EVALUATION OF ATRACURIUM NEUROMUSCULAR BLOCKADE IN PAEDIATRIC PATIENTS WITH BURN INJURY

Transcription

1 Br. J. Anaesth. (1988), 60, EVALUATION OF ATRACURIUM NEUROMUSCULAR BLOCKADE IN PAEDIATRIC PATIENTS WITH BURN INJURY A. K. MILLS AND J. A. J. MARTYN Depolarizing neuromuscular blocking drugs such as suxamethonium, can cause lethal hyperkalaemia when administered to burned patients [1, 2]. On the other hand, patients who have suffered thermal injury require increased doses of non-depolarizing myoneural blockers to achieve a given degree of neuromuscular blockade [2-5]. Atracurium is one such relatively new nondepolarizing neuromuscular blocking agent, which has a short to intermediate duration of action [6,7]. Its cardiovascular effects are minimal, even when large doses are administered in contrast to the long-acting neuromuscular blockers such as tubocurarine and pancuronium [8-10]. Therefore, in clinical conditions such as burns in which suxamethonium is contraindicated [1], the use of atracurium may provide a suitable therapeutic alternative. The present study was performed to determine if children with burns exhibit resistance to the neuromuscular effects of atracurium ; to study the relationship of resistance to time after burn and size of burn; and to construct dose-response curves for minor, moderate and major burns. SUMMARY The neuromuscular effects of atracurium were studied in acutely burned children, and in children at least 3 years after burn injury (controls). Thirty-one children were studied a total of 48 times. During nitrous oxide-narcotic anaesthesia, a single dose of atracurium was administered in each study and the twitch suppression recorded. Dose-response curves were established using least-squares regression techniques. The ED 95 of atracurium in children recovered from burn injury was 0.27 mg kg' 1, which is similar to that of normal children reported previously. During the first week of burn injury, the ED 95 was 0.3 mg kg' 1 irrespective of burn size, and was not significantly different from controls. After the first week of injury, in children with 20-60% body surface burn, the ED 95 was twice normal, while in those with greater than 60% body surface burn, it probably may be increased up to 3.0 mg kg' 1. PATIENTS AND METHODS Institutional approval was obtained from the Subcommittee on Human Studies, Committee on Research, and all studies conformed to ethical standards as delineated in the Declaration of ALEX K. MILLS, M.D. ; J. A. JEEVENDRA MARTYN,* M.D., F.F.A.R.C.S. ; Department of Anaesthesiology, Harvard Medical School, and the Anaesthesia Services of Shriners Burns Institute and Massachusetts General Hospital, Boston, Massachusetts U.S.A. Accepted for Publication: September 14, *Address for correspondence: Clinical Pharmacology Laboratory, Department of Anaesthesia, Massachusetts General Hospital, Boston, Massachusetts U.S.A. Helsinki. Thirty-one children of either sex, ranging in age from 6 months to 17 years, were studied a total of 48 times. The mean age and weight of each group is given in table I. The study patients were divided into five groups for purposes of analysis: Group 1. Recovered burned patients coming for plastic reconstructive surgery (controls). These patients were injured at least 3 years before the study. Group 2. Patients with acute burns who were studied within 7 days of injury. The mean (+SEM) size of burn in this group was 38 (±7)% (range 18-70%) total body surface area (TBSA).

