Although mivacurium is reported to be metabolized predominantly by plasma cholinesterase (pseudocholinesterase, acylcholine acylhydrolase, International Enzyme Commission number 184.108.40.206) , two investigations have revealed either no relationship  or low correlation  between the activity of this enzyme (within the normal range) and mivacurium infusion rates needed to maintain target twitch depression. In contrast, Lien et al.  reported that clearance of the two potent stereoisomers of mivacurium correlated with plasma cholinesterase activity, suggesting that activity of this enzyme should influence mivacurium infusion requirements. In each previous study of the relationship of plasma cholinesterase activity to mivacurium infusion rates, mivacurium was infused to maintain 90%-99% twitch depression. Because the dose-effect (concentration-effect) relationship for muscle relaxants flattens as effect exceeds 90%, the steady-state plasma concentration to maintain 99% twitch depression would greatly exceed that to maintain 90% twitch depression. Thus, mivacurium plasma concentrations (and consequently infusion rates) might vary markedly when 90%-99% twitch depression is targeted. Failure of previous investigators to demonstrate a strong relationship between mivacurium infusion rates and plasma cholinesterase activity was therefore not surprising, but we believed it important to ascertain whether such a relationship does in fact exist. To address this issue, we infused mivacurium at a minimum of two rates per subject, targeting both 50% and 90% twitch depression. The relationship between infusion rates and twitch depression was then fit by the Hill Equation toestimate the infusion rates that would have produced 50% and 90% twitch depression.
With approval of our local institutional review board, and after obtaining informed consent, we studied 14 ASA grade I and II patients, aged 18-58 yr, scheduled for elective surgery. Patients exceeding 130% of ideal body weight, those with renal, hepatic, neuromuscular, and/or electrolyte disorders, or those taking medications known to interfere with neuromuscular function were excluded.
After midazolam, 1-2 mg intravenously (IV), anesthesia was induced with fentanyl, 3-4 micro gram/kg IV, and propofol, 2-3 mg/kg IV. Tracheal intubation was performed without neuromuscular block  and ventilation was controlled to maintain end-tidal PCO2 at 30-35 mm Hg. Anesthesia was maintained with nitrous oxide, 70%, and isoflurane, 1% end-tidal (Datex Ultima, Helsinki, Finland). Electrocardiogram, SpO2, non-invasive blood pressure, and esophageal temperature were monitored continuously. Esophageal temperature was maintained between 35.5 and 37.0 degrees C. After induction of anesthesia, 5 mL of venous blood was obtained to determine plasma cholinesterase activity and dibucaine inhibition (SmithKline Beecham Clinical Laboratories, Van Nuys, CA) photometrically using acetylthiocholine as a substrate.
After induction of anesthesia, the ulnar nerve was stimulated via subcutaneous needle electrodes at the wrist. Supramaximal stimuli of 0.2 ms duration were delivered in a train-of-four at 2 Hz every 12 s (Digistim II; Neuro Technology, Houston, TX). Preload was maintained at 200-400 g. The evoked twitch tension of the adductor pollicis muscle was measured using a calibrated force transducer (Myotrace, Houston, TX) and amplified (DC Bridge Signal Conditioner; Gould Electronics, Valley View, OH). Twitch tension was digitized (NB-M10-16; National Instruments, Austin TX), displayed (LabView, National Instruments), and recorded on-line. In addition, a strip chart recorded the evoked twitch tension (TA240, Gould Electronics). End-tidal isoflurane concentration was stable for >20 min and the first twitch response of each train (T1) was stable for >10 min (the control twitch tension) prior to mivacurium administration.
Mivacurium was infused at 1.0 micro gram centered dot kg-1 centered dot min-1 until twitch tension stabilized for >10 min. Mivacurium infusion rate was adjusted (based on the Hill equation; see Appendix), targeting to maintain twitch tension at 50% of control. If twitch tension stabilized outside the range 47%-53% of control, the mivacurium infusion rate was adjusted so that two infusion rates, one producing <50% twitch depression, the other producing >50% twitch depression, were obtained. Steady-state infusion was defined as an infusion rate maintained for >15 min with twitch tension stable for >10 min. When steady state was attained, the mivacurium infusion was again adjusted, now targeting 90% (88%-92%) twitch depression. Typically, mivacurium was infused at 4-5 different rates for each subject; steady-state twitch depression was attained at two or three of these infusion rates.
For each subject, the Hill equation [Appendix, Equation 3] was fit to these values for mivacurium infusion rate and steady state twitch depression using an iterative least-squares nonlinear regression solver (Microsoft Excel Solver User's Guide, Version 3.0a. Redmond, Microsoft Corporation, 1991). We then determined the infusion rates expected to produce steady state 50% and 90% twitch depression (IR (50) and IR90, respectively) and the coefficient of the Hill equation, gamma. Values for IR50 and IR90 were estimated only if an infusion produced steady-state twitch depression in the target range or two or more infusions produced steady-state twitch depression bracketing the target range. For example, if infusions produced 45%, 55%, and 91% twitch depression, both IR50 and IR90 would be estimated. However, if infusions produced 51% and 67% twitch depression, IR90 would not be estimated. Values for twitch depression with mivacurium, 1 micro gram centered dot kg-1 centered dot min-1, IR50, IR90, and gamma were then compared to plasma cholinesterase activity by analysis of linear regression. Statistical significance was accepted when P < 0.05.
All subjects had normal values for plasma cholinesterase activity and dibucaine inhibition. For 11 subjects, three steady-state mivacurium infusions were obtained Figure 1; for the remaining three subjects, only two steady-state infusions were obtained. IR50 could be estimated for all subjects Figure 2 and IR90 for 13 (in one subject only two steady-state infusions were obtained with maximum twitch depression of 69%). With mivacurium, 1.0 micro gram centered dot kg-1 centered dot min-1, twitch depression decreased with increased activity of plasma cholinesterase (r2 = 0.43, P = 0.015; Figure 3). Both IR50 (r2 = 0.51, P < 0.005) and IR90 (r2 = 0.48, P < 0.01; Figure 4) correlated with plasma cholinesterase activity. The Hill coefficient, gamma, averaged 2.8 (SD = 0.4) and did not correlate with plasma cholinesterase activity (r2 < 0.01, P > 0.99).
We observed a strong relationship between plasma cholinesterase activity and the infusion rate to depress twitch tension 50% and 90%. However, we could explain only 48%-51% (based on values for r2) of the variability in infusion rates as a function of the variability in plasma cholinesterase activity. At steady state, the infusion rate of a muscle relaxant equals the product of its plasma clearance and the steady-state plasma concentration. Since neuromuscular junction sensitivity varies between patients, even if mivacurium's plasma clearance correlated exactly with plasma cholinesterase activity, infusion rates to produce specific degrees of neuromuscular block should not correlate exactly with plasma cholinesterase activity. Recently, Lien et al.  reported that plasma clearance of the two potent stereoisomers of mivacurium varies with plasma cholinesterase activity but that the values for r2 were only 0.32-0.33. Considering the additional variability in infusion rates as a function of neuromuscular junction sensitivity, it is surprising that we were able to demonstrate such a strong relationship between infusion rates and plasma cholinesterase activity.
In contrast to the strong relationship we observed between mivacurium infusion rates and plasma cholinesterase activity, Ali et al.  observed only a minimal relationship (r2 = 0.18) and Ostergaard et al. . reported no correlation. Whereas we targeted 50% and 90% twitch depression and calculated IR (50) and IR90, both Ali et al.  and Ostergaard et al.  maintained 90%-99% twitch depression. If the Hill Equation isvalid for mivacurium and its pharmacokinetics are first-order (and using the mean value for gamma determined in the present experiments), then the steady-state infusion rate to maintain 99% twitch depression exceeds that to maintain 90% twitch depression by a factor >2. In that plasma cholinesterase activity varied only two-to threefold in these patients, the variability in the target neuromuscular block is sufficient to explain the limited relationship other investigators have observed between plasma cholinesterase activity and infusion rate. In addition, Ali et al.  adjusted the infusion rate frequently and determined average values maintaining the target range of neuromuscular depression (whereas we maintained a constant infusion rate for sufficient periods to attain steady-state neuromuscular depression).
Savarese et al.  reported "poor correlation of the duration of action of bolus doses of mivacurium with the plasma cholinesterase activity of individual subjects." Duration of action of a bolus dose of a muscle relaxant is a complicated function of neuromuscular junction sensitivity, distribution volumes, and plasma clearance of the drug such that even twofold changes in plasma clearance of a muscle relaxant might not influence recovery from single bolus doses. Even if there were an exact correlation between plasma cholinesterase activity and mivacurium clearance, the range of plasma cholinesterase activities in Savarese et al.'s  patients is sufficiently small to limit the likelihood of their observing an effect of plasma cholinesterase on duration of action.
