmaco-kinetics; pharmacology; pseudocholinesterase; stereoisomers.

c Acta Anaesthesiologica Scandinavica 46 (2002)

of the two active isomers of mivacurium (11). Nostudies of mivacurium metabolites’ pharmacokineticshave been reported in elderly patients.

The purpose of the present study was to evaluateboth the pharmacodynamics of mivacurium and thepharmacokinetics of the three mivacurium isomersand their metabolites, in young adult and elderly pa-tients.

Patients and methods

Seventy patients, ASA physical status 1–2, scheduledfor elective ear or nose operations entered the study.Thirty-five were young adult patients (18–40 years)and 35 were elderly patients (age 65 years). Beforeinduction of anesthesia a blood sample was taken,and according to the biochemical analysis all patientshad normal pChe phenotype and low to normal pCheactivity (13). Females of childbearing potential and

Mivacurium in adults and elderly patients

patients with a history of neuromuscular, cardio-vascular, renal or hepatic disorders were excludedfrom the study, as were patients receiving drugs thatmight affect the neuromuscular transmission. All pa-tients gave their informed consent, and the local Eth-ics Committee approved the study.

Premedication consisted of diazepam 0.1–0.2 mg/kgorally. Anesthesia was induced with fentanyl 1–4 mg/kg, diazepam 0.05–0.15 mg/kg and thiopentone 3–6mg/kg, and maintained with 66% nitrous oxide inoxygen and supplementary doses of fentanyl and di-azepam.

Perioperative monitoring included electrocardio-gram, pulseoximetry, capnography and non-invasiveblood pressure every minute for the first 3 min follow-ing mivacurium administration and every 5 minthereafter. The blood pressure cuff and the intra-venous line used for drug and fluid administrationwere on the same arm. Ventilation was adjusted tomaintain normocapnia (end tidal CO2 4.5–5.6 kPa).The rectal and peripheral skin temperatures weremeasured and maintained at greater than 35æC and32æC, respectively (14). After the induction of anes-thesia, a second intravenous line was inserted in thearm for blood sampling (used for monitoring) and athird line in the foot for mivacurium infusion.

The mechanical twitch was recorded using a Myo-graph 2000 (Biometer International, Denmark). The ul-nar nerve was stimulated at the wrist using surfaceelectrodes and 1-Hz single twitch stimulation (14).When supramaximal stimulation was achieved and theresponse to stimulation was stable for 5 min the stimu-lation pattern was changed to train-of-four (TOF).

PharmacodynamicsDose–response studyAccording to the protocol 32 young adult and 32elderly patients were allocated to receive one of fourdifferent doses of mivacurium: 0.04, 0.06, 0.08, or 0.09mg/kg. After including six patients receiving the 0.08and the 0.09 mg/kg doses it became clear, however,that these doses resulted in a complete response toTOF nerve stimulation abolition. The doses weretherefore reduced to 0.03 and 0.05 mg/kg, respec-tively. The six patients who had received the 0.08 and0.09 mg/kg doses were excluded from the dose–re-sponse study and replaced by another six randomlyselected patients.

Continuous infusion studyIn 65 patients an additional bolus dose of mivacuriumwas given up to a total of 0.1 mg/kg, when maximumblock was achieved following the initial dose (equal

685

twitch height of the first twitch (T1) in two TOFtrains). Tracheal intubation was performed at a maxi-mum T1 depression following this second bolus dose.Following the spontaneous recovery of the T1 5–10%of start control value, a continuous infusion of mivac-urium was started at an initial rate of 6 mg/kg/min. Itwas subsequently adjusted to maintain a neuro-muscular block within the range of 91–99% T1 de-pression. At the end of the surgical procedure theneuromuscular block was allowed to recover spon-taneously.

Neuromuscular monitoringThe maximum T1 depression following the first doseof mivacurium and all recovery data were calculatedusing start control values (14). Monitoring was con-tinued until both 90% T1 recovery and a TOF ratio of0.70 were obtained. The time from the second mivacu-rium dose injection to the first response to TOF andduration to 10, 25 and 90% T1 recovery to a TOF ratioof 0.70 were determined. The interval 25–75% (timefrom 25 to 75% T1 recovery) after the end of the in-fusion were measured (14).