2 ATRACURIUM IN BURNS 451 TABLE I. Physical characteristics of patients studied. Values are expressed as mean + SEM Group 1 Group 2 Group 3 Group 4 Group 5 Burn size (% TBSA) n Age (yr) Weight (kg) (0) < > ± ± ± ±0.6 30± ± Group 3. Patients with burns covering less than 20% TBSA (minor). The mean (±SEM) burn size of this group was 14 (±2)% TBSA. Group 4. Patients with burns between 20 % and 60% TBSA (moderate). The mean ( + SEM) burn size of this group was 37 (±4)% TBSA. Group 5. Patients with burns greater than 60 % TBSA (major). The mean ( + SEM) burn size of this group was 51 (±3)% TBSA. Patients in groups 2-5 were studied at least 1 week after and up to 60 days after burn injury. No patient had any serious medical problem, neuropathy or myopathy before sustaining their burn injury. There was no cardiac, hepatic or renal failure as indicated by clinical signs or laboratory tests including measurements of the concentrations of SGOT, SGPT, BUN and creatinine. Patients were excluded from the study if they were receiving any drug known to alter the response to neuromuscular blocking drugs, and these included drugs such as antibiotics and theophylline [11, 12]. However, patients received cephalosporin and aminoglycoside antibiotics immediately after the termination of the study. Premedication consisted of diazepam mg kg" 1 by mouth or morphine mg kg" 1 i.m. or i.v., or both drugs. The very young patients, in whom there was no venous access before the induction of anaesthesia, received methohexitone mg kg" 1 rectally. The ECG, arterial pressure, temperature and end-tidal carbon dioxide concentration (Hewlett-Packard) were monitored in each patient. Measurements of arterial pressure were obtained using the auscultatory method except in those patients who had indwelling arterial cannulae. Temperature was recorded by a telethermometer probe. Anaesthesia was induced with thiopentone 4-6 mg kg" 1 i.v. and 60 % nitrous oxide in oxygen by mask. In infants with no venous access, halothane was always the anaesthetic of choice, given by mask for a short period (< 5 min) until a vein was cannulated; in these patients, studies were not commenced for a minimum of 10 min after the discontinuation of the halothane. It was assumed that this duration of exposure and wash out of halothane would not have any significant effects on dose-response curves to atracurium. Narcotic agents (morphine or fentanyl i.v.) were administered as anaesthetic supplements. Ventilation was controlled to maintain end-tidal carbon dioxide tension between 4.26 and 5.85 kpa. After the induction of anaesthesia, the ulnar nerve was stimulated with a Grass S48 stimulator through 25-gauge needles inserted to a forearm which was immobilized on a padded board. The thumb was linked to a Grass FT03 force transducer, and the strength of contraction of the adductor pollicis was measured as twitch height on a Hewlett-Packard strip-chart recorder. Single square-wave stimuli of 0.15 ms duration were administered at 0.15 Hz. Once a stable plane of anaesthesia, normal haemodynamic values and a constant twitch height were achieved and sustained for 5 min, atracurium was administered. The determination of ED 50, ED 70 and ED 95 was made by use of the "single-dose" technique described previously [13, 14] since the incremental dose technique results in artefactually larger estimates of ED values for drugs such as atracurium which have a rapid recovery time. Thus only one predetermined arbitrary dose was administered per patient and the resultant twitch depression was recorded. Twitch depression was measured as the difference between the control twitch height and the height of the twitch once it had stabilized after the dose of atracurium. Twitch height was considered to be stable when five twitches were of the same height. At this point, the study was terminated. Further doses of neuromuscular blocking drug were given if clinically indicated, but these doses were not taken into consideration for the calculation of ED values because of the uncertainty of how much atracurium had been metabolized from the initial dose.

