Anesth Analg 2001;93:339-344
© 2001 International Anesthesia Research Society
PEDIATRIC ANESTHESIA
The Pharmacoeconomics of Neuromuscular Blocking Drugs: A Perioperative Cost-Minimization Strategy in Children
William M. Splinter, MD, FRCPC, and
Lisa A. Isaac, MD, FRCPC
Departments of Anaesthesia, Children’s Hospital of Eastern Ontario and The University of Ottawa, Ottawa, Ontario, Canada
Address correspondence and reprint requests to Dr. William Splinter, Department of Anaesthesia, Children’s Hospital of Eastern Ontario, 401 Smyth Rd., Ottawa, Ontario, Canada K1H 8L1. Address e-mail to splinter{at}cheo.on.ca
 |
Abstract
|
|---|
The purpose of this investigation was to compare the costs of intermediate-acting neuromuscular blocking drugs in children during routine ambulatory surgery. We studied 200 healthy, 2–10-yr-old children undergoing elective dental restorative surgery. During Part 1 of the study, children received an inhaled anesthetic with halothane and nitrous oxide, whereas in Part 2, the anesthetic was IV propofol with nitrous oxide. The study drugs were atracurium, cisatracurium, mivacurium, rocuronium, and vecuronium. Patients were initially administered 2x the effective dose for 95% of the study drug. After recovery to 10% of baseline neuromuscular function, the neuromuscular blockade was rigidly maintained with an infusion of the study drug at about 10% of baseline function. Neuromuscular drug costs were approximated as drug usage x cost/unit. The initial drug costs were not substantially different for both Parts 1 and 2, but over time, mivacurium became the most expensive drug and cisatracurium the least expensive. In conclusion, based on current costs, cisatracurium is the least expensive intermediate-acting neuromuscular drug.
IMPLICATIONS: For children undergoing minor ambulatory procedures of 1–2 h, and continuous intraoperative neuromuscular blockade is indicated, cisatracurium currently is the least expensive drug.
|
Introduction
|
|---|
"Fiscal policy" has been a health care catch phrase for the past few years, and has dramatically affected health care management and health care workers. There have been many policy changes in health care in an effort to minimize costs in hospitals, including hospital closures and amalgamations. The pharmaceutical industry has been particularly affected by these changes. Although the proportion of hospital operating costs directed to anesthetic pharmaceuticals is small, approximately 0.24% of hospital budget (1), they are a great proportion of variable anesthetic supply costs. Strategies are being developed that minimize anesthetic costs and maintain the quality of health care. Because neuromuscular blocking (NMB) drugs constitute approximately 30% of the total anesthesia drug budget in the United States (2), these drugs are appropriate targets for cost-minimization strategies in anesthesia.
There have been almost no cost-minimization studies published that have relevance to NMB drugs (3). A cost comparison of mivacurium and rocuronium for adult ambulatory surgery concluded that mivacurium was less expensive than rocuronium for ambulatory surgery of 1–2-hr duration, i.e., $26.00 versus $42.00 (US).1
There are also numerous, nonpeer-reviewed reports by the pharmaceutical industry. Further, unbiased investigations are indicated.
Anesthetics are among the numerous factors that can affect the potency of NMB drugs. Inhaled anesthetics have variable, synergistic effects on NMB drugs, whereas most IV anesthetics, which are used in "balanced anesthesia" have little effect on NMB action. Because each class of anesthetic may have an effect on the potency of NMB drugs, their use in an investigation of NMB drugs must be strictly controlled. Also, comparisons among studies are often difficult because of differences in anesthetics used. In the current investigation, we evaluated and compared the perioperative cost of intermediate-acting NMB drugs in children during both inhaled anesthesia with halothane and IV anesthesia with propofol in children. For brief operations of <30 min, we expected mivacurium to be least expensive because it often does not require reversal. As the length of anesthesia increased, rocuronium or cisatracurium was expected to be the least expensive. For prolonged operations, the cost of cisatracurium was expected to be least, because it has a small cost when used for maintenance of neuromuscular blockade.
|
Methods
|
|---|
This two-part investigation was performed with the approval of the Research Ethics Committee of Children’s Hospital of Eastern Ontario and with written consent from parent or guardian, and if indicated, patient assent. Healthy patients aged 2–10 yr, of ASA physical status I–II, who underwent elective dental restorative surgery were enrolled in this study. Patients were excluded from the study if they had known allergies or contraindication to any of the drugs that were to be administered during this investigation. Other exclusions included gastroesophageal reflux, anomaly of the arm (right or left) precluding the use of electromyelogram (EMG) monitoring, or a disease process that may affect NMB activity, such as a neuromuscular disorder, including Duchenne’s muscular dystrophy, disuse atrophy, paralysis, peripheral neuropathies, and seizures.
Patients were assigned using a computer-generated random number table to receive either cisatracurium, mivacurium, rocuronium, vecuronium, or atracurium as the muscle relaxant. Standard patient monitoring included electrocardiogram, noninvasive blood pressure cuff, temperature probe (axillary), pulse oximetry, capnography, respiratory monitors, inhaled anesthetic monitors, and inspired and expired oxygen monitors. All patients received premedication with oral midazolam 0.5 mg/kg 20–30 min before anesthesia. IV access was established under nitrous oxide (50%) analgesia in the operating room (OR). Induction of general anesthesia was by inhalation of halothane and nitrous oxide in Part 1 of our study and by IV propofol, 3.5 mg/kg, and nitrous oxide (50%) in oxygen in Part 2. Maintenance of general anesthesia was with 70% nitrous oxide in oxygen plus halothane (expired concentration 0.8%) in Part 1 and with 70% nitrous oxide in oxygen plus propofol (10 mg · kg-1 · h-1) in Part 2. Halothane concentration and propofol infusions were adjusted to maintain systolic blood pressure within 10 mm Hg of baseline. The doses of NMB drugs used in this study were assumed to have minimal effects on hemodynamic variables. Baseline blood pressures were defined as the average of the first blood pressure in the OR and the systolic blood pressure measured in the day-care surgical unit. Temperature was monitored and maintained between 36.0° and 37.5°C to prevent the influence of hypothermia on the duration of neuromuscular blockade. End-tidal carbon dioxide was maintained between 35 and 45 mm Hg.
Neuromuscular blockade was assessed with accelerography. The adductor pollicis was monitored in all patients, and we specifically followed the train-of-four stimulation at 2 Hz, 0.1-ms duration of stimulus repeated every 20 s.
One of five nondepolarizing NMB drugs was administered in a double-blinded manner IV after establishing a stable baseline EMG recording immediately after induction of anesthesia. Endotracheal intubation was facilitated with an intubating dose (2x the effective dose for 95% [ED95]) of the study drug, as detailed in Table 1 (4–15). This was followed by a maintenance infusion of the study drug when the EMG had reached 10% of baseline twitch height. The initial maintenance doses are shown in Table 1 (16–24). The NMB infusion was adjusted every 2 min on an as-needed basis to maintain NMB at approximately 90% blockade. More specifically, if the twitch height was >95%, the infusion was decreased by 50%, and if the twitch height was <85%, the infusion was increased by 50%. The objective of these multiple adjustments was to maintain all patients at almost identical twitch height, i.e., 90%.
Intraoperative IV fluids were Ringer’s lactate at standard rates, i.e., 1/2 the deficit during the first hour, plus maintenance fluids. Patients did not receive atropine at induction or during maintenance of anesthesia. Muscle relaxants were not administered during the last 10 min of surgery.
Upon completion of surgery, the nitrous oxide and halothane or infusions of propofol were discontinued, and residual neuromuscular blockade was reversed with 0.015 mg/kg atropine and 0.5 mg/kg edrophonium if the twitch height was less than a 70% recovery according to the NMT monitor (25). With adequate recovery of neuromuscular activity, the endotracheal tube was removed after spontaneous ventilation had returned and before the return of upper airway reflexes. The patients were transferred to the recovery room after receiving 100% oxygen for 30 s. Routine postoperative orders were not altered by this study.
NMB drug costs were approximated as follows: NMB drug usage x acquisition cost/unit. Subsequently, the cost of NMB reversal was then added to the cost of NMB drug. At the time of the study, the acquisition cost in Canadian (CDN) dollars of atropine was $0.006/mg, whereas the acquisition cost of edrophonium was $0.044/mg. Thus, the acquisition cost for NMB reversal was $0.50 for a 10-kg patient.
Age and weight were compared with one-way analysis of variance and two-way Student’s t-tests with a Bonferroni correction factor. Gender was compared with 
2 analysis. An acceptable 
error was set at 0.05.
Sample size was estimated to be 100 for each part of the study. We initially set a sample size that would accurately predict muscle relaxant requirements in children. For the purpose of the current investigation, we defined accurate as a mean (µ) value for muscle relaxant use at 45 min after intubation that has a standard error of the mean (SEM) of 
5% of the mean. We expected this would require 15–20 patients per group to obtain such an accurate assessment of the muscular relaxant requirements based on our pilot studies.
|
Results
|
|---|
Two hundred subjects were enrolled in this study. Subjects were excluded from analysis because of major protocol violations such as drug administration pump failure, administration of nonstudy drugs, e.g., IV induction in Part 1. The patients within Parts 1 and 2 of the study had similar demographic data (Table 2). Endotracheal intubation was readily facilitated in all patients. Most patients who received mivacurium did not require neuromuscular reversal drugs, whereas most patients in the other four groups did (Table 2). Halothane usage in Part 1 and propofol usage in Part 2 were similar among the NMB study groups. Patients had uneventful recovery from anesthesia and surgery. Endotracheal extubation was uneventful for all patients, i.e., there were no emergency airway problems. Postoperative vomiting was negligible, with only one patient in each Part of the study vomiting in the recovery room. The time from initial bolus to starting the maintenance infusion was similar for the groups studied, except mivacurium (Table 2). Adjustments in infusion rates varied according to drug. Rocuronium required more adjustments (6 ± 3) than atracurium or cisatracurium (3 ± 3), P < 0.01.
The NMB drug usage is listed in Table 3. Drug usage was converted to cost by multiplying ‘drug usage in mg per kg’ by ‘drug cost in $CDN per mg drug.’ The cost of NMB drugs at the time of this study in $CDN/mg was: atracurium $0.356/mg, cisatracurium $0.9975/mg, mivacurium $0.512/mg, rocuronium $0.255/mg, and vecuronium $1.874/mg. Initial NMB drug costs were not substantially different, but with infusion of medication, mivacurium became the most expensive and cisatracurium the least expensive (Table 4). The NMB drug cost plus the cost of NMB reversal for all patients except those who received mivacurium are in Table 5. Initial drug costs (NMB plus reversal) were least for mivacurium, but by 30 min, cisatracurium was least expensive.
View this table:
[in this window]
[in a new window]
|
Table 5. Cumulative Drug Costs over Time for a 10-kg Patient (NMB Drug Plus Reversal, Where Applicable; Cost $CDN/10 kg)
|
|
|
Discussion
|
|---|
In this investigation, we observed decreased NMB acquisition costs with cisatracurium during both inhaled and IV anesthesia in children. This cost minimization was not time related, so was present throughout the anesthetic. If one accounts for the cost of NMB reversal drugs, the initial cost of mivacurium was least, but mivacurium’s cost rapidly increased and surpassed cisatracurium by 30 minutes after induction.
The universality of the results of a study is important. Neuromuscular blockade during dental surgery was studied because dental surgery is a common, uncomplicated operation and homeostasis is readily achievable during surgery. Potential confounding factors, such as hypothermia, blood loss, and third-space fluid loss, are easily kept to a minimum. Another factor that alters the potency of NMB drugs and the relevancy of this study is patient age (26). Children are typically divided into neonate, i.e., less than 1 month; infant, i.e., 1–24 months; child, i.e., 2–10 years; and youth/adolescent, i.e., 13–18 years. For the current investigation, we studied the most common age group in pediatric anesthesia, 2–10 years.
Cost analysis for each of these drugs was not straightforward. Numerous models may be used (27). Factors that affect cost include the initial or loading doses required for endotracheal intubation, cost of maintenance doses, and the cost of the reversal of muscle relaxants. For each drug, there was a curve generated that described time versus the amount of the drug in milligrams used. The amount of drug used was converted into dollars by multiplying by the unit drug cost. It must be noted that in today’s markets, drug costs, especially NMB drugs, are in a fluid state, so the graphs illustrating costs need to be repeated each time there is a notable unit price change.
The amount of drug used for each patient was calculated and drug wastage was also tabulated. It was decided that drug wastage was of secondary importance. The decision to exclude drug wastage may be considered controversial. The true cost of a drug includes both the drug administered plus drug wastage. Unfortunately, drug wastage cannot be calculated reliably. There are no standard methods for determining waste. The amount of drug wasted is both operator-dependent and varies from one institution to another. Within our institution, some anesthesiologists withdraw several doses of a medication in advance of usage just in case it is needed, whereas others do not. Obviously, waste varies from one anesthesiologist to another at our site. Similar variations in practice occur at other sites. Guidelines for the use of pharmaceutical drugs have been produced (3), and these are readily customized for any institution, but it is not reasonable to expect each institution to have similar guidelines for usage and wastage. Finally, pharmaceutical drugs are not supplied to users in identical vials from one jurisdiction to another. In one country, mivacurium may be supplied in 20-mg multidose vials, whereas in another, single-dose 10-mg ampoules may be used. For a patient requiring a 12-mg dose, the jurisdiction with multidose vials would likely have minimal wastage, whereas the single dose ampoules would have rather substantial waste, i.e., 8 mg or 40%.
The recommended doses of NMB drugs used in the OR are calculated by performing dose-response studies which include least-squares linear regression of the log-probit transformation of the administered doses. With these curves, both ED50 and ED95 are determined (4,28,29). The dose of NMB drug administered to a patient is primarily based on the patient’s age, the anesthetic being used, and ED95 for the drug to be used. It is important to note that the ED95 used in calculating the dose of NMB drug to be administered to the patient must have been calculated during similar clinical conditions. That is, an ED95 calculated during balanced anesthesia is only applicable to other patients receiving a balanced anesthetic. For the current study, we used published ED95 values obtained during balanced anesthesia or inhaled anesthesia. If these data were not available, we used data from inhaled anesthesia and assumed, based on other investigations, that the ED95 during balanced anesthesia is 120% that measured during inhaled anesthesia.
Our observed pharmacodynamics were similar to that reported in the literature. Duration of action of bolus injections was as expected (5,7,9) and infusion rates were often as predicted after a few adjustments. Balanced anesthesia required a larger dose for intubation and maintenance when compared with inhaled anesthesia. Thus, as expected, costs of neuromuscular blockade were more during balanced anesthesia than during inhaled anesthesia.
During this study, our NMB monitoring was with accelerography. This provided accurate data for the measurement of neuromuscular blockade. The alternative is the less popular and more bulky forced transduction monitor, which requires immobilization.
The literature reports a variety of infusion rates to maintain twitch height 90%–99% depressed. We chose infusion rates that would depress twitch height 90%, which represents common clinical practice during anesthesia. Also, this neuromuscular block is easily reversed at the end of the procedure. Infusion rates were increased or decreased 50% every 2 minutes, when indicated, to maintain the EMG at 90% depression and have a precise prediction of NMB drug requirements. We predicted that the EMG would reach 10% of baseline height anywhere from 10–45 minutes after the initial bolus of NMB drug. Also, we expected numerous adjustments of the NMB infusions, because the pharmacodynamics and pharmacokinetics of NMB drugs are quite variable in children.
Patients in this study received NMB drugs by an atypical clinical technique. In practice, most anesthesiologists administer an intubating dose of an NMB drug followed by intermittent boluses on an as-needed basis. This technique leads to marked fluctuations in blood levels and NMB, and is a rather imprecise estimation of true need. To obtain a more precise determination of NMB drug need, blockade was maintained with an infusion. The infusion was adjusted if needed every two minutes, which is a rather labor-intensive approach. This led to tight control of NMB and almost identical levels of NMB in all subjects and, finally, an accurate estimate of need for all of the patients studied.
We expected that only the patients who received mivacurium would recover spontaneously from the neuromuscular blockade and would not require edrophonium and atropine. Surprisingly, the occasional patient in other groups recovered rather rapidly from their neuromuscular blockade and apparently did not require reversal of residual blockade.
Many children undergo anesthesia each year and receive muscle relaxants. The most economical nondepolarizing muscle relaxant given by infusion available for cases lasting 30–120 minutes currently is cisatracurium. We evaluated cost of drugs with similar safety profiles. There are times when a particular muscle relaxant is clinically superior to other drugs in its class, and drug cost should not then be considered. Also, we did not evaluate the cost of rare, unpredictable adverse reactions, such as prolonged muscle relaxation after mivacurium in a patient with atypical pseudocholinesterase activity. The costs of other uncommon side effects, such as inadequate reversal, could be easily considered similar among the groups studied.
In conclusion, the cost of intermediate-acting NMB drugs during inhaled and balanced anesthesia were considered in children undergoing a typical operation. After determining direct drug usage, acquisition drug costs were determined at the time of the study. At the time of the study, cisatracurium was the least expensive drug. If an NMB reversal drug was required, mivacurium was slightly less expensive initially, but, by 30 minutes after the start of the NMB, cisatracurium was least expensive. Clinicians may readily calculate their local direct drug costs by determining the product of their local acquisition costs and the predicted drug use from Table 3. When appropriate, other costs, such as the cost of reversal drugs and drug wastage, should be added.
|
Acknowledgments
|
|---|
Financial support for this project was received from the Children’s Hospital of Eastern Ontario Research Institute.
|
Footnotes
|
|---|
1 Joshi GP, Garg SA, Yu SY. Cost comparison of mivacurium and rocuronium for ambulatory anesthesia [abstract]. Anesthesiology 1996;85:A1034. 
|
References
|
|---|
-
Hawkes C, Miller D, Martineau R, et al. Evaluation of cost minimization strategies of anaesthetic drugs in a tertiary care hospital. Can J Anaesth 1994; 41: 894–901.[Abstract/Free Full Text]
-
DeMonaco HJ, Shah AS. Economic considerations in the use of neuromuscular blocking drugs. J Clin Anesth 1994; 6: 383–7.[ISI][Medline]
-
Lubarsky DA, Glass PSA, Ginsberg B, et al. The successful implementation of pharmaceutical practice guidelines. Anesthesiology 1997; 86: 1145–60.[ISI][Medline]
-
Goudsouzian NG, Gionfriddo M, Rudd GD. Safety and efficacy of atracurium in adolescents and children anesthetized with halothane. Anesthesiology 1983; 59: 459-62.[ISI][Medline]
-
Brandom BW, Rudd GD, Cook DR. Clinical pharmacology of atracurium in paediatric patients. Br J Anaesth 1983; 53: 117S.
-
Pittet J-F, Melis A, Rouge J-C, et al. Effect of volatile anesthesia on vecuronium-induced neuromuscular blockade in children. Anesth Analg 1990; 70: 248–52.[Abstract/Free Full Text]
-
Fisher DM, Miller RD. Neuromuscular effects of vecuronium (ORG NC45) in infants and children during N2O, halothane anesthesia. Anesthesiology 1983; 58: 519–23.[ISI][Medline]
-
Meistelman C, Loose JP, Saint-Maurice C, et al. Clinical pharmacology of vecuronium in children. Br J Anaesth 1986; 58: 996–1000.[Abstract/Free Full Text]
-
Goudsouzian NG, Martyn JJ, Liu LM, Gionfriddo M. Safety and efficacy of vecuronium in adolescents and children. Anesth Analg 1983; 62: 1083–8.[Abstract/Free Full Text]
-
Woelfel SW, Brandom BW, McGowan FX, et al. Clinical pharmacology of mivacurium in pediatric patients less than two years old during nitrous oxide-halothane anesthesia. Anesth Analg 1993; 77: 713–20.[Abstract/Free Full Text]
-
Goudsouzian NG, Alitimoff JK, Eberly C, et al. Neuromuscular and cardiovascular effects of mivacurium in children. Anesthesiology 1989; 70: 237–42.[ISI][Medline]
-
From RP, Pearson KS, Choi WW, et al. Neuromuscular and cardiovascular effects of mivacurium chloride (BW B1090U) during nitrous oxide-fentanyl-thiopentone and nitrous oxide-halothane anaesthesia. Br J Anaesth 1990; 64: 193–8.[Abstract/Free Full Text]
-
Shorten GD, Crawford MW, St. Louis P. The neuromuscular effects of mivacurium chloride during propofol anesthesia in children. Anesth Analg 1996; 82: 1170–5.[Abstract]
-
Motsch J, Leuwer M, Bottiger BW, et al. Dose-response, time-course of action and recovery of rocuronium bromide in children during halothane anaesthesia. Eur J Anaesthesiol Suppl 1995; 11: 73–8.[Medline]
-
Woelfel SK, Brandom BW, Cook DR, Sarner JB. Effects of bolus administration of ORG 9426 in children during nitrous oxide-halothane anesthesia. Anesthesiology 1992; 76: 939–42.[ISI][Medline]
-
Kalli I, Meretoja OA. Infusion of atracurium in neonates, infants and children: a study of dose requirements. Br J Anaesth 1988; 60: 651–4.[Abstract/Free Full Text]
-
Meretoja OA, Wirtavouri K, Taivainen T, Olkkola KT. Time course of potentiation of mivacurium by halothane and isoflurane in children. Br J Anaesth 1996; 76: 235–8.[Abstract/Free Full Text]
-
Brandom BW, Sarner JB, Woefel SK, et al. Mivacurium infusion requirements in pediatric surgical patients during nitrous oxide-halothane and during nitrous-oxide-narcotic anesthesia. Anesth Analg 1990; 71: 16–22.[Abstract/Free Full Text]
-
Alikimoff JK, Goudsouzian NG. Continuous infusion of mivacurium in children. Br J Anaesth 1989; 63: 520–4.[Abstract/Free Full Text]
-
Meretoja OA, Olkkola KT. Pharmacodynamics of mivacurium in children, using a computer-controlled infusion. Br J Anaesth 1993; 71: 232–7.[Abstract/Free Full Text]
-
Ridley SA, Braude N. Post-tetanic count and intense neuromuscular blockade with vecuronium in children. Br J Anaesth 1988; 61: 551–6.[Abstract/Free Full Text]
-
Brandom BW, Cook DR, Woelfel SK, et al. Atracurium infusion requirements in children during halothane, isoflurane and narcotic anesthesia. Anesth Analg 1985; 64: 471–6.[Abstract/Free Full Text]
-
Tobias JD. Continuous infusion of rocuronium in a paediatric intensive care unit. Can J Anaesth 1996; 43: 353–7.[Abstract/Free Full Text]
-
Woelfel SW, Dong M-L, Brandom BW, et al. Vecuronium infusion requirements in children during halothane-narcotic-nitrous oxide, isoflurane-narcotic-nitrous oxide and narcotic-nitrous oxide anesthesia. Anesth Analg 1991; 73: 33–8.[Abstract/Free Full Text]
-
Baxter MRN, Bevan JC, Samuel J, et al. Postoperative neuromuscular function in pediatric day-care patients. Anesth Analg 1991; 72: 504–8.[Abstract/Free Full Text]
-
Meretoja OA, Wirtavuori K. Influence of age on the dose response relationship of atracurium in paediatric patients. Acta Anaesthesiol Scand 1988; 32: 614–8.[ISI][Medline]
-
Einarson TR, Shear NH, Oh PI. Models for pharmacoeconomic analysis. Can J Clin Pharmacol 1997; 4: 25–9.
-
Katz RL, Stirt J, Murray AL, Lee C. Neuromuscular effects of atracurium in man. Anesth Analg 1982; 61: 730–4.[Abstract/Free Full Text]
-
Fisher DM, Fahey MR, Cronelly R, Miller RD. Potency determination for vecuronium (ORG NC45): comparison of cumulative and single dose techniques. Anesthesiology 1982; 57: 309–10.[ISI][Medline]
Accepted for publication April 11, 2001.
Top
Abstract
Introduction