PharmacokineticsSixteen young adult and 16 elderly patients with anexpected anesthetic duration of more than 90 minwere randomly selected for the pharmacokineticstudy. Later, three patients (two young adult and oneelderly) were excluded because of insufficient bloodsampling. In six young adult and in three elderly pa-tients the plasma concentration data for the cis-transisomer during the infusion were close to or below thelower limit of quantification of the plasma assay;these data were excluded.

Plasma concentrations of mivacuriumVenous blood samples (5 ml) were collected immedi-ately before mivacurium administration, 1.2 and 5 minfollowing the initial dose, and before and at 1, 2, 5, 10and 20 min following the second bolus dose. Bloodsamples were also collected immediately before andat 1, 2, 5, 10, 15, 20 and 30 min after the start of themivacurium infusion, and at 30 min intervals duringthe infusion. If the infusion rate was changed, bloodsamples were collected immediately before and at 1,2 and 5 min following the change. Three samples werealso taken at 5 min intervals before the infusion wasterminated. Additional samples were collected im-mediately before and at 1, 2, 5, 10, 15, 30, 45, 60, 120,180, 240 and 360 min after the infusion was termin-ated. In less than 10 s the blood samples were trans-ferred into a vacutainer containing a cholinesterase in-

D. Østergaard et al.

hibitor (phospholine iodide). The samples were centri-fuged and the plasma was decented and frozen atª70æC. The ratio of the cis-cis, cis-trans and trans-transisomers in the clinical trial material used was approxi-mately 5.8, 35.5 and 58.8%, respectively (data are fromthe certificate of analysis, Glaxo Wellcome). The con-centration of each of the three isomers were deter-mined by a stereospecific high performance liquidchromatographic method with fluorometric detectionand a stepped gradient (by the Department of Bioana-lysis and Drug Metabolism, Glaxo Wellcome ResearchLaboratories, Beckingham, UK). The drug assay wasautomated (ASPEC, Gilson). The coefficient of vari-ation was 12% and the lowest level of quantificationwas 2.5 ng/ml. Calibration was linear over the range5–1000 ng/ml (15).

Pharmacokinetic dataThe plasma concentrations of the individual isomerswere used to construct plasma concentration vs. timeprofile of the drug. Non-compartmental pharmaco-kinetic parameters were determined for individualpatient assay data.

Cmax (the maximum concentration) was taken di-rectly from the plasma concentrations. AUC 0-t (areaunder the plasma concentration vs. time curve fromzero to the last measurable concentration at time t) forboth isomers and metabolites were calculated by thelinear trapezoidal method, taking the plasma concen-tration at the time of injection to be zero. The elimin-ation rate constant (l) was estimated by log-linear re-gression for the terminal slope of the concentration vs.time profile. The value of l was used to extrapolateconcentrations beyond the last measurable concen-tration: Ct to infinite time. The elimination half-life(t1/2l Ω ln2/l) and the AUC0-≤ (area under the plasmaconcentration curve vs. time curve from zero to timeinfinity (AUC0-≤Ω AUC0-t π Ct/l) for the isomers andmetabolites, the total plasma clearance (CL Ω dose/AUC0-≤), the clearance at steady state (Clss Ω rate ofinfusion/concentration at steady state) and the vol-ume of distribution (VDl Ω CL/l) for the isomerswere calculated.

Statistical analysesTo determine mivacurium’s potency, the T1 de-pressions (%) were transformed into probits and plot-ted against the logarithm of the mivacurium dose. Thedose–response relationship was determined separ-ately for the two age groups by linear regressionanalysis. Probit values of 0% and 100% T1 depressionwere replaced by 0.5% and 99.5% values, respectively(14).

