There was no significant difference between pain scores for fentanyl versus sufentanil (Fig. 4). The significant heterogeneity at the 19- to 24-hour time interval was likely driven by the single discordant study that favored sufentanil.37 Subgroup analysis was not feasible given the small number of studies.
Other comparisons involving morphine versus either sufentanil or oxycodone did not reveal any differences (data not shown). There were also no differences detected by the single adult study comparing morphine to hydromorphone and 2 pediatric studies that compared the same opioids.41–43
Postoperative Nausea and Vomiting
There was a higher incidence of PONV in patients receiving morphine compared with fentanyl (OR = 1.91; 95% CI, 1.14–3.18; Fig. 5). The absolute risk reduction for this result was 10.46%, which corresponds to a number needed-to-treat (NNT) of 9.6 patients (95% CI, 5.9–26.2). No other differences in PONV for other opioids were detected in pooled (Fig. 5) and subgroup analyses.
Patients receiving morphine experienced more pruritus than those receiving fentanyl, although the data were not statistically significant (OR = 1.64; 95% CI, 0.98–2.76; Fig. 6). Two studies in this group were not pooled. First, we excluded the data from Tsui et al.31 because an antipruritic was randomly administered to all study subjects. Second, the study by Vallejo et al.25 was the only trial to report pruritus on a continuous scale, and no difference between the groups was detected. With respect to other opioids and subgroup analyses (data not shown), no further differences were detected.
Respiratory Depression, Hypotension, and Sedation
Regardless of the opioids being compared, the incidence of respiratory depression and hypotension was low across most trials, with 0 to 2 patients generally experiencing the outcome in each group. In addition, the number of patients experiencing sedation was generally similar between groups in the same trial but varied markedly among trials. For example, 1 trial reported sedation in 98 of all 105 patients33 and another reported sedation in 2 of 64 patients.27 The only exception to this is the study by Chrubasik et al., where 0 and 9 of 20 patients receiving morphine and fentanyl were sedated, respectively.32 Given the similarity in respiratory depression, hypotension, and sedation among studies, it is unsurprising that pooled and subgroup analyses did not reveal any significant differences in these outcomes (data not shown).
Total Opioid Consumption
Total opioid consumption was only compared for studies that used equipotent infusion rates and was standardized per 24 hours. As a caveat, not all studies reported the SD associated with the infusion rate, and in some cases it had to be estimated. There was slightly lower opioid consumption in the morphine group among trials comparing morphine versus fentanyl (1.22 mg of morphine equivalent; 95% CI, 0.27–2.18; P = 0.012; Fig. 7). There were no differences in opioid consumption in studies comparing fentanyl to sufentanil and insufficient data to perform this analysis for the other groups.
Cho et al.42 showed a higher incidence of pruritus in the sufentanil group compared with the fentanyl group (P = 0.026). Goodarzi43 showed a higher incidence of pruritus, urinary retention, and respiratory depression in the morphine group compared with both fentanyl and hydromorphone.
The single study that compared morphine versus hydromorphone found a higher incidence of pruritus in those receiving morphine in the first 24 hours postoperatively (44.4% vs 11.5%, respectively; P < 0.01) with no other differences detected.41
Adjuncts to Analgesia
Dyer et al.23 reported significantly higher use of supplementary IV morphine in the epidural sufentanil group compared with the epidural morphine group. In addition, despite detecting no difference in VAS pain scores, the pediatric study by Cho et al.42 showed that patients in the fentanyl group required more rescue analgesia than those receiving sufentanil (6/32 subjects compared with 0/32 subjects; P = 0.012). Otherwise, the use of analgesic adjuncts was similar in the rest of the studies.
Assessment of Publication Bias
Funnel plots were constructed for the analyses presented in Figures 2 to 6 (Supplemental Digital Content 2, http://links.lww.com/AA/A932). Egger’s regression was nonsignificant (P > 0.05) for all these analyses, and all the plots were symmetric with the exception of that corresponding to Figure 3. The symmetry in these plots suggests that there was no publication bias in this review. This asymmetry in the plot for the studies in Figure 3 may have been due to the small number of studies included in that analysis, given the nonsignificant Egger’s regression.