3 452 Statistical analysis of the relationship between the percentage of twitch suppression and the logarithm of the dose was performed using a multiple regression model that provided a common slope for all groups and a separate intercept for each group. Combining the data in this way yielded more stable estimates of the common slope and of the residual variability. Graphical examination of the data revealed no evidence of systematic departure from straightline dependence (within group) of twitch suppres- BRITISH JOURNAL OF ANAESTHESIA sion on the logarithm of dose. Thus it was not necessary to transform the percentage of twitch suppression (for example to a probit scale). A more comprehensive multiple regression model, providing a separate slope for each group, permitted a test for the adequacy of the common-slope model; the difference between the two models was not significant (P = 0.47 for atracurium). Point estimates and 95 % confidence intervals for specified effective doses (e.g. ED 50 ) were obtained by a straightforward modification of the method described by Finney [15] for probit analysis Dose (mg kg' 1 ) FIG. 1. Atracurium dose v. twitch depression in controls and acutely burned children. In the first week (#), the dose-response curves were similar to controls ( ). After this time, there was a relative resistance to the effects of atracurium which was related to the size of the burn. The slopes of the curves are similar, but the intercepts are significantly different between minor (A)> moderate (O) and major ( ) bums. These groups were also significantly different (P < 0.001) when compared with controls and burned patients studied within 1 week of injury. RESULTS A plot of twitch suppression v. dose of atracurium for all groups is shown in figure 1. The ED 50, ED 75, and ED 85 values for each group are shown in table II, together with the 95% confidence limits for each ED value. It should be noted that the ED 75 and ED 95 values for group 5 were obtained by extrapolation beyond the data points. Patients in group 1 (controls) had a dose-response curve essentially the same as that reported previously for normal children [6]. Group 2 (those studied within 7 days of injury) demonstrated a dose-response curve that was not significantly different from the controls. In these patients there was no correlation between size of burn and ED values. The children in the other groups showed a relative resistance to the neuromuscular effects of atracurium, apparently related to the magnitude of the burn. Although the slopes were not different (P = 0.43), the intercepts of the minor, moderate and major burns were signifi- TABLE II. ED values of atracurium in burned children and controls. Numbers in parentheses represent lower and upper 95% confidence limits, respectively. + Values obtained by extrapolation of line beyond data points. ***Intercepts significantly different (P < 0.001) from each other and from groups 1 and 2 ED 50 ED 75 ED 95 Group 1 (Controls) Burns < 1 Week Group 2 (18%-7O% TBSA Burns > 1 Week Group 3 (<20% TBSA) Group 4 (20-60% TBSA) Group 5 (>60% TBSA) ( ) ( ) ( ) ( ) 1.25 ( ) (0.145-O.0214) ( ) (0.203-O.292) ( ) 2.07t ( ) ( ) ( ) 0.367*** ( ) 0.573*** ( ) 3.093t*** ( )

4 ATRACURIUM IN BURNS 453 cantly different (P < 0.001) from the controls, and from the burned patients studied within 7 days. The intercepts of minor, moderate and major burns were also significantly different from each other. There were no major changes in haemodynamic indices associated with the administration of atracurium. No injuries or complications resulted from the study. DISCUSSION This study has demonstrated that children with thermal injury are resistant (hyposensitive) to the effects of atracurium. This resistance was not observed in the first week after injury, irrespective of the area of the burn. After this period, however, resistance was observed, the magnitude of which was related to the size of burn. In other words, the greater the area of the burn, the greater were the ED values. Our study is consistent with another recent report indicating that burned patients are resistant to the neuromuscular effects of atracurium, and that this hyposensitivity does not develop until about 1 week after burn [5]. This latter study was performed in adults and did not establish ED values for twitch inhibition. Our study, therefore, adds to the previous work in that it was performed in children and establishes ED values for children with minor, moderate and severe burns. The ED 95 of 0.27 mg kg" 1 in our control patients was similar to that reported for normal children (0.3 mg kg" 1 ) [6]. During the first week after burn, the ED 95 was 0.3 mg kg" 1, a value not significantly different from that obtained in the control group. In group 2 there was no correlation between burn size and ED values. After the first week of injury the ED 95 in patients with < 20 % burn was shifted 1.2-fold (120%). The ED 95 value for patients with 20-60% TBSA burns was approximately twice that of the control patients, while the ED 95, extrapolated beyond the data points (calculated), for patients with burns