686

Summary statistics (mean, SD), median, range and95%CI were used to describe the neuromuscular data.The efficacy data were reasonably normally distrib-uted. Comparisons between age groups were basedon t-tests with a pooled variance estimate. PΩ0.05 wasused to define a statistically significant difference. Anoverall mean infusion rate was calculated for each agegroup from the average infusion rates of the individ-ual patients. Pearson’s correlation coefficient wasused to evaluate whether a linear relationship existedbetween the infusion rate and the pChe activity.

Results

Table 1 shows the demographic and biochemical data.All patients were within 20% of their ideal bodyweight (14). No statistically significant differences inweight, height or pChe activity were seen between thegroups.

Estimated creatinine clearances were significantlylower in the elderly than in the young adults: 62 (17)vs. 121 (24) ml/min (mean, SD).

Pharmacodynamic dataDose–response studyThere was no statistically significant difference in thepotency of mivacurium in the young adult andelderly patients. The estimated ED50 and ED95 were0.030 mg/kg and 0.053 mg/kg, and 0.033 mg/kg and0.061 mg/kg, in young adult and elderly patients, re-spectively. In Fig. 1 the composite dose–response linesand the confidence limits are given.

Continuous infusion studyThere was no statistically significant difference in dur-ation of infusion or mean infusion requirement in thetwo groups of patients (Table 2). A stable 91–99% T1

depression was obtained after the first four 5-minperiods both in the young adult and elderly patients,with a median infusion rate of 6.0 mg/kg/min in bothgroups of patients for the first 105 min. Only a fewpatients received an infusion for more than 120 min.A correlation was found between the pChe activityand the infusion rate of mivacurium (rΩ0.43;P0.001). Spontaneous recovery from the neuro-muscular block was significantly prolonged in theelderly patients compared with the young adult pa-tients (Table 2). All patients were able to maintainheadlift for 5 s before leaving the operation theaterand again at discharge from the recovery room.

An adverse reaction possibly related to mivacuriumwas only seen in one patient, who had had two brief

Mivacurium in adults and elderly patients

Table1

Demographic and biochemical data in young adult (18–40years) and elderly patients (65years).

Group Number of Males/ Age Weight Height Plasma cholinesterase Dibucainepatients Females years kg cm activity U/l n

Young adults 35 9/26 31 (18–40) 74.8 (53–105) 178 (160–192) 799 (386–1221) 84 (82–85)Elderly 35 17/18 72 (65–83) 70.4 (48–95) 171 (154–190) 807 (447–1734) 85 (84–86)

Medians and ranges are given. Reference values for plasma cholinesterase activity and for the dibucaine number are 660–1620 U/L and 79–87 L, respectively (13).

Table2

Infusion requirements, duration of action and recovery data in young adult (18–40years) and elderly patients (65years).

Group Duration of Median infusion Duration Duration TOF Interval75% infusion rate 0.70 25–75%10% 25% 90%

min mg/kg/min

Young adults 86 6.0 3.9 6.0 16.7 16.5 7.2(25–227) (4.6–10.6) (0–12.5) (1.4–16.7) (9.4–31.9) (9.5–31.5) (3.6–14.0)

Elderly 95 6.0 5.0 8.5 21.6 21.0 10.6(28–207) (3.8–10.6) (0–15.9) (0–20.9)* (9.3–43.9)* (8.4–38.5)* (6.2–20.0)*

Median (ranges) and number of patients (n) are given. Data are presented according to the good clinical research practice (GCRP) rules forpharmacodynamic studies in neuromuscular blocking agents (14).*Significant difference in recovery from the neuromuscular block between the two groups (P0.05).TOF, train-of-four ratio.

episodes of mild bronchospasme, which were success-fully treated with terbutaline.

