When the priming principle (PP) was first introduced by Mehta et al. (1) and Schwarz et al. (2), they suggested that the onset of neuromuscular block could be markedly accelerated if an intubating dose of relaxant was preceded by a priming dose (PD) administered a few minutes earlier. Schwarz et al. suggested a PD of 0.015 mg/kg vecuronium (approximately 30% of the 95% effective dose [ED95]). Mehta et al. offered a PD dose of 0.015 mg/kg pancuronium (approximately 25% of the ED95). As articles describing the PP multiplied, no consensus regarding the correct PD or priming interval was achieved, but many authors concluded that the PD should approximate a dose 10% of the intubating dose (20% of the ED95). In an editorial that attacked these protocols as potentially unsafe, Donati (3) argued that the maximal safe PD of a nondepolarizing relaxant should be equivalent to 10% of the ED95, not a dose twice as large. Despite Donati’s warning, authors continue to recommend PDs as large as 30% of a drug’s ED95(4,5). By using current pharmacodynamic theory and knowledge of drug variability, we thought it might be instructive to estimate the possible consequences associated with different priming or precurarizing doses of nondepolarizing neuromuscular relaxants.
Because the dose-response curves of all nondepolarizing drugs are essentially parallel (6), the peak neuromuscular block induced by a given fraction of an ED95 dose should be essentially the same for all relaxants. The time to manifest this effect, however, will vary depending on a drug’s speed of onset. Unfortunately, the time to maximal effect of 10%–30% of ED95 doses is not only difficult to measure, but also cannot be extrapolated from larger doses. The difference in time to peak effect between a dose ≥1× ED95 and a precurarizing or PD is best appreciated with computer simulation. We analyzed this difference for a representative drug (vecuronium) for which we have what we believe to be a realistic and reliable model.
It is widely recognized that in any normal population sample, there is wide variability in the response to a given dose of a neuromuscular blocking drug. For most nondepolarizing relaxants, it seems that the sd about the ED50 approximates 20%–25% of the mean value (coefficient of variation, 20%–25%) (7–9). Thus, 1 patient in 50 (−2.05 sd) may have an ED50 only half of the commonly cited value.
A straight line can be defined if the slope and the coordinates of a single point are supplied. Thus, the dose-response relationship of a drug is described if the ED50 and slope of the log dose/logit plot are provided. If these variables are known, the neuromuscular effect of any dose of relaxant may be predicted from the Hill equation.
Taking published values for rocuronium as a representative example, we calculated the effect of administering 10%, 20%, or 30% of the ED95 dose on the indirectly evoked response of the adductor pollicis muscle in a population normally distributed with respect to drug sensitivity.
Our computer simulation for onset time uses the equations of Sheiner et al. (10) and uses pharmacokinetic variables derived from the work of Sohn et al. (11) and Shanks (12) (Table 1). With this model, the calculated ED95 of vecuronium is 0.049 mg/kg, with an ED50 of 0.027 mg/kg. The model predicts a time to 90% of peak effect after a single ED95 dose of 3.20 min, very similar to the value (3.35 min) we have previously reported (13). The model predicts a clinical duration (bolus to 25% T1 recovery) after a 2× ED95 bolus of 39 min and a 25% to 75% recovery interval of 13 min. We believe that these pharmacodynamic predictions are a realistic representation of the potency, onset time, and duration of action of vecuronium in an average subject. We used this model to determine the time to achieve 90% of peak effect after a worst-case (30% of an ED95, 0.0146 mg/kg) PD of vecuronium.
Our results appear in Table 2. A dose of rocuronium equivalent to 10% of the usually cited ED95 will rarely produce a measurable neuromuscular effect. As this dose is increased, the potential for inducing clinical weakness rapidly escalates. In 1 in 50 individuals, the usual recommendation of 10% of the intubation dose (2× ED95) will produce measurable neuromuscular depression.
