| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
 
|
|
|
||
|
|
© 2000 International Anesthesia Research Society
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
This study prospectively determined whether there was a learning curve with the use of remifentanil, as indicated by decreased hemodynamic variability, improved recovery profile, and decreased incidence of opioid-related adverse events with increasing experience. Patients undergoing diverse surgical procedures (outpatient [n = 1340] and inpatient [n = 560]) were enrolled by investigators (n = 190) who had no previous experience with remifentanil use. Each investigator enrolled 10 patients. A standardized protocol for administration of remifentanil was used. Data were analyzed to determine differences between the first three patients and the last three patients enrolled for each anesthesiologist in the study. There were no differences in hemodynamic variables between the first triad and the last triad in either outpatients or inpatients. Requirements for hypnotic drugs and the doses of remifentanil used were also similar between groups. Analgesic medications administered at the end of surgery and in the postanesthesia care unit (PACU) were similar between groups, except that the last triad in the outpatient group received smaller doses of fentanyl compared with the first triad. Times to response to verbal command, tracheal extubation, and operating room discharge did not differ between groups. However, patients in the last triad undergoing outpatient surgery had shorter times to eligibility for PACU discharge, but times to eligibility for discharge home did not differ. The overall incidence of all adverse events (i.e., hypotension, hypertension, muscle rigidity, respiratory depression, apnea, nausea, and vomiting) was less in the last triad as compared with the first triad. When analyzed separately, only the incidence of vomiting (in the outpatient group) was decreased in the last triad as compared with the first triad. This study suggests that there is a learning curve that aids reduction of minor adverse effects associated with the use of analgesic medications administered at the end of surgery in outpatients, which might have reduced the incidence of postoperative vomiting and the duration of PACU stay.
Implications: This study demonstrated that anesthesiologists rapidly acquire the ability to use remifentanil with limited experience. However, there is a learning curve that aids reduction of minor adverse effects associated with the use of analgesic medications administered at the end of surgery in outpatients, which might have reduced the incidence of postoperative vomiting and the duration of postanesthesia care unit stay.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Remifentanil is a new, potent µ-opioid receptor agonist with pharmacodynamic properties similar to other µ-opioid receptor agonists (e.g., fentanyl, sufentanil, and alfentanil). However, in contrast to other opioids, remifentanil is rapidly metabolized by nonspecific esterases in the blood and tissues. It has a context-sensitive half-time of approximately 3 min and elimination half-time of approximately 10 min. These times are significantly shorter than corresponding times for alfentanil, sufentanil, or fentanyl (1,2). Because of the unique pharmacokinetic characteristics of remifentanil, it may have numerous advantages. However, an educational program (3) may be necessary, and perhaps a learning curve associated with its use.
Honkavaara and Paloheimo (4) performed a post hoc analysis of data from a multicenter study of propofol to evaluate whether there was a learning effect with its use. They found that, compared with patients enrolled in the earlier part of the study, those enrolled later received a larger dose of propofol at the induction of anesthesia, earlier administration of the last additional dose, and an earlier recovery from anesthesia. They concluded that there was a learning effect with the use of propofol. Similarly, the incidence of adverse events should be more frequent during the initial use of new drugs, possibly as a result of a learning effect (5). Because there are so little published data available concerning physicians’ learning processes in adopting anesthetic drugs, especially those whose pharmacokinetics differ so greatly from established paradigms, we designed this study to investigate whether there was a learning curve associated with the initial use of remifentanil.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
This study complied with the Declaration of Helsinki (as amended) and local institutional review board approval, and written, informed consent was obtained for each patient. There were 190 investigators within the contiguous 48 United States and Hawaii who took part in this study. None of these investigators had previous experience with the use of remifentanil. Each investigator completed a 3- to 5-h educational program regarding the use of remifentanil. The information provided included the pharmacokinetics of remifentanil, the clinical studies evaluating its use, the drug interactions between remifentanil and other anesthetics, and the need for providing postoperative analgesia before discontinuation of remifentanil infusion. Investigators used a remifentanil-based anesthetic technique in 10 consecutive qualified patients, under open-label conditions. Investigators were designated a priori as an outpatient (n = 134) or inpatient (n = 56) site and were to enroll only patients undergoing outpatient or inpatient procedures, respectively.