Pharmacokinetic studyDuring the continuous infusion of mivacurium a newsteady state plasma concentration was achieved forthe cis-trans and the trans-trans isomers within 10 minof infusion rate changes. Contrary to this, the plasmaconcentration of the cis-cis isomer never reachedsteady state. Therefore, no steady state clearance forthis isomer could be estimated. No significant differ-ence in the concentration of the cis-cis isomer at theend of the infusion was found between the youngadult and elderly patients: 70 and 60 ng/ml, respec-tively. Table 3 summarizes the pharmacokinetic dataof the three isomers. No statistical significant differ-ence in Cmax and AUC was seen in the young adult andelderly patients, although there was a trend towardshigher Cmax values in the young adult patients. Nostatistical significant difference in the clearances of thethree isomers was seen in the young adult and elderlypatients. There was a statistically significant corre-lation between the pChe activity and the clearance ofthe cis-trans ( rΩ0.42; PΩ0.04) and trans-trans (rΩ0.50;PΩ0.006) isomers, but not of the cis-cis isomer (rΩ0.12;PΩ0.54)Table 3.

There was a trend towards a longer eliminationhalf-life in the elderly for all isomers, but this was not

687

statistically significant. The 95%CI for the cis-cis, cis-trans and the trans-trans isomer was 0.77–1.57, 0.72–2.04 and 0.92–2.86, respectively; indicating that theelimination half-life could have been from 57–186%longer in the elderly patients.

All isomers showed small volumes of distributionduring the elimination phase. The data is, however, se-verely affected by the poor characterization of the ter-minal phase in some patients and by high estimates forplasma clearances. The higher VD estimates for theelderly patients were only statistically significant forthe trans-trans isomer, most probably because of thelarge variances found for the other isomers.

The plasma concentration of the cis quarternary alco-hol metabolite was too low to allow for reliable esti-mates of the half-life or AUC data. For the three othermetabolites, the plasma concentration increased dur-ing the infusion period and Cmax occurred near the endof the infusion (Table 4). For all metabolites, Cmax wasdependent on the duration of infusion. No significantdifferences were seen in Cmax between the young adultand elderly patients. The plasma concentrations of thethree major metabolites were measurable for 6 h with abi-exponential decline following the termination of themivacurium infusion. The elimination half-lives for thethree major metabolites were prolonged by approxi-mately 30% in the elderly patients. This difference wasstatistically significant for the two monoester metabo-

D. Østergaard et al.

Table3

Estimated maximum concentration (Cmax), area under curve (AUC0–≤), steady state concentration (Css), steady state clearance CLss, totalclearance (CL), elimination half-life (t1/2l) and volume of distribution during the terminal elimination phase (VDl) of each isomer in 14 youngadult (18–40years) and 15 elderly patients (65years).

Isomer Cmax AUC0 – ≤ Css Clss Cl t1/2l VDl

ug/ml ng/ml hour ng/ml ml/kg/min ml/kg/min min l/kg

Cis-cisYoung Adults 101 191 – – 4.4 56 0.31

(77–170) (51–321) (3.0–5.5) (35–75) (0.24–0.54)Elderly 76 152 – – 4.5 65 0.37

(42–144) (38–435) (2.7–9.8) (15–131) (0.20–0.96)

Cis-transYoung adults 192 42 25 92 94 2.1 0.32

(67–390) (12–75) (10–43) (53–190) (51–144) (1.1–6.1) (0.12–1.17)Elderly 107 28 17 133 133 2.6 0.50

(27–518) (5–131) (9–54) (57–264) (47–167) (1.2–9.5) (0.10–2.55)

Trans-transYoung adults 448 124 67 53 51 2.7 0.19

(179–791) (34–196) (32–114) (32–121) (30–98) (0.4–9.0) (0.06–0.49)Elderly 308 82 46 77 67 4.3 0.44*

(83–1069) (44–364) (34–148) (33–141) (29–135) (1.4–9.5) (0.10–1.10)

Medians and ranges are given.*Statistically significant difference, P0.05.

lites (cis and trans 879) (Table 4). The relative exposureto each metabolite per unit dose was estimated by di-viding the AUC by the total dose. The average relativeexposure was 20% higher in the elderly than in theyoung adult patients. This difference was statisticallysignificant for all metabolites.