Because the slopes of the dose-response relationship for all nondepolarizing neuromuscular blocking drugs are essentially parallel (6), we believe that our calculations and conclusions based on rocuronium can be more generally applied. Thus, the results in Table 2 should be identical for all nondepolarizing relaxants if the doses are expressed in ED50 equivalents rather than in milligrams per kilogram.
After an ED95 dose of vecuronium (Fig. 1), 90% of the peak effect (86% T1 depression) was seen in 3.2 min; maximal effect (95% T1 depression) took 7.5 min. Thirty percent of an ED95 produces not 29% block (0.30 × 95%), but <5% twitch depression. As with any subparalyzing dose, its maximal effect also takes 7.5 min to develop. However, the time from drug administration to 90% of the peak effect after this dose was 5.9 min, rather than 3.2 min.
We think that the term “intubation dose” (implying 2× ED95) is imprecise and potentially misleading. For example, because of its slow onset, the commonly advocated intubation dose of cisatracurium is 0.15 to 0.20 mg/kg (14). Because the ED95 of cisatracurium approximates 0.05 mg/kg, 10% of this drug’s intubation dose constitutes 30%–40% of the ED95(15). Even 10% of the smaller recommendation would result in 25% twitch depression in 1 patient in 10. The clinical consequences of twitch depression of this magnitude have not been investigated. The relationship between twitch height and the train-of-four ratio during onset and offset of block are quite different. Little fade may be seen during onset, despite profound neuromuscular block (16). In awake volunteers receiving a mivacurium infusion titrated to a train-of-four ratio of 0.70, problems with swallowing, grip strength, vision, and masseter strength were all quite apparent, although twitch height never decreased by more than 10% from control values (17).
In many circumstances, unexpected sensitivity to a precurarizing or PD of relaxant will result in nothing worse than an unpleasant experience for the patient (heavy eyelids, blurred vision, and difficulty in swallowing). These responses are not uncommon. Howardy-Hansen et al. (18) reported these findings in 70% of subjects receiving 20% of an ED95 dose of either pancuronium or gallamine. However, the potential for more serious adverse consequences is real. Engbaek et al. (19) describe a healthy awake volunteer given vecuronium 0.005 mg/kg (10% of the usual ED95). This dose resulted in 75% twitch depression and caused the patient to be unable to swallow. This subject obviously represents an extreme outlier. All the same, if this individual had been at risk for aspiration and had received this dose before a rapid-sequence intubation protocol, the consequences could have been catastrophic.
The variability in response noted previously has another corollary. Just as some individuals will be quite sensitive to nondepolarizing blockers, other will show resistance. Those individuals with an ED50 1 or 2 sds more than the mean may show little discernible effect from a usual and customary dose. In these individuals, a defasciculating dose of relaxant (10% of the ED95) or even a larger PD (20% of the ED95) of a nondepolarizing relaxant will produce less than its intended effect. The safety and dependability of the PP is thus very much subject to the laws of probability.
The previous discussion makes an assumption that may not be true. It postulates that the priming interval is sufficiently long to allow the peak effect of the PD to manifest itself. This raises a question: what is the optimal priming interval? This is probably most easily studied with computer simulation. It is generally assumed that for less than fully paralyzing doses of neuromuscular blockers, the time to peak effect is independent of dose (20,21). This assertion, although true, is misleading. The time to maximal effect for an amount ≤;1× ED95 is dose independent, the time to 90% of peak effect is not. Neuromuscular blocking activity is not linearly related to drug concentration, but it is governed by the Hill equation.
The consequences of this are illustrated in Figure 1. After a bolus of vecuronium, the peak concentration in the effect compartment will occur in 7.5 minutes regardless of dose. After a 1× ED95 dose of vecuronium, 90% of the drug’s peak effect is seen in slightly more than three minutes. However, when the dose is reduced to 30% of the ED95, it takes almost six minutes for the drug to achieve 90% of its maximal effect. Although we have previously reported that the time to peak effect after a 1× ED95 dose of vecuronium was 5.5 minutes (13), we believe that this computer simulation probably gives a better approximation of what is actually transpiring at the effect site. Clinically it is very difficult to differentiate small differences in the depth of neuromuscular block from background “noise.” The above analysis has important implications for the anesthetist who wants to use the PP to accelerate the onset of neuromuscular block. Let us assume that a clinician wants to initiate laryngoscopy within 60 seconds of giving an intubating dose of vecuronium. This second, and larger, dose of relaxant should logically be given one minute before the PD achieves 90% of its full effect, i.e., five minutes after the PD. Earlier attempts at intubation will not take full advantage of the initial dose of relaxant.
The previous simulation demonstrates that when small fractions of an ED95 are administered, the time to 90% of peak effect is only slightly shorter than the time to maximal T1 depression. We have not attempted to simulate the optimal priming interval for rocuronium or cisatracurium because we have not found a pharmacokinetic/pharmacodynamic model of these drugs that in our experience is a reliable and realistic predictor of clinical onset/offset. Nevertheless, the optimal priming interval will obviously depend on a drug’s intrinsic onset profile. After a single ED95 dose, we previously reported that the time for peak effect for rocuronium was 4.1 minutes, and for cisatracurium it was 7.0 minutes (13). Thus, an intubating dose of cisatracurium should be preceded by a priming interval of at least six minutes, but as little as three minutes may be required for rocuronium.
The results of our analysis strongly support Donati’s suggestion (3) that PDs or precurarizing doses of nondepolarizing relaxants should not exceed 10% of the drug’s ED95. It is probably fortuitous that the usual precurarizing dose of d-tubocurarine (3 mg, or approximately 0.05 mg/kg) represents almost exactly 10% of the ED95(22). Even in subjects −2.3 sd from the mean (first percentile), twitch height would remain at 98.5% of control with this dose. Once this mass of drug is exceeded, the potential for misadventure increases rapidly. In a normal population of healthy patients, sensitivity to the neuromuscular effects of these drugs is very variable. One patient in 20 will exhibit a measurable decrease in twitch height with the usual PD of 20% of the ED95 (10% of the intubation dose). If this dose is increased by 50%, this segment of the population will experience profound weakness (twitch depression of almost 40%) from the PD.
It may be argued that this analysis fails to account for a drug’s onset time. Because the majority of clinicians probably do not wait more than three minutes after a PD before inducing anesthesia, most drugs (rocuronium is the exception) will not have reached their maximal effect at this time. This is true, but we think it misses the point. If an anesthetist is not going to wait for the PD to have its full effect, we would ask, why not just give a slightly larger initial bolus and not prime at all? Rocuronium in doses not exceeding 2× ED95 avoids the potential problems outlined previously and still allows expeditious tracheal intubation without priming.
When the PP was first described in 1985, the only neuromuscular blocking drug available with a rapid onset of action was succinylcholine. Priming was attractive because it seemed to fill a real clinical need. To quote Schwarz et al. (2) : “With the described technique, comparable intubating conditions could be obtained just as rapidly with vecuronium as with succinylcholine chloride, without subjecting the patients to the side effects of and the complications occasionally encountered with succinylcholine.” This early enthusiasm has been tempered with the passage of time. We think our analysis provides a theoretical basis for questioning the utility and safety of the PP.
1. Mehta MP, Choi WW, Gergis SD, et al. Facilitation of rapid endotracheal intubations with divided doses of nondepolarizing neuromuscular blocking drugs. Anesthesiology 1985; 62: 392–5.
2. Schwarz S, Ilias W, Lachner F, et al. Rapid tracheal intubation with vecuronium: the priming principle. Anesthesiology 1985; 62: 388–91.
