This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (22) Citing Articles via Google Scholar Google Scholar Articles by Myles, P. S. Articles by Buckland, M. R. Search for Related Content PubMed PubMed Citation Articles by Myles, P. S. Articles by Buckland, M. R. Related Collections Economics and Health Care Research Heart Anesthetic Techniques Pharmacology

---

CARDIOVASCULAR ANESTHESIA

Remifentanil, Fentanyl, and Cardiac Surgery: A Double-Blinded, Randomized, Controlled Trial of Costs and Outcomes

Paul S. Myles, MB.BS, MPH, MD, FCARCSI, FANZCA^*, Jennifer O. Hunt, RN^*, Helen Fletcher, RN^*, Jennifer Watts, RN Bbus, PGradDip.Hlth.Eco, MCom(EC), David Bain, MB.BS, FANZCA^*, Andrew Silvers, MB.BS, FANZCA^*, and Mark R. Buckland, MB.BS, FANZCA^*

*Department of Anaesthesia & Pain Management, Alfred Hospital; and Departments of Anaesthesia and Epidemiology & Preventive Medicine and the Centre for Health Program Evaluation, Monash University, Clayton, Victoria, Australia

Address correspondence and reprint requests to A/Prof. Paul S. Myles, Department of Anaesthesia & Pain Management, Alfred Hospital, Commercial Road, Prahran, Victoria, 3181 Australia. Address e-mail to p.myles{at}alfred.org.au

	Abstract

Top Abstract Introduction Methods Results Discussion Conclusions References

 
Remifentanil may be beneficial in patients undergoing coronary artery bypass graft surgery, by promoting hemodynamic stability, reducing drug requirements, and attenuating the neurohumoral "stress response." We enrolled 77 cardiac surgical patients in a double-blinded, randomized trial and randomly allocated them to one of three groups: remifentanil infusion at 0.83 µg · kg^-1 · min^-1 (Group R); fentanyl bolus, small dose, at 12 µg/kg (Group FLD); and fentanyl bolus, moderate dose, at 24 µg/kg (Group FMD). We found a significant difference in the median time to tracheal extubation: Group FLD, 6.5 h; Group R, 7.3 h; and Group FMD, 9.7 h (P = 0.025). Group R patients had similar times to those of Groups FLD (P = 0.14) and FMD (P = 0.30). Group FLD patients had a longer length of hospital stay (P = 0.030). Patients in Group R had a significantly infrequent rate of hypertension but a frequent rate of hypotension (P < 0.01). The urinary cortisol excretion was larger in Group FLD patients (P < 0.0005), and urine flow was smaller (P < 0.0005). Remifentanil was associated with a propofol dose reduction (P = 0.0005) and a concomitant higher bispectral index (P = 0.032). Three Group FLD patients, but none in groups FMD and R, had postoperative myocardial infarctions (P = 0.032). Remifentanil has larger drug acquisition costs but does not increase the total hospital costs associated with cardiac surgery.

IMPLICATIONS: Remifentanil did not significantly reduce the duration of tracheal intubation after cardiac surgery. Remifentanil, when compared with fentanyl (total doses of approximately 15 and 28 µg/kg), blunts the hypertensive responses associated with cardiac surgery but is associated with more hypotension; when compared with fentanyl 15 µg/kg, remifentanil reduces cortisol excretion. Larger-dose opioids (remifentanil 0.85 µg · kg^-1 · min^-1 or fentanyl 28 µg/kg) were associated with a decreased rate of myocardial infarction after cardiac surgery.

	Introduction

Top Abstract Introduction Methods Results Discussion Conclusions References

 
Goals of anesthesia for coronary artery bypass graft (CABG) surgery include hypnosis, hemodynamic stability, and neurohumoral stress ablation (1) and may include early tracheal extubation (2,3). Fast-track cardiac surgery is partly dependent on smaller-dose opioid regimens (2,3), and although these may be associated with increased hemodynamic responses during surgery, there does not appear to be an increased risk of complications (3).

Remifentanil is a satisfactory adjunct during cardiac surgery (4–8), but it is uncertain whether it should replace other established opioids. Given its unique pharmacokinetic profile, large-dose remifentanil can be used to ablate stress responses and should allow faster recovery after cardiac surgery. This approach may reduce perioperative complications and length of stay. Nevertheless, remifentanil is a relatively expensive drug, and it is unclear whether the additional cost of acquisition can be outweighed by savings in total hospital costs. We therefore studied the costs and outcomes of a remifentanil regimen compared with two separate fentanyl regimens for cardiac surgery.