greater than 60% TBSA burns was almost 10 times that of control values. Such a resistance to the neuromuscular effects of atracurium has not been seen in any other clinical pathological state. A combination of factors might account for this significant shift to the right in the dose-response curve for atracurium. These factors include altered kinetics [16], altered protein binding [17, 18] and, possibly, the presence of circulating substances in the plasma [19, 20]. Altered kinetics which might influence dose include changes in central volume of distribution and 24-h urinary excretion [16]. Enhanced drug binding to plasma is brought about by increases in alpha x -acid glycoprotein [17, 18]. Circulating factors which may shift the dose-response curve to the right include changes in the concentrations of cyclic nucleotides, prostaglandins and other unidentified substances [11, 19, 20]. More importantly, there seems to be a generalized increase in the number of nicotinic acetylcholine receptors in all skeletal muscles following burn injury [21]. Therefore, the changes in acetylcholine receptors at the skeletal muscle following burns are akin to the changes seen following denervation [1]. Clinically, electromyographic studies in burned patients also confirm the presence of a denervation-like state which includes positive sharp waves and fibrillation potentials [22]. Such a change in target organ sensitivity could account for the aberrant responses to both suxamethonium and the nondepolarizing drugs [2]. An important advantage of atracurium is its short duration of action and minimal cardiovascular effects compared with tubocurarine, dimethyl-tubocurarine and pancuronium. A good substitute for suxamethonium for use in emergency situations such as a "full stomach" or in laryngospasm is presently not available for use in burned patients [2]. High doses of pancuronium and metocurine administered together produce rapid onset of paralysis, but the time to recovery to 25 % of control twitch height is greater than 80 min [23]. High-dose atracurium may prove useful as a substitute for suxamethonium in these emergency situations. A dose of atracurium 1.5 mg kg" 1 (6 x ED 95 in normal individuals) has been administered to normal patients and 95 % twitch depression was achieved in approximately 60 s [9]. This was associated with a decrease in arterial pressure of only (mean + SD) mm Hg. Although arterial pressure was not monitored in a systematic, timed fashion, we did not observe any clinically significant decrease in our patients when high doses were administered. A recent study, in adult burned patients, confirms that high-dose atracurium does not cause significant hypotension [24]. These authors, however, used only 0.8 mg kg" 1 in both burned and unburned patients. A 20 % decrease in arterial pressure was seen in the normal patients whereas, in burned patients, a

5 454 BRITISH JOURNAL OF ANAESTHESIA 5% decrease was observed. The reason for the absence of significant hypotension in burned patients is subject to speculation. Clinical study with larger doses (> 1.5 mg kg" 1 ) to evaluate onset times and haemodynamic changes is warranted, since atracurium 0.8 mg kg" 1 will not cause as rapid an onset of paralysis in burned patients as suxamethonium. High-dose vecuronium may also prove useful [25]. In burns, another theoretical advantage of atracurium over other presently available nondepolarizing neuromuscular blocking drugs is its lack of dependence on the kidney or liver for breakdown [26]. Furthermore, the metabolic breakdown products of atracurium have no effect on the neuromuscular junction. Thus, when burned patients have impaired renal or hepatic function, atracurium may be superior to other drugs. On the other hand, burned patients have altered kinetics of almost every drug studied [4]. Serum concentrations of specific and non-specific esterases are depressed [27, 28]. The effects of these alterations in enzyme concentrations on the pharmacokinetics of atracurium in burned patients need to be clarified in future studies. This study has established dose-response curves for burned paediatric patients, and confirmed those for normal patients. Most thermally injured patients do show relative resistance to atracurium, and the degree of resistance appears to be related to extent of burn and the time after the injury. The ED 95 for control subjects is 0.27 mg kg" 1 and is not significantly different from the ED 95 of burned patients studied within 1 week of injury, irrespective of burn size. In patients with moderate burns, the ED 95 is twice normal, while in patients