Patients (1340 outpatients and 560 inpatients) of any ASA physical status scheduled for elective surgical procedures under general anesthesia of at least 30-min duration and requiring tracheal intubation were enrolled in the study. Except for the administration of opioids, premedication was unrestricted. Anesthesia was induced with either IV propofol or thiopental according to the usual practice of the investigator. Investigators were taught in the educational session that concurrent administration of remifentanil could reduce the dose of hypnotic that was needed. The administration of remifentanil followed its US-approved labeling (Ultiva®; GlaxoWellcome Inc., Research Triangle Park, NC).
An infusion of 1 µg · kg-1 · min-1 remifentanil (0.5 µg · kg-1 · min-1 for patients older than 65 yr) was started simultaneously with the administration of the anesthetic used to induce anesthesia. For patients 30% heavier than ideal body weight, remifentanil was dosed according to ideal body weight, not actual body weight. The remifentanil infusion could be reduced at any time in the event of hypotension and was reduced to 0.25 µg · kg-1 · min-1 after tracheal intubation and could then be titrated (up or down) in anticipation of a stressful stimulus or for hemodynamic instability. Upward titration was either a 1-µg/kg bolus dose or an increase in infusion rate of 25%–100% (i.e., initially, 0.0625–0.25 µg · kg-1 · min-1). Remifentanil infusions were limited to a maximum of 2 µg · kg-1 · min-1, and after two upward titrations, the investigators could titrate the hypnotic drugs.
Anesthesia was maintained with isoflurane or propofol infusion and N2O or air in O2, according to the investigators’ usual practice. Initial inspired concentrations of isoflurane were 0.2%–0.8%, in pursuit of end-tidal concentrations of 0.2% and 0.4%, in the presence and absence of N2O, respectively. Infusion rates for propofol depended on the use of N2O and patient’s age. For patients receiving N2O, the suggested propofol infusion rates were 50–75 µg · kg-1 · min-1 and 50–100 µg · kg-1 · min-1 for patients older and younger than 55 yr, respectively. In patients not receiving N2O, the suggested infusions for propofol were 75–100 µg · kg-1 · min-1 and 100–150 µg · kg-1 · min-1 for patients older and younger than 55 yr, respectively. Muscle relaxants and reversal drugs were administered at the discretion of the anesthesiologist.
All patients received longer-acting analgesic medications (e.g., morphine 10–15 mg, fentanyl 1–2 µg/kg, IV) 20–30 min before the anticipated end of surgery to provide postoperative analgesia. Remifentanil infusion was discontinued with the last suture or last surgical manipulation in the case of closed procedures. Other anesthetic drugs and N2O were discontinued according to the anesthesiologist’s usual practice.
Age, sex, ASA physical status, and duration of anesthesia were recorded. Heart rate and arterial blood pressure were recorded just before the induction of anesthesia, at loss of consciousness, tracheal intubation, start of surgery, skin closure, and on arrival in the postanesthetic care unit (PACU). The time from end of surgery to response to a verbal command, extubation, discharge from the operating room, eligibility for discharge from the PACU, and (for outpatients) eligibility for discharge from the hospital were also recorded. Concomitant drugs and their route and time of administration were recorded.
Investigators reported adverse events occurring at any time during the study. Adverse events were reported without consideration of causality, even if unrelated to the study drug or test procedures. The frequency, duration, and severity of all adverse events were recorded. A specific subset of adverse events met a regulatory definition of "serious" (i.e., an adverse event that was fatal, life-threatening, prolonging hospitalization, or causing unscheduled hospitalization).
Seven specific adverse events were prospectively identified, which corresponded with known pharmacological properties of potent opioids. The quantitative criteria for these adverse events were: hypertension: systolic blood pressure (SBP) 
180 mm Hg, diastolic blood pressure (DBP) 110 mm Hg, or requiring treatment with a pharmacologic drug; hypotension: SBP 
80 mm Hg or requiring treatment with a pharmacologic drug; muscle rigidity of any degree; respiratory depression: respiratory rate 8 breaths/min for 60 s or more; apnea for 15 s; and nausea or vomiting: as reported by the patient, investigator, or otherwise recorded in the chart. Nonspecific adverse events were converted to standard terms by using MIDASTM (GlaxoWellcome Inc.), a proprietary dictionary that is similar to COSTARTTM. Adverse events were categorized by body system, clinical severity, and drug-attributability.
Patients were stratified by clinical setting (i.e., outpatient or inpatient). The sample size was determined by the number of investigators and patients per investigator required to estimate adverse event rates with a standard error of <5% and to detect a 5% or larger overall difference in incidence of adverse events with 80% power at an 
level of 5%. For this purpose, investigators were the primary sampling unit, and their patients were repeated observations for each investigator. Because of variations within each investigator (e.g., surgical procedures and length of surgery), initial experience included the first three patients and end experience included the last three patients. If the difference in adverse event rates was large, then analysis techniques for repeated measures were used to explore the trend in incidence by patient order. Sample size calculations for modeling this trend indicated that 10 patients per investigator would be adequate. Data were analyzed to determine whether the first three patients enrolled by each investigator had greater hemodynamic changes, delayed recovery or discharge from the PACU or discharge home (for outpatients), or higher incidence of adverse events as compared with the last three patients enrolled by the investigator. The Cochran-Mantel Haenszel test was used for comparing adverse event frequency, the analysis of variance test for comparing the means, and the Cox proportional hazards model for comparing recovery times. A P value of <0.05 was considered statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The demographic characteristics of patients, the duration of anesthesia, and the ASA physical status were well balanced for the first (1–3) and last (8–10) triads of patients within the ordered population (Table 1). The types of surgical procedures were similarly well balanced in the ordered population. For outpatients, these comprised 78.2% and 79.4% superficial, 21.8% and 20.1% nonlaparoscopic abdominal, and 0% and 0.5% thoracic, for 1–3 and 8–10 subsets, respectively. For inpatients, surgery types were 54.5% and 50.0% superficial, 43.1% and 46.4% nonlaparoscopic abdominal, and 2.4% and 3.6% thoracic, for 1–3 and 8–10 subsets, respectively.
|
The doses of propofol or thiopental for the induction of anesthesia and the maintenance doses of propofol or isoflurane were similar for the first and last triads (Table 2). The total remifentanil requirements were also similar in the first and the last triads (Table 2). There were no significant differences in hemodynamic variables (i.e., heart rate and systolic and diastolic blood pressure) in the different subgroups except that the last triad of inpatients had significantly smaller systolic blood pressure at the time of skin closure (Table 3). There were no differences in the use of labetalol, esmolol, ephedrine, phenylephrine, atropine, or glycopyrrolate in the subgroups (data not shown).
|
|
With respect to analgesic medications for postoperative pain management, outpatients in the last triad received significantly smaller fentanyl doses (P < 0.01) compared with the first triad (Table 4). However, there were no differences in the use of other analgesics in the various subgroups (Table 4). The times to response to verbal command, tracheal extubation, and operating room discharge did not differ between the subgroups (Table 5). However, the patients in the last triad undergoing ambulatory surgery had significantly shorter times (by 11 min) to eligibility for PACU discharge. The times to eligibility for discharge home in outpatients were similar between the subgroups (Table 5).
|
|
The last triad had a statistically less frequent overall adverse event frequency as compared with the first triad of patients. However, when specific adverse events were analyzed separately, the only adverse event to achieve statistical significance was the incidence of vomiting in outpatients (Table 6). Of the five cases of muscle rigidity reported, three occurred in the first triad of inpatients, and two occurred in the last triad of outpatients (an overall incidence of 0.04%). The frequencies for both muscle rigidity and apnea were too low for any meaningful comparison between the surgical patient groups. The incidence of nonspecific adverse events were diverse and occurred at low frequencies.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Although previous studies have described learning curves with respect to gaining manual skills in regional anesthesia techniques, tracheal intubation, and arterial cannulation (6,7), or the use of new equipment (8), few studies have investigated the effects of learning with the use of new drugs (4,5). This large study was designed to prospectively investigate whether there was a learning curve associated with the use of remifentanil as a part of balanced anesthesia technique in a diverse patient population. The study design we used was similar to that reported by Honkavaara and Paloheimo (4) and Roizen and Toledano (9) in which they compared the efficacy and recovery characteristics of propofol in the group of patients enrolled early with those enrolled later in the study. In our study, patients were analyzed by clinical setting (i.e., outpatient or inpatient), because of the diverse characteristics of the anesthetic technique used for these patient populations and potential differences in adverse events associated with these techniques.