We assumed that mivacurium's plasma concentration was at steady state after 15 min of constant rate infusion. Mivacurium's two potent stereoisomers (cistrans and trans-trans) comprise >90% of the administered drug and, because of their rapid clearance by plasma cholinesterase, have elimination half-lives <2 min . Thus, we maintained the same infusion rate for a sufficient number of half-lives that plasma concentrations of these stereoisomers were presumably constant. The third stereoisomer of mivacurium, ciscis, has a smaller plasma clearance and a longer terminal half-life (52.9 +/- 19.8 min) compared to the other stereoisomers . Although the cis-cis isomer will cumulate during a constant infusion, it is unlikely to contribute significantly to twitch depression: it comprises only 4%-8% of administered drug  and in animals (although not studied in humans) its potency is 0.1 that of the other isomers . Although we have no information regarding the rate constant for equilibration between plasma concentrations and effect (Keo), the lack of changes in twitch depression during the 10-min period of stability suggests that effect compartment concentrations of the two potent stereoisomers of mivacurium were stable.
In summary, by targeting 50% and 90% twitch depression rather than >90% twitch depression and by estimating infusion rates needed to produce these degrees of neuromuscular depression, we demonstrated a strong relationship between the activity of plasma cholinesterase (for subjects with values in the normal range) and mivacurium infusion rates. This relationship is consistent with the belief that the major elimination pathway of mivacurium is hydrolysis by plasma cholinesterase . Although plasma cholinesterase values are not typically available for clinical purposes, these values, if known, might guide clinicians toward appropriate mivacurium infusion rates during clinical anesthesia. Our study does not examine whether subjects whose plasma cholinesterase values lie outside the normal range [e.g., subjects with liver disease  or those with genetic variants of plasma cholinesterase  in whom duration of mivacurium is prolonged] can have mivacurium dosing guided by plasma cholinesterase values.
Determination of the Infusion Rate Predicted to Produce 50% and 90% Twitch Depression
According to the Hill equation: Equation 1 where Ce is the mivacurium concentration in the effect compartment, Ce50 is the effect site concentration depressing twitch tension 50%, and gamma is the Hill coefficient that governs the sigmoidal relationship between concentration and effect. At steady state, Cp and Ce are proportional. Equation 1 can therefore be rewritten as: Equation 2 where Cp50 is the steady-state plasma concentration depressing twitch tension 50%. If the pharmacokinetics of mivacurium are first order, Cp at steady state is proportional to infusion rate and Equation 2 can be rewritten as: Equation 3 where IR50 is the steady-state infusion rate depressing twitch tension 50%. Equation 3 can be solved for IR50: Equation 4 During each study, because that patient's value for gamma is unknown, we assumed that gamma equals 3 (determined in pilot studies). Thus, we infused mivacurium at a rate estimated to produce 50% twitch depression [IR50 (predicted)]: Equation 5 Similarly, the infusion rate estimated to depress twitch tension 90% [IR90 (predicted)] was estimated as: Equation 6
1. Cook DR, Stiller RL, Weakly JN, et al. In vitro metabolism of mivacurium chloride (BW B1090U) and succinylcholine. Anesth Analg 1989;68:452-6.
2. Ostergaard D, Jensen FS, Jensen E, et al. Influence of plasma cholinesterase activity on recovery from mivacurium-induced neuromuscular blockade in phenotypically normal patients. Acta Anaesthesiol Scand 1992;36:702-6.
3. Ali HH, Savarese JJ, Embree PB, et al. Clinical pharmacology of mivacurium chloride (BW B1090U) infusion: comparison with vecuronium and atracurium. Br J Anaesth 1988;61:541-6.
4. Lien CA, Schmith VD, Embree PB, et al. The pharmacokinetics and pharmacodynamics of the stereoisomers of mivacurium in patients receiving nitrous oxide/opioid/barbiturate anesthesia. Anesthesiology 1994;80:1296-1302.
5. Scheller MS, Zornow MH, Saidman LJ. Tracheal intubation without the use of muscle relaxants: a technique using propofol and varying doses of alfentanil. Anesth Analg 1992;75:788-93.
6. Savarese JJ, Ali HH, Basta SJ, et al. The clinical neuromuscular pharmacology of mivacurium chloride (BW B1090U). A shortacting nondepolarizing ester neuromuscular blocking drug. Anesthesiology 1988;68:723-32.
7. Maehr RB, Belmont MR, Wray DL, et al. Autonomic and neuromuscular effects of mivacurium and its isomers in cats [abstract]. Anesthesiology 1991;75:A772.
8. Cook DR, Freeman JA, Lai AA, et al. Pharmacokinetics of mivacurium in normal patients and in those with hepatic or renal failure. Br J Anaesth 1992;69:580-5.
© 1995 International Anesthesia Research Society
9. Ostergaard D, Jensen FS, Jensen E, et al. Mivacurium-induced neuromuscular blockade in patients with atypical plasma cholinesterase. Acta Anaesthesiol Scand 1993;37:314-8.