Summary of Key Findings
This novel meta-analysis identified 24 RCTs that compared various continuous infusions of epidural opioids for postoperative analgesia and side effects. Most trials (19/24) compared the clinically relevant opioids morphine, fentanyl, and sufentanil. Overall, there were no clinically significant differences in analgesia. There was an increased rate of PONV (OR = 1.91; 95% CI, 1.14–3.18; and NNT 9.6; 95% CI, 5.9–26.2) and pruritus (OR = 1.64; 95% CI, 0.98–2.76) among patients receiving morphine versus fentanyl. There were no other differences in opioid side effects detected, including respiratory depression, which had a low event rate. Limitations of our analysis included the heterogeneity in surgical populations, differences in outcome reporting among studies, and overall paucity of trials.
It was challenging to combine the data for analgesia in a manner that is both clinically meaningful and not misleading. For instance, a pooled analysis ignoring the type of surgery would have been better powered but at the cost of equating the bony pain of orthopedics to the laboring pain of a parturient.9,45 Furthermore, by virtue of their interaction with different parts of the body, certain surgical procedures give rise to higher rates of side effects, such as the increased risk of respiratory complications after thoracotomy.18 With this rationale in mind, we opted to pool only data involving the same type of surgery and at least 3 trials for adequate power.
Very few individual studies24,25 reported statistically significant differences in analgesia that met the 20-mm VAS threshold suggested by Farrar et al.46,47 for clinically meaningful pain reduction. These isolated findings did not persist when studies were pooled overall and in subgroup analysis by type of surgery. Importantly, the use of analgesic adjuncts was similar between groups (for all but 2 trials),23,42 as was overall total opioid consumption.
Several factors may account for this lack of difference in analgesia. First, all studies reported epidural titration according to the patients’ VAS pain scores using either a bolus or increase in infusion rate of the study solution (by patient-controlled analgesia or staff administration) or analgesic adjunct pro re nata. Although the standard of care, such protocols were problematic when interpreting pain scores because they minimize any difference between the opioids being compared. As an alternative to pain scores, time to first rescue analgesia and total opioid consumption can be considered.48,49 Unfortunately, none of the studies reported the former. With respect to the latter, the small magnitude of the reduced morphine consumption (1.2 mg of morphine equivalent; 95% CI, 0.27–2.18) from the morphine versus fentanyl analysis would not be considered clinically relevant.
Opioid Side Effects
With respect to secondary outcomes, our meta-analysis revealed more similarities than differences among the side effect profiles of opioids being compared. Notably, morphine resulted in more PONV (OR = 1.91; 95% CI, 1.14–3.18) and perhaps pruritus (OR = 1.64; 95% CI, 0.98–2.76) relative to fentanyl. The NNT for the former is <10 patients, which is arguably clinically significant in settings where epidural analgesia is performed in high volume.
The general lack of differences in secondary outcomes is unsurprising given both the dose-dependent nature of opioid side effects and the similarities in total opioid consumption across all studies.13 Nonetheless, morphine still gave rise to more PONV and pruritus than fentanyl, although the latter was not statistically significant. This higher rate of side effects may reflect morphine’s propensity for cephalad migration due to its higher hydrophilicity relative to fentanyl.5 Furthermore, although both fentanyl and morphine can give rise to pruritus directly via opioid receptor stimulation, morphine and its derivatives have an additional mechanism of doing so via inducing histamine release.45
Similarly, the lack of difference in analgesic properties or side effects between sufentanil and fentanyl may be accounted for by their closely shared lipophilic nature, which permits similar penetration of the blood–brain barrier and onset of central nervous system effects (effect-site equilibration time of 6.2 minutes for sufentanil versus 6.8 minutes for fentanyl).50 In contrast, better powered analyses (in other clinical contexts) have found differences in analgesia and side effects between sufentanil and fentanyl that were not observed herein.51 Biologic plausibility for such a difference is suggested by animal studies that show sufentanil has a much larger spinal volume of distribution compared with fentanyl, which instead preferentially distributes to the epidural space and fat.52
A meta-analysis on intrathecal opioids (in combination with local anesthetics) found similar results to the study presented herein, with higher rates of PONV with intrathecal morphine compared with fentanyl.52 This study addressed a different clinical question in that it assessed epidural opioids that were given in combination with local anesthetic and compared head to head with another opioid in a randomized fashion.
There are several limitations to our analysis. Article assessors were not blinded, and we did not include non-English language studies. Only one study published in Italian53 that compared morphine to sufentanil was excluded for this reason. Unfortunately, only the abstract for this trial is accessible.53 Although analgesia was found to be equivalent, the incidence of pruritus and PONV was an unspecified amount higher with morphine (P < 0.0001).53 Without a proper full-text appraisal of this study, it is difficult to ascertain how our results would be affected. We also specified a priori that we would exclude crossover trials due to the potential for inadequate washout of longer-acting opioids, but no trials had to be excluded for this reason.
We considered epidural anesthesia-specific quality factors such as surgery-appropriate level of epidural catheter insertion and confirmation of epidural placement. With regard to the former, “catheter-congruent” epidural anesthesia (relative to the surgical incision level) results in decreased side effects and patient morbidity.54 Although only two-thirds of studies described catheter-congruence, the one-third that did not were mainly concentrated in single studies and had minimal impact on our pooled analyses. With respect to the latter, the majority of studies did confirm placement, and it is hoped that the standard of care was followed in cases where it was left undescribed.
There was a paucity of trials comparing certain opioids and much variability in the surgical populations and outcome reporting among trials. Therefore, some of our analyses (particularly those related to analgesia) were relatively underpowered because data could not be pooled in full. In addition, our subgroup analyses relied on reported equianalgesic ratios for epidural opioids,14 which some authors argue are widely variable.55 Thus, the results of any dependent analyses may differ if repeated using a different set. Furthermore, in contrast to the 10 trials comparing morphine to fentanyl, there was a relative dearth of data comparing other opioids and thus less power to detect any differences. Last, due to differences in the timing of pain assessment across studies, we grouped pain scores by time. These bands of time created difficulty with detecting differences in analgesia at finer time points. The comparison of the same subjects across many time points also may have introduced an inherent autocorrelation in the results, obscuring any differences in analgesia.
In summary, this meta-analysis did not demonstrate any convincing or clinically meaningful differences in analgesia or total opioid consumption among the opioids studied. It is reasonable to conclude that fentanyl gives rise to less PONV and perhaps pruritus compared with morphine. As most trials focused on morphine versus fentanyl and fentanyl versus sufentanil, further studies are required to elucidate any differences in analgesia and side effect profiles between other opioids.
All authors are members of the McMaster Epidural Research Group.
Name: Nayer Youssef, MD.
Contribution: This author was responsible for identifying the clinical question and performing the literature search.
Attestation: Nayer Youssef approved the final manuscript and attests to the integrity of the original data and the analysis reported in this manuscript.
Name: David Orlov, MD.
Contribution: This author was responsible for identifying the clinical question and performing the literature search.
Attestation: David Orlov approved the final manuscript. David Orlov attests to the integrity of the original data and the analysis reported in this manuscript.
Name: Tristan Alie, MSc, MD.
Contribution: This author was responsible for extracting methodological data and identifying the approach for data extraction from graphical sources.
Attestation: Tristan Alie approved the final manuscript and attests to the integrity of the original data and the analysis reported in this manuscript.
Name: Matthew Chong, MD.
Contribution: This author was responsible for extracting numerical data, performing the data analysis, and interpreting the results.
Attestation: Matthew Chong approved the final manuscript and attests to the integrity of the original data and the analysis reported in this manuscript.
Name: Ji Cheng, PhD.
Contribution: This author analyzed the data and reviewed drafts of the manuscript.
Attestation: Ji Cheng approved the final manuscript and attests to the integrity of the original data and the analysis reported in this manuscript.
Name: Lehana Thabane, PhD.
Contribution: This author supervised the analysis and reviewed drafts of the manuscript.
Attestation: Lehana Thabane approved the final manuscript.
Name: James Paul, MD, MSc, FRCPC.
Contribution: This author is the principal investigator and supervised the completion of the project. James Paul will act as the archival author.
Attestation: James Paul approved the final manuscript.
This manuscript was handled by: Spencer S. Liu, MD.
1. Block BM, Liu SS, Rowlingson AJ, Cowan AR, Cowan JA Jr, Wu CL. Efficacy of postoperative epidural analgesia: a meta- analysis. JAMA. 2003;290:2455–63
2. Wu CL, Cohen SR, Richman JM, Rowlingson AJ, Courpas GE, Cheung K, Lin EE, Liu SS. Efficacy of postoperative patient-controlled and continuous infusion epidural analgesia versus intravenous patient-controlled analgesia with opioids: a meta-analysis. Anesthesiology. 2005;103:1079–88
3. Wheatley RG, Schug SA, Watson D. Safety and efficacy of postoperative epidural analgesia. Br J Anaesth. 2001;87:47–61
4. Gourlay GK, Cherry DA, Cousins MJ. Cephalad migration of morphine in CSF following lumbar epidural administration in patients with cancer pain. Pain. 1985;23:317–26
5. Bromage PR, Camporesi EM, Durant PA, Nielsen CH. Rostral spread of epidural morphine. Anesthesiology. 1982;56:431–6
6. de Leon-Casasola OA, Lema MJ. Postoperative epidural opioid analgesia: what are the choices? Anesth Analg. 1996;83:867–75
7. Roy SD, Flynn GL. Solubility and related physicochemical properties of narcotic analgesics. Pharm Res. 1988;5:580–6
8. Haynes RB, Wilczynski N, McKibbon KA, Walker CJ, Sinclair JC. Developing optimal search strategies for detecting clinically sound studies in MEDLINE. J Am Med Inform Assoc. 1994;1:447–58
9. Lowe NK. The nature of labor pain. Am J Obstet Gynecol. 2002;186:S16–24
10. Jadad AR, Moore RA, Carroll D, Jenkinson C, Reynolds DJ, Gavaghan DJ, McQuay HJ. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials. 1996;17:1–12
11. Coppe E, Willaert J. Postoperative analgesia for major abdominal surgery with continuous thoracic epidural infusion of bupivacaine with sufentanil, versus bupivacaine with morphine. A randomized double blind study. Acta Anaesthesiol Belg. 1992;43:131–7
12. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche PC, Ioannidis JP, Clarke M, Devereaux PJ, Kleijnen J, Moher D. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. J Clin Epidemiol. 2009;62:e1–34
13. Chaney MA. Side effects of intrathecal and epidural opioids. Can J Anaesth. 1995;42:891–903
14. Cousins MJ, Bridenbaugh PO, Carr DB, Horlocker TT Cousins and Bridenbaugh’s Neural Blockade in Clinical Anesthesia and Pain Medicine. 20084th ed. China. Philadelphia, PA Lippincott Williams & Wilkins
15. Chen JY, Ko TL, Wen YR, Wu SC, Chou YH, Yien HW, Kuo CD. Opioid-sparing effects of ketorolac and its correlation with the recovery of postoperative bowel function in colorectal surgery patients: a prospective randomized double-blinded study. Clin J Pain. 2009;25:485–9
16. Tramèr M, Moore A, McQuay H. Omitting nitrous oxide in general anaesthesia: meta-analysis of intraoperative awareness and postoperative emesis in randomized controlled trials. Br J Anaesth. 1996;76:186–93
17. Unruh AM. Gender variations in clinical pain experience. Pain. 1996;65:123–67
18. Nomori H, Kobayashi R, Fuyuno G, Morinaga S, Yashima H. Preoperative respiratory muscle training. Assessment in thoracic surgery patients with special reference to postoperative pulmonary complications. Chest. 1994;105:1782–8
19. Manworren RC, Hynan LS. Clinical validation of FLACC: preverbal patient pain scale. Pediatr Nurs. 2003;29:140–6
20. Cohen MM, Cameron CB, Duncan PG. Pediatric anesthesia morbidity and mortality in the perioperative period. Anesth Analg. 1990;70:160–7
21. Kim JY, Lee SJ, Koo BN, Noh SH, Kil HK, Kim HS, Ban SY. The effect of epidural sufentanil in ropivacaine on urinary retention in patients undergoing gastrectomy. Br J Anaesth. 2006;97:414–8
22. Broekema AA, Veen A, Fidler V, Gielen MJ, Hennis PJ. Postoperative analgesia with intramuscular morphine at fixed rate versus epidural morphine or sufentanil and bupivacaine in patients undergoing major abdominal surgery. Anesth Analg. 1998;87:1346–53
23. Dyer RA, Anderson BJ, Michell WL, Hall JM. Postoperative pain control with a continuous infusion of epidural sufentanil in the intensive care unit: a comparison with epidural morphine. Anesth Analg. 1990;71:130–6
24. Valairucha S, Maboonvanon P, Napachoti T, Sirivanasandha B, Suraseranuvongse S. Cost-effectiveness of thoracic patient-controlled epidural analgesia using bupivacaine with fentanyl vs bupivacaine with morphine after thoracotomy and upper abdominal surgery. J Med Assoc Thai. 2005;88:921–7
25. Vallejo MC, Edwards RP, Shannon KT, Kaul B, Finegold H, Morrison HL, Ramanathan S. Improved bowel function after gynecological surgery with epidural bupivacaine-fentanyl than bupivacaine-morphine infusion. Can J Anaesth. 2000;47:406–11
26. Berti M, Fanelli G, Casati A, Lugani D, Aldegheri G, Torri G. Comparison between epidural infusion of fentanyl/bupivacaine and morphine/bupivacaine after orthopaedic surgery. Can J Anaesth. 1998;45:545–50
27. Gedney JA, Liu EH. Side-effects of epidural infusions of opioid bupivacaine mixtures. Anaesthesia. 1998;53:1148–55
28. Tsueda K, Mosca PJ, Heine MF, Loyd GE, Durkis DA, Malkani AL, Hurst HE. Mood during epidural patient-controlled analgesia with morphine or fentanyl. Anesthesiology. 1998;88:885–91
29. Saito Y, Uchida H, Kaneko M, Nakatani T, Kosaka Y. Comparison of continuous epidural infusion of morphine/bupivacaine with fentanyl/bupivacaine for postoperative pain relief. Acta Anaesthesiol Scand. 1994;38:398–401
30. White MJ, Berghausen EJ, Dumont SW, Tsueda K, Schroeder JA, Vogel RL, Heine MF, Huang KC. Side effects during continuous epidural infusion of morphine and fentanyl. Can J Anaesth. 1992;39:576–82
31. Tsui SL, Chan CS, Chan AS, Wong SJ, Lam CS, Jones RD. Postoperative analgesia for oesophageal surgery: a comparison of three analgesic regimens. Anaesth Intensive Care. 1991;19:329–37
32. Chrubasik J, Wüst H, Schulte-Mönting J, Thon K, Zindler M. Relative analgesic potency of epidural fentanyl, alfentanil, and morphine in treatment of postoperative pain. Anesthesiology. 1988;68:929–33
33. Fischer RL, Lubenow TR, Liceaga A, McCarthy RJ, Ivankovich AD. Comparison of continuous epidural infusion of fentanyl-bupivacaine and morphine-bupivacaine in management of postoperative pain. Anesth Analg. 1988;67:559–63
34. Jeon HR, Chae WS, Lee SJ, Lee JH, Cho SH, Kim SH, Jin HC, Lee JS, Kim YI. A comparison of sufentanil and fentanyl for patient-controlled epidural analgesia in arthroplasty. Korean J Anesthesiol. 2011;60:41–6
35. Lutti MN, Vieira JL, Eickhoff DR, de Carli D, de Carvalho MA. Patient controlled analgesia with fentanyl or sufentanil in the postoperative period of knee ligament reconstruction: comparative study. Rev Bras Anestesiol. 2002;52:166–74
36. Wilhelm AJ, Dieleman HG. Epidural fentanyl and sufentanil for intra- and postoperative analgesia. A randomized, double-blind comparison. Pharm World Sci. 1994;16:7–12
37. Cohen S, Amar D, Pantuck CB, Pantuck EJ, Goodman EJ, Widroff JS, Kanas RJ, Brady JA. Postcesarean delivery epidural patient-controlled analgesia. Fentanyl or sufentanil? Anesthesiology. 1993;78:486–91
38. Geller E, Chrubasik J, Graf R, Chrubasik S, Schulte-Mönting J. A randomized double-blind comparison of epidural sufentanil versus intravenous sufentanil or epidural fentanyl analgesia after major abdominal surgery. Anesth Analg. 1993;76:1243–50
39. Yanagidate F, Dohi S. Epidural oxycodone or morphine following gynaecological surgery. Br J Anaesth. 2004;93:362–7
40. Backlund M, Lindgren L, Kajimoto Y, Rosenberg PH. Comparison of epidural morphine and oxycodone for pain after abdominal surgery. J Clin Anesth. 1997;9:30–5
41. Chaplan SR, Duncan SR, Brodsky JB, Brose WG. Morphine and hydromorphone epidural analgesia. A prospective, randomized comparison. Anesthesiology. 1992;77:1090–4
42. Cho JE, Kim JY, Kim JE, Chun DH, Jun NH, Kil HK. Epidural sufentanil provides better analgesia from 24 h after surgery compared with epidural fentanyl in children. Acta Anaesthesiol Scand. 2008;52:1360–3
43. Goodarzi M. Comparison of epidural morphine, hydromorphone and fentanyl for postoperative pain control in children undergoing orthopaedic surgery. Paediatr Anaesth. 1999;9:419–22
44. Jaffe RA, Samuels SI Anesthesiologist’s Manual of Surgical Procedures. 20094th ed. Philadelphia, PA Lippincott Williams & Wilkins
45. Chelly JE, Ben-David B, Williams BA, Kentor ML. Anesthesia and postoperative analgesia: outcomes following orthopedic surgery. Orthopedics. 2003;26:s865–71
46. Farrar JT, Berlin JA, Strom BL. Clinically important changes in acute pain outcome measures: a validation study. J Pain Symptom Manage. 2003;25:406–11
47. Farrar JT, Portenoy RK, Berlin JA, Kinman JL, Strom BL. Defining the clinically important difference in pain outcome measures. Pain. 2000;88:287–94
48. McQuay HJ. Pre-emptive analgesia. Br J Anaesth. 1992;69:1–3
49. Ong CK, Lirk P, Seymour RA, Jenkins BJ. The efficacy of preemptive analgesia for acute postoperative pain management: a meta-analysis. Anesth Analg. 2005;100:757–73
50. Scott JC, Cooke JE, Stanski DR. Electroencephalographic quantitation of opioid effect: comparative pharmacodynamics of fentanyl and sufentanil. Anesthesiology. 1991;74:34–42
51. Davey J, Turner RM, Clarke MJ, Higgins JP. Characteristics of meta-analyses and their component studies in the Cochrane Database of Systematic Reviews: a cross-sectional, descriptive analysis. BMC Med Res Methodol. 2011;11:160
52. Pöpping DM, Elia N, Marret E, Wenk M, Tramèr MR. Opioids added to local anesthetics for single-shot intrathecal anesthesia in patients undergoing minor surgery: a meta-analysis of randomized trials. Pain. 2012;153:784–93
53. Gianferrari P, Clara ME, Borghi B, Marzullo A, Voltolina M, De Paolis P, Montone N. [Sufentanil vs morphine combined with ropivacaine for thoracic epidural analgesia in major abdominal surgery]. Minerva Anestesiol. 2001;67:155–9
54. Brown AK, Christo PJ, Wu CL. Strategies for postoperative pain management. Best Pract Res Clin Anaesthesiol. 2004;18:703–17
55. Pereira J, Lawlor P, Vigano A, Dorgan M, Bruera E. Equianalgesic dose ratios for opioids. A critical review and proposals for long-term dosing. J Pain Symptom Manage. 2001;22:672–87
Supplemental Digital Content
© 2014 International Anesthesia Research Society