Because of its unique pharmacokinetic characteristics, remifentanil may have a more pronounced learning curve. Furthermore, in contrast to other commonly used opioids (i.e., fentanyl and sufentanil), remifentanil is most effectively administered by continuous IV infusion. This may further increase the learning requirements associated with the use of remifentanil. However, the requirements of hypnotic drugs (i.e., propofol or isoflurane) and remifentanil during the maintenance of anesthesia were similar in the first and the last triads of patients. In addition, there were no differences in the hemodynamic variables and the immediate recovery profile (i.e., response to verbal command, tracheal extubation, and discharge from the operating room).
In contrast to the observations of Honkavaara and Paloheimo (4) and Roizen and Toledano (9), with propofol we did not observe a learning effect with the use of remifentanil. It is possible that part of this result might be attributed to an educational program for all investigators before the start of the study. However, this effect could only be shown by comparing the outcome differences between the investigators who did or did not complete the education program. In the previous studies (4,9), no common group educational event occurred before the drug introduction. However, our results are in agreement with Roizen and Toledano (9), who suggested that learning requirements might be less in pharmaceutical studies. The lack of learning effect may be a result of the common educational session or the stringent study protocol that specified the detailed guidelines for remifentanil infusion, and significant oversight and monitoring may have precluded the anesthesiologists from judging appropriate requirements independently (i.e., learning through practice), thereby decreasing the learning effect. However, such a protocol occurred with propofol (4,9). Thus, either the educational session or drug differences must account for differences in the learning between propofol (4,9) and this drug.
Unlike other opioids, the pharmacodynamic effects of remifentanil rapidly dissipate with the discontinuation of its infusion. The lack of residual analgesic effects of remifentanil means that measures must be taken to provide postoperative pain relief before discontinuation of remifentanil (11). The transition between intraoperative analgesia with remifentanil and postoperative analgesia with other analgesic techniques may be challenging. Strategies used for postoperative pain management depend on the anticipated degree of pain, the concurrent anesthetics to be used, the duration of surgery, and whether an inpatient or outpatient procedure is being performed (11). The use of longer-acting opioids at the end of the surgery may confound the incidence of opioid-related adverse side effects in the postoperative period. With increasing experience with the management of postoperative pain in patients receiving remifentanil-based anesthesia, the incidence of adverse events should reduce.
We found that the doses of fentanyl used for postoperative analgesia in the last triad of patients undergoing ambulatory surgery were significantly smaller than those used in the first triad. Furthermore, the incidence of postoperative vomiting was significantly reduced in the last triad of outpatients and these patient had shorter times to eligibility for PACU discharge. The shorter times to eligibility for PACU discharge may be explained by fewer emetic symptoms. The less frequent incidence of postoperative vomiting may be the result of reduced use of fentanyl in this patient population (12). However, unrelieved pain increases the incidence of postoperative nausea and vomiting, particularly in patients undergoing ambulatory surgery (13). Perhaps the reduced incidence in vomiting may be related to improved postoperative pain management as the experience with administration of longer-acting opioids increased. This suggests a learning effect with the use of analgesic medications administered at the end of surgery in outpatients, which might have reduced the incidence of postoperative vomiting and reduced the PACU stay. This suggests that the education before introduction of remifentanil could be concentrated on these aspects of the anesthetic technique.
In a Phase IV postmarketing study of propofol, Roizen and Toledano (9) found that physicians initially enrolled younger and healthier patients, and as they gained experience with propofol, they enrolled older or sicker patients. However, there was no difference in the demographic characteristics and the ASA physical status between the first three and the last three patients enrolled by each investigator in this study. We found that the overall incidence of adverse events in this study was lower than in previous Phase II and III studies (i.e., product labeling of remifentanil). Similar observations were reported by McLeskey et al. (14) in a Phase IV postmarketing study of propofol. The incidence of muscle rigidity observed in this study (0.04%) was smaller than that reported in the product labeling or Food and Drug Administration Phase II–III studies of remifentanil (6%–17%) (15). The reduced incidence of adverse events may be caused by availability of more information regarding the potential adverse events and increased education resulting in change in dosing practice.