	Methods

Top Abstract Introduction Methods Results Discussion Conclusions References

 
After Ethics Committee approval, we approached all consecutive eligible elective CABG surgical patients aged <75 yr and obtained written, informed consent. Patients were excluded if they were >100 kg, were allergic to trial medications, were considered at very high risk (defined as a Tu score of >9) (9), had uncontrolled hyper- or hypotension, had congestive cardiac failure or an ejection fraction of <25%, had atrioventricular or left bundle branch block detected on their preoperative electrocardiogram (ECG), or had a pacemaker in situ.

This was a prospective, randomized, double-blinded trial of three active treatment groups:

1. Group R: remifentanil infusion at 0.83 µg · kg^-1 · min^-1.
   
2. Group FLD: fentanyl bolus, small dose, at 12 µg/kg.
   
3. Group FMD: fentanyl bolus, moderate dose, at 24 µg/kg.
   

Patients were stratified before randomization into low risk (Tu score <2) (9) and high risk (Tu score 2–6) to maximize equality of groups. The clinical trials unit of our hospital pharmacy department prepared each of the following solutions according to the randomization code (table of random numbers) and maintained blinding of study drug preparations:

1. Group R: saline (placebo) in 20-mL syringes for the bolus and remifentanil 5 mg in 50-mL syringes (100 µg/mL) for the infusion.
   
2. Group FLD: fentanyl 500 µg in 20-mL syringes for the bolus and saline (placebo) in 50-mL syringes for the infusion.
   
3. Group FMD: fentanyl 1000 µg in 20-mL syringes for the bolus and saline (placebo) in 50-mL syringes for the infusion.
   

The patient’s usual medications were continued until the time of operation. All patients received a standard premedication of oral temazepam 10 mg, IM morphine 5 mg, and oxygen delivered via face mask at 5 L/min. Patient monitoring consisted of five-lead ECG, pulse oximetry, capnography, invasive arterial pressure, and pulmonary artery pressure. Correct ECG ST segment monitoring was confirmed (with definition of isoelectric line and J point). Intravascular catheters were inserted with local anesthesia before induction by using IV midazolam in 0.5-mg increments.

Bispectral index (BIS^®; Aspect Medical Systems Inc., Natick, MA) electrodes were applied according to the manufacturer’s instructions, and BIS^® A1050 monitoring commenced. All patients had an identical induction technique commencing with a propofol infusion at 8 mg · kg^-1 · h^-1 until loss of the eyelash reflex. An intravenous trial drug bolus was then administered at 0.32 mL/kg via the 20-mL syringes, and the trial drug infusion immediately commenced at 0.5 mL · kg^-1 · h^-1 by using the 50-mL syringes prepared by the pharmacy. Thus, the patients received one of the following: fentanyl 8 µg/kg (Group FLD), fentanyl 16 µg/kg (Group FMD), or remifentanil 0.83 µg · kg^-1 · min^-1 (Group R). Muscle relaxation and tracheal intubation followed 3 min after IV pancuronium 0.12 mg/kg. The propofol infusion was reduced to 5 mg · kg^-1 · h^-1 immediately after tracheal intubation. A further bolus of trial drug, 0.16 mL/kg, was given before sternotomy. Thus, patients in Groups FLD and FMD had received their complete allocated trial medication at this stage, with those in Group R continuing to receive remifentanil at 0.83 µg · kg^-1 · min^-1.

Further propofol and study drug dosage adjustments were standardized by protocol, according to adverse hemodynamic responses and BIS^® measurements. In general, this consisted of

1. a bolus of propofol and trial drug and increased propofol and trial drug infusion rates if signs of "light" anesthesia (hypertension, tachycardia, or BIS^® >60)
   
2. or reduced propofol and trial drug infusion rates if signs of "deep" anesthesia (hypotension or BIS^® <40).
   

All patients had a nitroglycerin (NTG) infusion commenced after the induction of anesthesia at 0.2 µg · kg^-1 · min^-1 (except during cardiopulmonary bypass [CPB]). A dose of NTG 0.4 µg · kg^-1 · min^-1 was used after removal of the aortic cross-clamp.

Patients’ lungs were ventilated with a tidal volume of 10 mL/kg, adjusting the respiratory rate to an end-tidal carbon dioxide of 32–34 mm Hg. CPB was standardized by using a crystalloid/colloid prime, membrane oxygenator, moderate hypothermia (32°C–34°C), and alpha-stat pH management.

Postoperative analgesia consisted of IV ketorolac 10 mg and morphine 1–2 mg/h. Neuromuscular block was reversed. Propofol sedation, if required, was commenced at 0.5 mg · kg^-1 · h^-1. Patients remained intubated and mechanically ventilated for transfer to the intensive care unit (ICU). Patients were weaned from mechanical ventilation as soon as they responded to verbal stimuli and were normothermic, hemodynamic stability had been established, and blood loss was satisfactory (<l00 mL/h). This adhered to standard fast-track protocols (2,3). Tracheal extubation occurred when the patient was awake and cooperative, with a respiratory rate between 10 and 20 breaths/min, and had satisfactory arterial blood gas analyses.

Hemodynamic variables were measured at set time points. Cardiac output was measured with the thermodilution method (10 mL of room temperature saline at end-expiration, in triplicate). The total amount of intraoperative vasoactive drugs and cardiac pacing requirements were recorded. The patient’s intraoperative urine volume was recorded for calculation of flow and cortisol excretion.

Continuous three-lead (II, aVL, V5) automated ST segment analysis was used to detect intraoperative myocardial ischemia. Ischemia was defined as ST segment depression >1 mm or elevation >2 mm at 60 ms after the J point, persisting for at least 2 min; each new episode was confirmed with a real-time visual check by the investigator. Transesophageal echocardiography was not routinely used. A 12-lead ECG was performed on admission to the ICU and daily for 3 days after surgery. Blood was taken at 6 and 36 h for troponin I levels. Acute myocardial infarction was diagnosed if new Q waves appeared in at least two adjacent ECG leads, as determined by a blinded cardiologist.

Sedative and analgesic requirements were documented for the first 24 h after ICU admission. Time to tracheal extubation (from ICU admission) and "fitness for discharge" from ICU were recorded. (ICU "fit for discharge" criteria were 1) awake and cooperative [Glasgow Coma Scale >12]; 2) extubated, ventilation satisfactory [respiratory rate <30 breaths/min, pH >7.30]; 3) temperature >36.0°C; 4) stable hemodynamics; 5) no life-threatening arrhythmias; 6) blood loss <200 mL/4 h; and 7) inotropes: dopamine <5 µg · kg^-1 · min^-1 [or equivalent]).

Cost analyses were performed by a health economist (JW), by using a data collection form designed to capture resource use in the operating theater (OT) and cardiothoracic ICU (10). Intraoperative resource use data were recorded, including the total procedure time, drugs, and doses administered. ICU resource use included recovery times, length of stay, and drug administrations. In addition, the time the patient was "fit for discharge" was recorded to take account of any exit blocks between ICU and the general ward.

The total hospital costs for 51 patients were extracted from the hospital’s clinical costing system according to the following cost categories: OT, ICU, ward nursing, pharmacy, imaging, pathology, and an "other" category, which included allied health and medical. For 23 pre-July 1999 patients, clinical costing data could not be extracted because the hospital’s system was not operating. For these patients, utilization data were extracted from the admission transfer discharge system, and the costs were estimated by using mean unit costs from the subsequent patients. Cost data were available from 76 patients.

The two sources of costing and resource utilization data (data collection form and hospital clinical costing data) enabled estimation of costing data in the areas where the trial anesthetic drug regimen may affect perioperative resource use, particularly the OT and ICU. A modeled OT and ICU cost was determined by substituting the relevant recorded activity from the hospital’s information system with length of stay and OT drug administration from the data collection form. For pharmacy costs, the actual drugs and amounts administered were recorded for all relevant drugs. Acquisition drug costs provided by the hospital’s pharmacy department were used to calculate OT drug costs. All modeled costs were substituted for data extracted from the clinical costing system. Total cost thus included clinical costing data for ward, pharmacy, pathology, imaging, and other costs and modeled costs for OT and ICU for 76 patients. These modeled costs provide a more precise total cost based on a more accurate recording of resource utilization in the OT and ICU. Hospital length of stay included the day of surgery. All costs are presented in Australian dollars (1999–2000), where AUS$1 US$0.55.

A preliminary estimate of sample size of 25 patients per group was based on an expected 30% reduction in time to tracheal extubation after surgery, on the basis of our unit’s current practice (mean difference 10 vs 7 h, SD 4 h; Type I error of 0.05; Type II error of 0.20). The primary end-points were time to tracheal extubation and length of hospital stay.

Normally distributed data were analyzed by using general linear models and are presented as mean (SD). The Tu score (9) and baseline hemodynamic variables were included as covariates in the analysis of hemodynamic data. Nonnormally distributed data were analyzed with Kruskal-Wallis analysis of variance and are presented as median (interquartile range); post hoc testing was with Student’s t-tests or Mann-Whitney U-tests. Associations were compared by using the Spearman rank order correlation, rho. Proportions were analyzed by ² and Yates’ correction or Fisher’s exact test and are presented as n (%). A P value <0.01 was considered significant for the analysis of the multiple hemodynamic variables. The Hochberg procedure was used for post hoc comparisons (11). All statistical analyses were performed with SPSS for Windows Version 9.0 (SPSS, Inc., Chicago, IL).

	Results

Top Abstract Introduction Methods Results Discussion Conclusions References

 
Eighty-seven patients were enrolled in this study, with subsequent exclusion of 10 patients (1 patient withdrew consent before surgery, 2 patients had their surgery deferred, and 7 patients failed to receive their allotted study medication). The groups were well matched for relevant perioperative factors (Table 1). No patient reported intraoperative awareness after direct questioning. One patient in Group FLD had a postoperative stroke. There was a significant difference in the time to tracheal extubation, and patients in Group FLD had a longer length of hospital stay (Table 2).

Mean and median total variable costs are reported for the three groups of patients, with the high-cost outlier cases included, in Table 3. These findings show that, although not statistically significant, Group FLD had a higher mean total cost compared with the other two groups, with a mean cost difference of approximately $1000. Further analyses of mean cost data are reported both with and without high-cost outliers (Tables 3 and 4).

Remifentanil was associated with a reduction in the need to give additional opioids (P = 0.0005) and significantly reduced anesthetic (propofol) requirements (Table 7). BIS^® scores were higher in Group R (P = 0.032). There were no differences in postoperative analgesic (P = 0.24) or sedative requirements (P = 0.21).

	Discussion

Top Abstract Introduction Methods Results Discussion Conclusions References

Patients in Group FLD were able to be extubated earlier than those in Group FMD, with patients in Group R being intermediate. Because there may be a relationship between the neurohumoral stress response and postoperative myocardial ischemia (1), large-dose opioid techniques have been the preferred technique in many centers. Our study suggests that remifentanil can provide this effect without prolonging recovery times. In contradistinction, small-dose fentanyl, although associated with shorter tracheal extubation times, did not reduce ICU or hospital length of stay.

We found that remifentanil provided better mitigation against hypertension than either moderate- or small-dose fentanyl, although it contributed to more episodes of hypotension requiring vasoconstrictor use. The larger dose of fentanyl afforded some degree of protection when compared with small-dose fentanyl. Other markers of cardiac performance, particularly the need for cardiac pacing or inotropic support, were not affected.

There was a highly significant reduction in cortisol excretion in Group R patients (and to a lesser extent those in Group FMD) compared with the small-dose fentanyl technique, suggesting greater neurohumoral stress ablation with remifentanil. A preliminary report¹ found that remifentanil blunted catecholamine secretion in CABG patients. This mechanism may explain the increased urine flow rates seen with remifentanil. In our study, both fentanyl groups required significantly more "top-up" opioid doses to treat light anesthesia or hemodynamic hyperresponsiveness. There were in fact lower average BIS^® scores in Group FLD patients, and this is consistent with the additional hypnotic required to maintain hemodynamic stability.

Our findings are generally consistent with those of Howie et al. (6), who found similar times to tracheal extubation with remifentanil and moderate-dose (approximately 22 µg/kg) fentanyl by using an isoflurane-based technique. They reported fewer hemodynamic disturbances with remifentanil but could not identify a relative reduction in catecholamine levels in this comparative study, which may be explained by the marked variance in their catecholamine data, suggesting that their study was underpowered for this end-point. In a separate article (7), their group reported similar resource utilization, but no explicit cost comparisons were performed. They did not include a smaller-dose fentanyl group, so our study provides additional information for those who use such a technique.

There was an apparent increased rate of myocardial infarction in Group FLD, although this was limited to three patients. Cheng et al. (3) found that early tracheal extubation was not associated with any demonstrable increase in postoperative complications.

This study suggest that the higher direct cost of remifentanil may be associated with cost savings as a result of a reduction in ward length of stay, particularly when comparing remifentanil with small-dose fentanyl. No significant difference in mean total variable cost among the three groups suggests that the decision to use remifentanil should not rely solely on the comparatively higher cost of the drug itself. These findings suggest that there may be downstream savings because of a reduction in postoperative general ward length of stay, thus countering the effect of the higher direct drug costs in the perioperative setting. A comparison of actual and "fit for discharge" times in the ICU suggests that there may be some savings to hospitals where ICU "exit blocks" are minimized, although simulation studies suggest that this may not be easily achieved (12).

Cost comparisons were limited by the sample size of our study. It is interesting to note that Engoren et al. (8) found no statistically significant difference in costs (P = 0.3) when comparing fentanyl, sufentanil, and remifentanil for cardiac surgery, but there appeared to be higher costs with fentanyl (US$7841) compared with remifentanil (US$6286). The decision to use a particular drug regimen is generally based on clinical factors; however, in output-based hospital funding environments, pressure is often applied to clinicians to minimize costs. Further studies incorporating a larger sample size to take resource differences into account may be valuable in influencing these decisions.

We chose to study two doses of fentanyl (12 and 24 µg/kg), aiming to represent current cardiac anesthetic practice, in which early tracheal extubation is a goal. It is possible that larger doses may be more effective in modifying the stress response, but they may also lead to a slower recovery profile. We did not find a significant difference in rates of myocardial ischemia; this may be related to the small incidence of this complication, so our study may have been underpowered to provide a reliable comparison, and other monitoring may have been more sensitive. Because we performed statistical tests on multiple end-points, there is an increased likelihood of a Type I error. Our study was not designed to identify smaller differences in some end-points (myocardial ischemia, ICU, and hospital discharge). We therefore urge considered interpretation by the reader and would recommend further investigation to confirm our results.

	Conclusions

Top Abstract Introduction Methods Results Discussion Conclusions References

 
Remifentanil, when compared with fentanyl (total doses of approximately 15 and 28 µg/kg, respectively), did not significantly reduce the duration of tracheal intubation after CABG surgery. However, remifentanil did blunt hypertensive responses and, when compared with fentanyl 15 µg/kg, cortisol excretion; this was at the expense of more episodes of hypotension. Larger doses of opioids (remifentanil 0.85 µg · kg^-1 · min^-1 or fentanyl 28 µg/kg) may be associated with a decreased rate of myocardial infarction.

	Acknowledgments

 
Dr. Myles is supported by a National Health and Medical Research Council Practitioner Fellowship award. This study was funded by the Alfred Hospital Wholetime Medical Specialists Scheme and Glaxo Wellcome.

We would like to thank the following for their cooperation with the requirements of the trial protocol: Drs. Tony Weeks, Roderick McRae, Sesto Cairo, and Howard Machlin; and Shin Choo, Clinical Trials Pharmacy Department. Special thanks to Sean Downer and Jeremy Hose from the hospital’s Clinical Costing Department.

	Footnotes

 
Presented in part at the 74th International Anesthesia Research Society Annual Meeting, Honolulu, HI, March 10–14, 2000.

¹ Howie MB, Michelson LG, Porembka DT, et al. Anesthesia induction with remifentanil for patients undergoing CABG [abstract]. Anesth Analg 1996;82:S190.