Discussion

The primary pharmacodynamic findings are thatthere was no difference in potency or infusion require-ments of mivacurium in the young adult and elderly

Fig.1. Scatterplot of the probit transformed twitch depression and the logarithm of the dose in young adult patients (A) and elderly patients (B).The composite dose–response lines (slope 6.427 and 7.098) and the confidence intervals are given.

688

patients, but that the elderly patients had a signifi-cantly slower rate of spontaneous recovery of neuro-muscular block. Pharmacokinetically, the clearancesand the elimination half-lives did not differ signifi-cantly in the two age groups, but the volume of distri-bution increased in the elderly patients. Also, theelimination half-lives of the metabolites were signifi-cantly longer in the elderly patients.

PharmacodynamicsWe used the single dose method to determine thedose–response relationship (the potency) of mivacuri-

Mivacurium in adults and elderly patients

um because a cumulative dose technique tends tounderestimate the potency of a short-acting drug suchas mivacurium (16). The ED95 found by us in youngadults are in good agreement with the results of pre-vious studies in adult patients using TOF stimulation,as in this study (3, 4). We found no statistically sig-nificant difference in potency between the two agegroups; neither did Valinthout et al. (7). Our estimatedED95 in both the young adult and elderly patients(0.053 and 0.061 mg/kg, respectively) is, however,lower than those found by Valinthout et al. (0.091 and0.096 mg/kg, respectively) (7). These differencesmight be caused by differences in methodologies.

The mean infusion rate (6 mg/kg/min) required tomaintain a 91–99% neuromuscular block did not dif-fer between the young adult and elderly patients forthe first 105 min. Only a few of the elderly patientsreceived a longer infusion. The infusion requirementfound in young adults is comparable to data fromother studies in adults during narcotic anesthesia (3,4, 6, 16), but slightly higher than reported by Goud-souzian et al. (8). This can be explained by differencesin methodology as Goudsouzian et al. used EMG. Ser-vin et al. (11), Dahaba et al. (17) and Goudsouzianet al. (8) all found lower infusion requirements inelderly compared with young adult patients, and alsofound that the necessary infusion rate seemed to de-crease with time. The duration of infusion was, how-ever, longer (4 h) (8, 17). A correlation was found be-tween the pChe activity and the infusion rate of miva-

Table4

Estimated maximum concentration (Cmax)), elimination half-lives(t1/2l) and area under curve AUC/dose of each mivacurium metabo-lite in 14 young adult (18–40years) and 15 elderly patients (65years).

Cmax t1/2lmin AUC/doseng/ml ml/h

Cis monoesterYoung adults 435 87 1660

(194–564) (63–103) (1247–2368)Elderly 424 107* 2026*

(209–839) (37–174) (1384–3464)Trans monoesterYoung adults 683 58 2112

(323–1051) (42–83) (1679–3381)Elderly 666 78* 2661*

(339–1240) (33–134) (1596–4365)Trans alcoholYoung adults 617 44 1800

(250–1752) (28–95) (943–3237)Elderly 599 61 2208*

(318–1215) (25–107) (1368–3288)

Medians and ranges are given.*Statistically significant difference between groups, P0.05.

689

curium both in the young adult and elderly patients;the higher the pChe activity the higher the infusionrate. This is in accordance with previous studies (2,18).

The times to different levels of T1 recovery in theyoung adult patients were comparable to those re-ported previously (9). The spontaneous recovery timewas significantly prolonged in the elderly comparedwith the young adult patients. This is in agreementwith the findings of Maddineni et al. (6) and Servinet al. (11). Whether or not a difference of 5–10 min inrecovery time is clinically relevant is, however, opento discussion. Plasma cholinesterase activity may de-crease with age (13). The prolonged recovery time inthe elderly cannot, however, be explained by differ-ences in pChe activity because we did not find anystatistically significant difference in pChe activity be-tween the two age groups.

PharmacokineticsIn elderly patients there was a trend towards lowerCmax values. One might speculate if the time to Cmax

was longer than 2 min in some of the elderly patientsand Cmax actually occurred within the 3-min periodwhere no blood samples were drawn. This would af-fect the area under the curve, and hence the clearanceof mivacurium in elderly patients would have beenoverestimated.

Total and steady state clearanceThe clearances we found in the young adult patientsare in agreement with data from studies using thesame technique (i.v. sampling) (9, 19, 20) and some-what higher than in studies using arterial sampling(10, 11). The differences might be explained by the dif-ferent sampling techniques. We used venous sam-pling and only limited sampling immediately after theinjection of mivacurium. Hence, the initial part of thearea under the plasma concentration curve could havebeen underestimated, and consequently the plasmaclearances and the volume of distributions are over-estimated. Lacroix et al. (10) reported the clearances ofthe trans-trans and the cis-trans to be overestimated by66 and 90%, respectively, whereas that of the cis-cisisomer was overestimated by only 10%. The import-ance of arterial sampling is of major importance fol-lowing a bolus dose and of less importance during acontinuous infusion.

No statistically significant difference in the clear-ance of the three isomers was seen between the twogroups of patients. Because of the large variation inclearances found between the individual patients wecannot exclude a type 2 error. Elderly patients often

D. Østergaard et al.

have a longer circulation time than do young adults,and this might cause an even greater decrease in thearea under the first moment curve and an even largeroverestimation of the plasma clearance and the vol-ume of distribution of the individual isomers of miva-curium. Our data is in contrast to Servin et al’s (11)findings of a 50% reduction in clearance of the activeisomers in elderly patients. The reduction in clearancedid not, however, cause a markedly impaired elimin-ation because of a simultaneous decrease in the distri-bution volume at steady state.

In accordance with previous studies (9, 19, 20) wefound a relationship between the pChe activity andthe clearances of the active isomers. For the cis-cis iso-mer there was only a weak correlation with pChe ac-tivity, and most probably the clearance of this isomeris predominantly renal (21).

Elimination half-lifeThe elimination half-lives in young adults found inour study were consistent with the findings of Lienet al. and Head Rapson et al. (9, 19, 20).

There was no statistically significant trend towardshigher elimination half-life values of the three isomersin the elderly patients. However, as the data variedlargely between the patients and for the cis-trans iso-mer it was difficult to obtain a sufficient number ofsamples before the lower plasma assay limit of quanti-fication was reached.

Volume of distributionThe VDl observed in the young adult patients is com-parable with previously reported data (9). In elderlypatients, the volume of distribution during the ter-minal phase (VDl) for the trans-trans isomer was sig-nificantly increased. Also, there was a tendency to-wards a higher VDl for the cis-trans and the cis-cisisomers. The volume of distribution for many drugs,for example cisatracurium, is known to increase inelderly patients (22, 23). The increased volume of miv-acurium distribution in the elderly is in contrast withthe findings of Servin et al. (11), who reported a de-creased Vdss in elderly patients.

MetabolitesThe plasma concentrations of all metabolites in-creased during the infusion. No significant differencein Cmax was found between the two age groups. How-ever, the elimination half-lives for the three majormetabolites were 20–40% longer in the elderly pa-tients. This might be a result of a lower renal clearancein the elderly patients. We found shorter eliminationhalf-lives of the trans monoester and trans alcohol in

690

the young adult patients than reported, following asingle dose of mivacurium (10). A statistical signifi-cant higher metabolite exposure in the elderly groupwas seen, when calculated as the AUC dose per unitdose. This is probably the result of prolonged half-lifein the elderly group. The metabolites, however, havelow pharmacological activity, and the higher metabo-lite exposure in the elderly patients probably has noclinical importance (24).

Relationship between pharmacodynamic andpharmacokinetic dataWe found a statistically significant longer duration ofaction and recovery time of mivacurium in the elderlypatients, whereas no difference in potency or infusionrequirements was seen. Most likely, the prolonged re-covery time in the elderly patients is a result of alteredpharmacokinetics of mivacurium, although we wereunable to detect a difference in elimination half-livesfor the three isomers of mivacurium. An increasedelimination half-life reflects a longer stay of mivacuri-um in the body, and hence a longer duration of action.This is in accordance with an increased volume of dis-tribution for the trans-trans isomer. A significantly in-creased elimination half-life of the cis- and transmonoester metabolite was seen. However, this isprobably of no clinical importance because metabo-lites have low pharmacological activity (24).

In a combined pharmacodynamic and pharmaco-kinetic study, Servin et al. (11) also found a prolongedduration of action in elderly patients, as well as a de-creased clearance of active isomers. Despite the de-crease in clearance, the elimination of mivacuriumwas not markedly modified because of a simul-taneous decrease in the distribution volume at steadystate.

Conclusion

The potency and mean infusion rate required to main-tain a stable level of neuromuscular block for at least105 min are similar in young adult and elderly pa-tients. Recovery from neuromuscular block is signifi-cant, though only moderately prolonged in elderlypatients. These pharmacodynamic findings cannotreadily be explained by our pharmacokinetic findings,but are most probably the result of longer eliminationisomer half-lives. The clearance and the eliminationhalf-lives of the three mivacurium isomers are similarin young adult and elderly patients, but the elimin-ation half-lives of the cis- and trans-monoester meta-bolites are longer in elderly patients.

Mivacurium in adults and elderly patients

AcknowledgementsThe study was supported by a grant from Glaxo Wellcome. Thepharmacokinetic analysis was performed by Glaxo Wellcome,UK, and the authors thank Graham Ridout, Glaxo Wellcome,UK for support and advice concerning the pharmacokineticanalysis. Also we thank Bodil Mathiesen, Rigshospitalet, for per-forming the pChe analysis.

References1. Savarese JJ, Ali HH, Basta SJ, Embree PB, Scott RPF, Sunder

N et al. The clinical neuromuscular pharmacology of mivac-urium chloride (BW B1090U). Anesthesiology 1988: 68: 723–732.

2. Ali HH, Savarese JJ, Embree PB, Basta SJ, Stout RG, BottrosLH et al.. Clinical pharmacology of mivacurium chloride(BW B1090U) infusion: comparison with vecuronum andatracurium. Br J Anaesth 1988: 61: 541–546.

3. Caldwell JE, Kitts JB, Heier T, Fahey MR, Lynam DP, MillerRD. The dose–response relationship of mivacurium chloridein humans during nitrous oxide- narcotic anesthesia in adultsurgical patients. Anesthesiology 1989: 70: 31–35.

4. Diefenbach C, Mellinghoff H, Lynch J, Buzello W. Mivacuri-um: dose–response relationship and administration by re-peated injection or infusion. Anesth Analg 1992: 74: 420–423.

5. Østergaard D, Jensen FS, Jensen E, Skovgaard LT, Viby Mog-ensen J. Influence of plasma cholinesterase activity on recov-ery from mivacurium-induced neuromuscular blockade inphenotypically normal patients. Acta Anaesthesiol Scand1992: 36: 702–706.

6. Maddineni VR, Mirakhur RK, McCoy EP, Sharpe TD.Neuromuscular and haemodynamic effects of mivacuriumin elderly and young patients. Br J Anaesth 1994: 73: 608–612.

7. Vanlinthout LEH, Booij LDHJ, van Egmond J, Robertson EN.Age related differences in magnitude and complete recoveryof mivacurium induced neuromuscular block. Anesthesiology1995: 83: A897.

8. Goudsouzian N, Chakravorti S, Denman W, Schwartz A,Yang HS, Cook DR. Prolonged mivacurium infusion inyoung and elderly adults. Can J Anesth 1997: 44: 955–962.

9. Lien CA, Schmith VD, Embree PB, Belmont MR, WarginWA, Savarese JJ. The pharmacokinetics and pharmacodyn-amics of the stereoisomers of mivacurium in patients re-ceiving nitrous oxide/opioid/barbiturate anesthesia. Anes-thesiology 1994: 80: 1296–1302.

