To the Editor:--The studies of Hart et al. [1]and Szenohradsky et al. [2]raise a number of issues regarding antagonism of mivacurium blockade. The following comments may be helpful.

The above studies were conducted using the original technique of Miller and Cronnelly, [3–5]in which an infusion of relaxant is given to maintain a steady-state level of paralysis. An antagonist is given without changing the rate of relaxant infusion to compare ease of antagonism of various relaxants and/or potency and duration of effect of the antagonists. The technique may have been constructed originally to obtain purely pharmacodynamic measurements, in the absence of any kinetic factors. The problem with the model is its questionable clinical relevance, especially with respect to its application to mivacurium.

Kinetic factors are of prime importance in any evaluation of antagonism of residual nondepolarizing blockade. What clinicians need to know is how much antagonist is necessary to restore normal function within a period of time pertinent to practice. In this respect, the studies of Hart et al. [1]and Szenohradsky et al. [2]may be misleading. Both studies would seem more pertinent if the infusion of mivacurium had been discontinued and then the antagonist given. The then-observed speed of reversal would provide data that would be useful to practitioners.

Because the active isomers of mivacurium have half-lives of less than 2 min, [6]the timing of reversal of mivacurium infusions is key. If the clinician simply waits 2 min after discontinuation of a mivacurium infusion, the plasma concentration will have decreased to approximately 50% of levels present beforehand, thereby making antagonism more facile. (By analogy, no anesthesiologists have been taught to antagonize block by any other relaxant soon after giving a "topping-up" dose.)

Reversal of all relaxants, including mivacurium, should be viewed as a summation of two processes: the clearance of the relaxant plus the pharmacodynamics of antagonism. In the case of mivacurium, the high rate of clearance due to plasma cholinesterase hydrolysis likely contributes especially to a rapid antagonism [7–10]of residual blockade. In any discussion of the influence of various factors on the kinetics or dynamics of mivacurium, including reversal, the major consideration is the possible effects of the interventions on plasma cholinesterase activity. This includes the anticholinesterases given for reversal. Both neostigmine and edrophonium do partially inhibit plasma cholinesterase. [7,8]They presumably do not stop the hydrolysis of mivacurium, however, which is why antagonism is relatively rapid.

Keeping the latter point in mind, explanations may be given for several observations or situations. For example, the increase in plasma concentrations of all mivacurium isomers found by Hart et al. [1]after edrophonium injection during infusion of mivacurium may be explained by an inhibition of plasma cholinesterase. [7].

Reduced plasma cholinesterase activity resulting from genetic or environmental factors likely will result in slower reversal of mivacurium in any patient when such a situation exists, because, in such a case, the large contribution of mivacurium clearance to the pharmacodynamics of reversal will be decreased. We suspect that, in the undocumented cases speculated on by Hart et al. [1]in which "incomplete or delayed" reversal of mivacurium by edrophonium was observed, an undiscovered reduction of plasma cholinesterase activity may have been contributory; in any event, other classic causes of slow or delayed reversal should be sought, such as temperature or electrolyte disturbances and concomitantly administered drugs. This commentary is especially pertinent to the rare patients who are homozygotes for the atypical form of plasma cholinesterase. Reversal of mivacurium in these people, though effective, may be relatively slow (as in the case of any long-lasting relaxant), because the contribution of drug (mivacurium) clearance to the rate of reversal is probably relatively minor in these individuals. [11,12]Because the plasma cholinesterase in these people is rather inactive to begin with, the administration of anticholinesterases to them is moot.

As Beemer et al. showed, [13]the testing of antagonism of any relaxant at depths of block greater than 90% twitch inhibition is probably inappropriate, because the probability of antagonism to train-of-four > 70% is less than 10%. Hart et al. [1]base one of the premises of their study on an abstract by Kao et al.* in which mivacurium antagonism was evaluated at 99% blockade of the electromyographic twitch:(1) Kao et al. do not state whether mivacurium infusion was discontinued;(2) At this depth of block, no twitch would be ordinarily recordable by mechanomyography nor palpable by clinical observation.

A final comment on the structure activity requirements of plasma cholinesterase hydrolysis should help clear up another element of apparent confusion: Mivacurium is a true choline ester, wherein complete hydrolysis yields a dibasic acid and two quaternary amino alcohols; this feature makes the molecule a substrate for the enzyme. Other quaternary esters, such as atracurium, do not yield acids and quaternary amino alcohols on hydrolysis of the ester linkage and are therefore not substrates. Consequently, Hart et al. [1]seem to be way out on a limb in speculating on metabolism of mivacurium by "other esterases or cholinesterases." As they have shown, [14]mivacurium infusion dosage requirements correlate well with plasma cholinesterase activity. The half-life of mivacurium in plasma in vitro and in vivo is less than 2 min, [6]indicating not only that metabolism and clearance occur largely in plasma but also that other possible processes cannot constitute rate-limiting factors.

In summary, in deciding whether or how to "reverse" (i.e., to accelerate recovery from) residual mivacurium blockade, clinicians should use commonly accepted rules of practice, as with other nondepolarizers: Administration should be stopped and spontaneous recovery should be allowed to proceed to a point when two or more twitches on TOF stimulation are palpable to ensure effective antagonism by anticholinesterases. Clinicians also may apply new predictive information (such as the 5–25% twitch recovery interval [15]) to better guide administration of mivacurium and to help predict whether an antagonist will be necessary to accelerate recovery. If the choice of antagonism is made, the infusion of mivacurium should simply be discontinued, the line flushed, a sign of beginning recovery observed on TOF monitoring, and the antagonist given. In 90% or more of cases, however, the patient may be offered rapid spontaneous recovery from mivacurium, providing an added option and a safety factor that does not exist in practice with other relaxants.

John J. Savarese, M.D., Cynthia A. Lien, M.D., Matthew R. Belmont, M.D., Department of Anesthesiology, The New York Hospital-Cornell Medical Center, 525 East 68th Street, A-1050, New York, New York 10021.

*Kao YJ, Le N, Barker SJ: Neostigmine prolongs profound neuromuscular blockade induced by mivacurium in surgical patients (abstract). ANESTHESIOLOGY 1993; 79:A929.


Hart PS, Wright PMC, Brown R, Lau M, Sharma M, Miller RD, Gruenke L, Fisher DM: Edrophonium increases mivacurium concentrations during constant mivacurium infusion, and large doses minimally antagonize paralysis. ANESTHESIOLOGY 1995; 82:912-8.
Szenohradsky J, Lau M, Brown R, Sharma ML, Fisher DM: The effect of neostigmine on twitch tension and muscle relaxant concentration during infusion of mivacurium or vecuronium. ANESTHESIOLOGY 1995; 83:83-7.
Miller RD, Van Nyhuis LS, Eger EI II, Vitez TS, Way WL: Comparative times to peak effect and durations of action of neostigmine and pyridostigmine. ANESTHESIOLOGY 1974; 41:27-33.
Cronnelly R, Morris RB, Miller RD: Edrophonium: Duration of action and atropine requirement in humans during halothane anesthesia. ANESTHESIOLOGY 1982; 57:261-6.
Miller RD, Larson CP Jr, Way WL: Comparative antagonism of d-tubocurarine-, gallamine-, and pancuronium-induced neuromuscular blockades by neostigmine. ANESTHESIOLOGY 1972; 37:503-9.
Lien CA, Schmith VD, Embree PB, Belmont MR, Wargin WA, Savarese JJ: The pharmacokinetics and pharmacodynamics of the stereoisomers of mivacurium in patients receiving nitrous oxide-opioid-barbiturate anesthesia. ANESTHESIOLOGY 1994; 80:1296-1302.
Lien CA, Belmont MR, Wray Roth D, Okamoto M, Abalos A: Pharmacokinetics and dynamics of mivacurium during spontaneous recovery and anticholinesterase-facilitated recovery (abstract). ANESTHESIOLOGY 1995; 83:A896.
Decic A, Munshi CA, Gandhi SK, Kampine JP: Antagonism of mivacurium neuromuscular block: Neostigmine versus edrophonium. Anesth Analg 1995; 81:1005-9.
Savarese JJ, Ali HH, Basta SJ, Embree PB, Scott RPF, Sunder N, Weakly JN, Wastila WB, El-Sayad HA: The clinical neuromuscular pharmacology of mivacurium chloride (BW B1090U), a short-acting nondepolarizing ester neuromuscular blocking drug. ANESTHESIOLOGY 1988; 68:723-32.
Naguib M, Abdulatif M, Al-Ghamdi A, Hamo I, Nouheid R: Dose-response relationships for edrophonium and neostigmine antagonism of mivacurium-induced neuromuscular block. Br J Anaesth 1993; 71:709-14.
Ostergaard D, Jensen FS, Jensen E, Skovgaard LT, Viby-Mogensen J: Mivacurium-induced neuromuscular blockade in patients with atypical plasma cholinesterase. Acta Anaesthesiol Scand 1993; 37:314-8.
Goudsouzian NG, d'Hollander AA, Viby Mogensen J: Prolonged neuromuscular block from mivacurium in two patients with cholinesterase deficiency. Anesth Analg 1993; 77:183-5.
Beemer GH, Goonetilleke PH, Bjorksten AR: The maximum depth of an atracurium neuromuscular block antagonized by edrophonium to effect adequate recovery. ANESTHESIOLOGY 1995; 82:852-8.
Hart PS, McCarthy GC, Brown R, Lau M, Fisher DM: The effect of plasma cholinesterase activity on mivacurium infusion rates. Anesth Analg 1995; 80:760-3.
Savarese JJ, Lien CA, Belmont MR: The accuracy of 5%-25% (T sub 1 -T sub 3) twitch recovery interval in predicting the speed of spontaneous recovery from mivacurium-induced neuromuscular blockade (letter). Anesth Analg 1995; 80:209-10.