	References

Top Abstract Introduction Methods Results Discussion Conclusions References

 

1. Hug CC, Shanewise JS. Anesthesia for adult cardiac surgery. In: Miller RD. Anesthesia, 4th ed. New York: Churchill Livingstone, 1994: 1757–810.
2. Myles PS, Weeks AM, Buckland MR, et al. Hemodynamic effects, myocardial ischemia and timing of extubation with propofol-based anesthesia for cardiac surgery. Anesth Analg 1997; 84: 12–9.[Abstract]
3. Cheng DCH, Karski J, Peniston Q, et al. Early tracheal extubation after coronary artery bypass graft surgery reduces costs and improves resource use: a prospective, randomized, controlled trial. Anesthesiology 1996; 85: 1300–10.[Web of Science][Medline]
4. Duthie DJR, Stevens JJWM, Doyle AR, et al. Remifentanil and pulmonary extraction during and after cardiac anesthesia. Anesth Analg 1997; 84: 740–4.[Abstract]
5. Olivier P, Sirieix D, Dasier P, et al. Continuous infusion of remifentanil and target-controlled infusion of propofol for patients undergoing cardiac surgery: a new approach for scheduled early extubation. J Cardiothorac Vasc Anesth 2000; 14: 29–35.[Web of Science][Medline]
6. Howie MB, Cheng D, Newman MF, et al. A randomized double-blinded multicenter comparison of remifentanil versus fentanyl when combined with isoflurane/propofol for early extubation in coronary artery bypass graft surgery. Anesth Analg 2001; 92: 1084–93.[Abstract/Free Full Text]
7. Cheng DC, Newman MF, Duke P, et al. The efficacy and resource utilization of remifentanil and fentanyl in fast-track coronary artery bypass graft surgery: a prospective randomized, double-blinded controlled, multi-center trial. Anesth Analg 2001; 92: 1093–102.
8. Engoren M, Luther G, Fenn-Buderer N. A comparison of fen-tanyl, sufentanil, and remifentanil for fast-track cardiac anesthesia. Anesth Analg 2001; 93: 859–64.[Abstract/Free Full Text]
9. Tu JV, Jaglal SB, Naylor CD. Multicenter validation of a risk index for mortality, intensive care unit stay, and overall hospital length of stay after cardiac surgery: Steering Committee of the Provincial Adult Cardiac Care Network of Ontario. Circulation 1995; 91: 677–84.[Abstract/Free Full Text]
10. Jackson T, Watts J, Lane L, Wilson R. Data comparability in patient level clinical costing systems. Casemix Q 1999; 1: 36–45.
11. Hochberg Y. A sharper Bonferroni method for multiple tests of significance. Biometrika 1988; 75: 800–2.[Abstract/Free Full Text]
12. Dexter F, Macario A, Dexter EU. Computer simulation of changes in nursing productivity from early tracheal extubation of coronary artery bypass graft patients. J Clin Anesth 1998; 10: 593–8.[Web of Science][Medline]

This article has been cited by other articles:

				T. Ledowski, B. Bein, R. Hanss, A. Paris, W. Fudickar, J. Scholz, and P. H. Tonner Neuroendocrine Stress Response and Heart Rate Variability: A Comparison of Total Intravenous Versus Balanced Anesthesia Anesth. Analg., December 1, 2005; 101(6): 1700 - 1705. [Abstract] [Full Text] [PDF]

				K. Rauf, A. Vohra, P. Fernandez-Jimenez, N. O'Keeffe, and M. Forrest Remifentanil infusion in association with fentanyl-propofol anaesthesia in patients undergoing cardiac surgery: effects on morphine requirement and postoperative analgesia Br. J. Anaesth., November 1, 2005; 95(5): 611 - 615. [Abstract] [Full Text] [PDF]

				P. S. Myles and D. McIlroy Fast-Track Cardiac Anesthesia: Choice of Anesthetic Agents and Techniques Seminars in Cardiothoracic and Vascular Anesthesia, March 1, 2005; 9(1): 5 - 16. [Abstract] [PDF]

				G. Cucchiaro, S. D. Markowitz, R. Kaye, N. S. Adzick, R. S. Litman, C. A. Stanley, and M. F. Watcha Blood Glucose Control During Selective Arterial Stimulation and Venous Sampling for Localization of Focal Hyperinsulinism Lesions in Anesthetized Children Anesth. Analg., October 1, 2004; 99(4): 1044 - 1048. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (22) Citing Articles via Google Scholar Google Scholar Articles by Myles, P. S. Articles by Buckland, M. R. Search for Related Content PubMed PubMed Citation Articles by Myles, P. S. Articles by Buckland, M. R. Related Collections Economics and Health Care Research Heart Anesthetic Techniques Pharmacology