3. Donati F. The priming saga: where do we stand now [editorial]? Can J Anesth 1988; 35: 1–4.
4. Stevens JB, Walker SC, Fontenot JP. The clinical neuromuscular pharmacology of cisatracurium versus vecuronium during outpatient anesthesia. Anesth Analg 1997; 85: 1278–83.
5. Puhringer FK, Scheller A, Kleinsasser A, et al. The effect of different priming doses on the pharmacodynamics of cisatracurium [in German]. Anaesthesist 2000; 49: 102–5.
6. Kopman AF, Klewicka MM, Neuman GG. An alternate method for estimating the dose-response relationships of neuromuscular blocking drugs. Anesth Analg 2000; 90: 1191–7.
7. Kopman AF, Klewicka MM, Ghori K, et al. Dose-response and onset/offset characteristics of rapacuronium. Anesthesiology 2000; 93: 1017–21.
8. Lebrault C, Chauvin M, Guirimand F, Duvaldestin P. Relative potency of vecuronium on the diaphragm and the adductor pollicis. Br J Anaesth 1989; 63: 389–92.
9. Meretoja OA, Wirtavuori K. Two-dose technique to create an individual dose-response curve for atracurium. Anesthesiology 1989; 70: 732–6.
10. Sheiner LB, Stanski DR, Vozeh S, et al. Simultaneous modeling of pharmacokinetics and pharmacodynamics: application to d-tubocurarine. Clin Pharmacol Ther 1979; 25: 358–71.
11. Sohn YJ, Bencini AF, Scaf AHJ, et al. Comparative pharmacokinetics and dynamics of vecuronium and pancuronium in anesthetized patients. Anesth Analg 1986; 65: 233–9.
12. Shanks CA. Pharmacokinetics of the nondepolarizing neuromuscular relaxants applied to calculation of bolus and infusion dosage regimens. Anesthesiology 1986; 64: 72–86.
13. Kopman AF, Klewicka MM, Kopman DJ, Neuman GG. Molar potency is predictive of the speed of onset of neuromuscular block for agents of intermediate-, short-, and ultra-short duration. Anesthesiology 1999; 90: 425–31.
14. Bluestein LS, Stinson LW Jr, Lennon RL, et al. Evaluation of cisatracurium, a new neuromuscular blocking agent, for tracheal intubation. Can J Anaesth 1996; 43: 925–31.
15. Belmont MR, Lien CA, Quessy S, et al. The clinical neuromuscular pharmacology of 51W89 in patients receiving nitrous oxide/opioid/barbiturate anesthesia. Anesthesiology 1995; 82: 1139–45.
16. McCoy EP, Connolly FM, Mirakhur RK, Loan PB. Nondepolarizing neuromuscular blocking drugs and train-of-four fade. Can J Anaesth 1995; 42: 213–6.
17. Kopman AF, Yee PS, Neuman GG. Correlation of the train-of-four fade ratio with clinical signs and symptoms of residual curarization in awake volunteers. Anesthesiology 1997; 86: 765–71.
18. Howardy-Hansen P, Chraemmer-Jorgensen B, Ording H, Viby-Mogensen J. Pretreatment with nondepolarizing muscle relaxants: the influence on neuromuscular transmission and pulmonary function. Acta Anaesthesiol Scand 1980; 24: 419–22.
19. Engbaek J, Howardy-Hansen P, Ording J, Viby-Mogensen J. Precurarization with vecuronium and pancuronium in awake, healthy volunteers: the influence on neuromuscular transmission and pulmonary function. Acta Anaesthesiol Scand 1985; 29: 117–20.
20. Donati F. Effect of dose and potency on onset. Anaesthesiol Pharmacol Rev 1993; 1: 34–43.
21. Healy TEJ, Pugh ND, Kay B, et al. Atracurium and vecuronium: effect of dose on the time of onset. Br J Anaesth 1986; 58: 620–4.
22. Ali HH, Savarese JJ. Stimulus frequency and dose response to d-tubocurarine in man. Anesthesiology 1980; 52: 36–9.