10. Lacroix M, Donati F, Varin F. Pharmacokinetics of mivacuri-um isomers and their metabolites in healthy volunteers afterintravenous bolus administration. Anesthesiology 1997: 86:322–330.

11. Servin F, Lavaut E, Peytavin G, Farinotti R, Desmonts JM. Apharmacokinetic and dynamic study of mivacurium in-fusion in elderly patients. Anesthesiology 1997: 87: A855.

12. Cook DR, Stiller RL, Weakly JN, Chakravorti S, BrandomBW, Welch RM. In vitro metabolism of mivacurium chloride

691

(BW B1090U) and succinylcholine. Anesth Analg 1989: 68:452–456.

13. Jensen FS, Skovgaard LT, Viby-Mogensen J. Identification ofhuman plasma cholinesterase variants in 6688 individualsusing biochemical analysis. Acta Anaesthesiol Scand 1995: 39:157–162.

14. Viby-Mogensen J, Engbæk J, Eriksson LI, Gramstad L, Jens-en E, Jensen FS, Koscielniak-Nielsen Z, Skovgaard LT, Øster-gaard D. Good clinical research practice (GCRP) in pharma-codynamic studies of neuromuscular blocking agents. ActaAnaesthesiol Scand 1996: 40: 59–74.

15. Viby-Mogensen J, Østergaard D, Donah F, Fisher D, HunterJ, Kampmann JP, Kopman A, Proost JH, Rasmussen SN,Skovgaard LT, Varin F, Wright PMC. Pharmacokineticstudies of neuromuscular blocking agents: Good Clinical Re-search Practice (GCRP). Acta Anaesthesiol Scand 2000: 44:1168–1180.

16. Ørding H, Skovgaard LT, Engbæk J, Viby-Mogensen. Dose–response curves for vecuronium during halothane and neu-rolept anaesthesia: Single bolus versus cumulative method.Acta Anaesthesiol Scand 1985: 29: 121–124.

17. Dahaba AA, Rehak PH, List WF. A comparison of mivacuri-um infusion requirements between young and elderly adultpatients. Eur J Anaesthesiol 1996: 13: 43–48.

18. Hart PS, McCarthy GJ, Brown R, Lau M, Fisher D. The effectof plasma cholinesterase activity on mivacurium infusionrates. Anesth Analg 1995: 80: 760–763.

19. Head-Rapson AG, Devlin JC, Parker CJR, Hunter JM.Pharmacokinetics of the three isomers of mivacurium andpharmacodynamics of the chiral mixture in hepatic cir-rhosis. Br J Anaesth 1994: 73: 613–618.

20. Head-Rapson AG, Devlin JC, Parker CJR, Hunter JM.Pharmacokinetics and pharmacodynamics of the three iso-mers of mivacurium in health, in end-stage renal failure andin patients with impaired renal function. Br J Anaesth 1995:75: 31–36.

21. Parker CJR, Hunter JM. The pharmacokinetics of mivacuri-um anaesthetic. Pharmacol Rev 1995: 3: 168–175.

22. Ornstein E, Lien CA, Matteo MD, Ostapkovich ND, Diaz J,Wolf KB. Pharmacodynamics and pharmacokinetics of cisa-tracurium in geriatric surgical patients. Anesthesiology 1996:84: 520–525.

23. Sorooshian SS, Stafford MA, Eastwood NB, Boyd AH, HullCJ, Wright PMC. Pharmacokinetics and pharmacodynamicsof cisatracurium in young and elderly patients. Anesthesi-ology 1996: 84: 1083–1091.

24. Belmont MR, Rubin LA, Lien CA, Tjan J, Savarese JJ. Mivac-urium anaesthetic. Anaesthetic Pharmacol Rev 1995: 3: 156–167.

Address:Doris ØstergaardDepartment of AnesthesiologyHerlev University Hospital2730 HerlevDenmarke-mail: dooe/herlevhosp.kbhamt.dk