In summary, this study demonstrated that anesthesiologists rapidly acquire the ability to use remifentanil with limited experience. However, there is a learning curve with the use of analgesic medications administered at the end of surgery in outpatients, which might have reduced the incidence of postoperative vomiting and the PACU stay. This suggests that the education before remifentanil usage should be concentrated on these aspects of the anesthetic technique. Differences in learning effects during the introduction of this drug compared with the previous studies with propofol imply that a common educational session with this drug greatly aided learning or that learning with this drug in practice is significantly faster than it was with propofol.
|
Acknowledgments |
|---|
We thank the SOURSE study investigators, study site coordinators, and patients who took part in this study. The study management assistance of Deborah Thompson, MPH, is greatly appreciated.
|
Footnotes |
|---|
This study was sponsored by Glaxo Wellcome Inc., Research Triangle Park, NC, under protocol number USAA4001.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
- Kapila A, Glass PSA, Jabobs JR, et al. Measured context-sensitive half-times of remifentanil and alfentanil. Anesthesiology 1995; 83: 968–75.[Web of Science][Medline]
- Egan TD, Lemmens JH, Fiset P, et al. The pharmacokinetics of the new short-acting opioid remifentanil (G187084B) in healthy adult male volunteers. Anesthesiology 1992; 76: 334–41.[Web of Science][Medline]
- Reves JG. Educational considerations for the clinical introduction and use of remifentanil. Anesth Analg 1999; 89: S4–6.
- Honkavaara P, Paloheimo M. Effect of learning during an anaesthesiological multicenter trial. Acta Anaesthesiol Scand 1998; 42: 1205–8.[Medline]
- Lew JKL, Spencer AA, Elton RA. Cross-sectional study of complications of inhalational anaesthesia in 16,995 patients. Anaesthesia 1991; 46: 810–5.[Medline]
- Kopacz DJ, Neal JM, Pollock JE. The regional anesthesia "learning curve": what is the minimum number of epidural and spinal blocks to reach competency? Reg Anesth 1996; 21: 182–90.[Web of Science][Medline]
- Konard C, Schupfer G, Wietlisbach M. Gerber H. Learning manual skills in anesthesiology: is there a recommended number of cases for anesthetic procedures? Anesth Analg 1998; 86: 635–9.[Abstract]
- Johnson C, Roberts JT. Clinical competence in the performance of fiberoptic laryngoscopy and endotracheal intubation: a study of resident instruction. J Clin Anesth 1989; 1: 344–9.[Medline]
-
Roizen MF, Toledano A. Technology assessment and the "learning contamination" bias. Anesth Analg 1994; 79: 410–2.
[Free Full Text] - Albrecht S, Schuttler J, Yarmush J. Postoperative pain management after intraoperative remifentanil. Anesth Analg 1999; 89: S40–5.
- Kehlet H, Rung GW, Callesen T. Postoperative opioid analgesia: time for a reconsideration? J Clin Anesth 1996; 8: 441–5.[Web of Science][Medline]
- Joshi GP. Pain management after ambulatory surgery. Ambulatory Surg 1999; 7: 3–12.
- Roizen MF, Stanley TH, Thisted RA, et al. How do anesthesiologists select patients when introducing a new drug into practice? Anesth Analg 1993; 77: S30–3.[Medline]
- McLeskey CH, Walawander CA, Nahrwold ML, et al. Adverse events in a multicenter Phase IV study of propofol: evaluation by anesthesiologists and postanesthesia care unit nurses. Anesth Analg 1993; 77: S3–9.
- Physician’s Desk Reference, 53rd ed. GlaxoWellcome Inc., Ultiva®. Montvale, NJ: Medical Economics Company, 1999:1233–8.
This article has been cited by other articles:
|
|
 |
|
  J. G. Reves Lessons on Learning About Learning Curves Anesth. Analg., November 1, 2000; 91(5): 1047 - 1048. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