with major burns it may be increased up to 10-fold (3.0 mg kg" 1 ). The pharmacokinetics and the cardiovascular and neuromuscular pharmacodynamics of high-dose atracurium as a substitute for suxamethonium in burned patients deserve further study. ACKNOWLEDGEMENTS This study was supported in pan by grants from the National Institutes of Health GM and from the Shriners Burns Institute. The authors acknowledge the secretarial assistance of Beverley Nuss. Statistical advice was provided by David Hoaglin, Ph.D., and organizational assistance by Gene Mahoney. REFERENCES 1. Gronert GA, Theye RA. Pathophysiology of succinylcholine-induced hyperkalemia. Aneslhesiology 1975; 43: Martyn JAJ, Goldhill DR, Goudsouzian NG. Clinical pharmacology of neuromuscular relaxants in burned patients. Journal of Clinical Pharmacology 1986; 26: Martyn JAJ, Szyfelbein SK, Ali HH, Matteo RS, Savarese JJ. Increased d-tubocurarine requirement following major thermal injury. Aneslhesiology 1980; 52: Martyn JAJ. Clinical pharmacology and drug therapy the burned patient. Anesthesiology 1986; 65: Dwersteg JF, Pavlin EG, Heimbach DM. Patients with burns are resistant to atracurium. Anesthesiology 1986; 65: Goudsouzian NG, Liu LMP, Gionfriddo M, Rudd GD. Neuromuscular effects of atracurium in infants and children. Aneslhesiology 1975: 62: Goudsouzian NG, Liu LMP, Cote CJ, Gionfriddo M, Rudd GD. Safety and efficacy of atracurium in adolescents and children anesthetized with halothane. Anesthesiology 1983; 59: 459^ Basta SJ, Savarese JJ, Ali HH, Moss J, Gionfriddo M. Histamine releasing potencies of atracurium, dimethyltubocurarine and tubocurarine. British Journal of Anaesthesia 1983; 55 (Suppl. 1): 105S-106S. 9. Lennon RL, Olsen RA, Gronert GA. Atracurium or vecuronium for rapid sequence endotracheal intubation. Anesthesiology 1986; 64: Scott RPF, Savarese JJ, Basta SJ, Embree P, Ali HH, Sunder N, Hoaglin DC. Clinical pharmacology of atracurium given in high dose. British Journal of Anaesthesia 1986; 58: Fuke N, Martyn JAJ, Kim C, Basta SJ. Concentration dependent interaction of theophylline with d-tubocurarine. Journal of Applied Physiology 1987; 62: Sokoll M, Gergis S. Antibiotics and neuromuscular function. Anesthesiology 1981; 55: Fisher DM, Fahey MR, Cronnelly R, Miller RD. Potency determination for vecuronium. Comparison of cumulative and single-dose techniques. Anesthesiology 1982; 57: rding H, Skovgaar J. Dose-response curves for vecuronium during halothane and neurolept anaesthesia: single dose versus cumulative method. Acta Anaesthesiologica Scandinavica 1985; 29: Finney DJ. Probit Analysis. 2nd edn. Cambridge: Cambridge University Press Section 19, 1952; 61: Martyn JAJ, Matteo RS, Greenblatt DJ, Lebowitz PW, Savarese JJ. Pharmacokinetics of d-tubocurarine in patients with thermal injury. Anesthesia and Analgesia 1982; 61: Leibel WS, Martyn JAJ, Szyfelbein SK, Miller KW. Elevated plasma binding cannot account for the burn related d-tubocurarine hyposensitivity. Aneslhesiology 1981; 54: Martyn JAJ, Abemathy DR, Greenblatt DJ. Plasma protein binding of drugs after severe burn injury. Clinical Pharmacology and Therapeutics 1984; 35: Storella RJ, Bierkamper GG, Martyn JAJ. The potency of non-depolarizing muscle relaxants is reduced by burn plasma. Anesthesiology 1986; 65: A292.

6 ATRACURIUM IN BURNS Tomera JF, Martyn JAJ. Increases in burn size and resulting changes at the neuromuscular junction in mice. Physiologist 1986; 29: Kim C, Fuke N, Martyn JAJ. Thermal injury to rat increases nicotinic acetylcholine receptors in the diaphragm. Anesthesiology 1988; (in press). 22. Mills AK, Shriefer T, Martyn JAJ. Electromyographic studies in patients with thermal injury. Anesthesiology 1986; 65: A Hagen J, Martyn JAJ, Szyfelbein SK, Goudsouzian N.G. Cardiovascular and neuromuscular responses to highdose pancuronium-metocurine in pediatric burned and reconstructive patients. Anesthesia and Analgesia 1986; 65: Dwersteg JF, Pavlin EG, Hascke R, Heimbach DM, Maclntyre PE. High dose atracurium does not produce hypotension in the burned patient. Anesthesiology 1986; 65: A Mills A, Martyn JAJ. Evaluation of vecuronium neuromuscular blockade in patients with thermal injury. Anesthesia and Analgesia 1987; 86: S Hughes R, Chappie DJ. The pharmacology of atracurium: A new competitive neuromuscular blocking agent. British Journal of Anaesthesia 1981; 53: Frohlich S. Serum cholinesterase deficiency in major burns. Burns 1978; 4: Viby-Mogensen J, Hanel HK, Hansen E, Grace J. Serum cholinesterase activity in burned patients: Anesthesia, suxamethonium and hyperkalemia. Ada Anaesthesiologica Scandinavica 1975; 19: