Secondary Logo

Journal Logo

A Randomized, Double-Blind, Placebo-Controlled Trial of Epidural Morphine Analgesia After Vaginal Delivery

Macarthur, Alison, MD, MSc*; Imarengiaye, Charles, MD; Tureanu, Luminita, MD; Downey, Kristi, MSc*

doi: 10.1213/ANE.0b013e3181c30f78
Obstetric Anesthesiology: Research Reports

BACKGROUND: Pain after vaginal delivery can interfere with the activities of daily living. We hypothesized that epidural medication administered after delivery would be of benefit for acute postpartum pain management. The objective of this study was to assess whether epidural morphine after vaginal delivery would reduce the analgesic requirements for perineal pain.

METHODS: This randomized, double-blind, placebo-controlled trial included 228 parturients who received epidural morphine, 2.5 mg, or epidural saline within 1 h of delivery. The primary outcome was the proportion of women who received additional analgesics in the first 24 h postpartum. We also evaluated the time to first request for analgesia, pain and satisfaction scores, and the incidence of side effects due to epidural morphine.

RESULTS: The majority of the 228 women participating in the study were Caucasian, primiparous patients >30 yr old. The proportion of women requiring additional analgesics was less among those who received epidural morphine (8 of 113; 7%) compared with saline (37 of 115; 32%), regardless of the degree of perineal trauma (RR = 0.22, 95% CI: 0.12–0.41). The relative risk reduction in receiving additional analgesics for primiparous patients receiving epidural morphine compared with saline was 0.25 (95% confidence interval [CI]: 0.13–0.49) and for multiparous women was 0.12 (95% CI: 0.02–0.63). The time to first request for analgesics was later for those who received morphine (mean 22.9 h, 95% CI: 22.2–23.7) compared with saline (mean 18.9 h, 95% CI: 17.4–20.4) (P < 0.0001). The side-effect incidence (pruritus, nausea and vomiting, and sedation) was not different between the 2 groups.

CONCLUSION: There was a 78% reduction in analgesic requirements in women given epidural morphine after vaginal delivery compared with placebo for both primiparous and multiparous patients. Women who receive epidural labor analgesia for vaginal deliveries and stay in the hospital for 24 h after delivery may benefit from postpartum administration of epidural morphine.

Published ahead of print November 12, 2009

From the *Department of Anesthesia, University of Toronto, Ontario, Canada; †Department of Anesthesiology, University of Benin Teaching Hospital, Benin, Nigeria; and ‡Department of Anesthesia, Northwestern University, Chicago, Illinois.

Accepted for publication August 25, 2009.

Published ahead of print November 12, 2009

Supported by Physicians Services Incorporated Quarterly Funding Competition.

Presented at the 35th annual meeting of the Society for Obstetric Anesthesia and Perinatology (SOAP), May 6, 2005, Palm Desert, CA.

Address correspondence and reprint requests to Alison Macarthur, MD, MSc, Department of Anesthesia, Mount Sinai Hospital, 600 University Ave., Room 1514, Toronto, ON, Canada M5G 1X5. Address e-mail to

In a previous study from our institution, we found that even women with little perineal trauma after vaginal delivery report significant interference in common daily activities, such as sitting and urinating, due to postpartum perineal pain.1 Thirty-seven percent of women with no perineal trauma or first-degree perineal lacerations described their perineal pain on the first day postpartum as distressing, horrible, or excruciating.1 The use of systemic, multimodal analgesics in the postpartum period is increasing as physicians recognize the safety and utility of drugs such as nonsteroidal antiinflammatory drugs and acetaminophen.2 However, systemic opioids are often required to achieve satisfactory analgesia, but at our institution these were only considered for women with extensive perineal trauma, such as third- or fourth-degree lacerations. Before this study, our institution's practice for analgesia after vaginal delivery was intermittent, administered-as-requested oral, IM, or IV medications.

As a follow-up to our previous study of acute postvaginal perineal pain, we hypothesized that all women using epidural labor analgesia would benefit from a single dose of postpartum epidural morphine. The basis of our hypothesis stemmed from the success of epidural and intrathecal morphine for postcesarean analgesia, and its routine use in North American obstetric anesthesia units.3 Epidural and intrathecal administration of morphine has the benefit of profound, prolonged analgesia at relatively low doses. The benefits compared with systemic opioids have been demonstrated in many postoperative analgesic settings.4,5 The objective of this study was to determine whether a single dose of epidural morphine given soon after a vaginal delivery would reduce postpartum perineal pain and reduce the use of IV or oral opioid analgesics in the early postpartum period.

Back to Top | Article Outline


This single-center, randomized, double-blind, clinical trial received IRB approval and was conducted at Mount Sinai Hospital, Toronto. Women eligible for inclusion in the study were those who delivered vaginally and who had received effective epidural analgesia for labor and delivery. Their informed, written consent to be involved in the study was obtained before delivery. Women receiving combined spinal-epidural analgesia for labor were excluded. The labor analgesia was not standardized for patients in the study but usually consisted of the following practice. Epidural analgesia was initiated with epidural bupivacaine 0.1%–0.125% and 50–100 μg of fentanyl, followed by a patient-controlled epidural analgesic with bupivacaine 0.0625% and fentanyl 2 μg/mL. Settings for the patient-controlled epidural analgesic pump included a background infusion between 8 and 10 mL/h, demand bolus of 5 mL, and lockout interval of 5–10 min. The infusion was continued until perineal suturing was finished.

The randomization schedule was stratified for primiparous and multiparous study subjects using a computer random number generator to select random permuted blocks of 8. The pharmacy department independently created the schedule and prepared sterile syringes of either epidural morphine or epidural saline in batches of 10 with assigned random numbers. Within 1 h of delivery, study subjects were randomized to receive 1 of 2 identical-appearing syringes epidurally with either preservative-free morphine 2.5 mg diluted to 10 mL, or 10 mL of preservative-free saline. The epidural catheter was removed just before the patient's transfer to the postpartum ward.

A standard anesthesia order set with monitoring, side-effect therapy, and analgesic guidelines was used for 24 h after discharge to the postpartum ward. Patients were monitored every hour for the first 12 h and every 2 h for the second 12 h by the nursing staff for sedation level, pruritus and nausea/vomiting, and visual analog scale (VAS) score for pain. All patients received scheduled acetaminophen (1000 mg every 6 h) and ibuprofen (400 mg every 6 h). In addition, women could request additional analgesics (opioids) for breakthrough perineal pain. The additional opioids included oral codeine (30–60 mg every 6 h as requested) or IM morphine (5–10 mg every 4 h as requested). The choice of supplemental opioid was made in cooperation with the patient and her nurse. Assessment of supplemental analgesia requirements was a 2-step process: patient and nursing opioid records were assessed at 24 h postpartum for counts of additional analgesics, and all patients were asked whether they had received supplemental medications beyond the acetaminophen and ibuprofen.

The primary outcome was the proportion of women in each group who received opioid analgesia in the first 24 h postpartum. Secondary outcomes included maternal VAS pain score at time of first request, using a 10-cm line anchored at 0 with the words “no pain” and the other end “worst pain imaginable,” maternal satisfaction with perineal analgesia during the first 24 h postpartum (5-point Likert scale), time to first request for supplemental analgesia after delivery, and the incidence of side effects in the first 24 h postpartum, including pruritus, nausea and vomiting, drowsiness, and urinary retention. Study participants were asked by nursing staff every 4 h whether they had pruritus, nausea or vomiting (yes or no), and were assessed for presence of drowsiness. Drowsiness was defined as respiratory rate <10 breaths/min or visual presence of moderate sedation or somnolence as assessed by the participant's nurse. Urinary retention was defined by the need for catheterization. Women with indwelling catheters placed immediately after delivery were excluded from this analysis. Subjects were telephoned 1 wk postpartum to assess whether any nonopioid or opioid analgesics were taken since hospital discharge.

Information on potential confounding variables was collected, including parity, gestational age, type of vaginal delivery (spontaneous or operative vaginal delivery, indicating use of vacuum or forceps), degree of perineal trauma as reported in the hospital delivery record, neonatal birth weight, maternal age, maternal body mass index, and self-identified cultural background (5-category scale of heritage used by Census of Canada). Perineal trauma was classified as minor and major trauma. Minor trauma was defined as no lacerations, periurethral lacerations, and first-degree lacerations. Major trauma was defined as second-, third-, and fourth-degree lacerations and episiotomies.

Sample size calculation was based on the incidence of women requiring opioid analgesia in the first 24 h postpartum who had no identified perineal trauma. This incidence was estimated at 63% in a previous study from our institution,1 and we wished to detect a 30% reduction in opioid analgesic use in the group of women who received epidural morphine compared with the epidural saline group (absolute difference of 19%). Using sample size calculations based on proportions using a 2-sided test, and accepting a type I error rate of 0.05 and a type II error rate of 0.20, 107 women were required in each group, with planned recruitment of 230 women for the study.

Adequacy of randomization was assessed by examining for any differences in the distribution of potential confounding variables in the 2 groups using the Student's t-test and Fisher's exact test for differences in means and proportions, respectively. A crude relative risk (RR) estimate for the analgesic-sparing effect of epidural morphine compared with placebo was calculated with 95% confidence intervals (CIs) as well as an adjusted estimate, adjusting for degree of perineal trauma and parity. Fisher's exact test for difference in proportions was used to compare the incidence of the primary and secondary outcomes. Time to first request for supplemental opioid analgesia was assessed using survival analysis, and other continuous variables were compared using the Student's t-test for difference in means. Significance level for the primary outcome was defined as P < 0.05, whereas secondary outcome P values were adjusted using Bonferroni corrections for multiple testing. All analyses were done using the program STATA (version 8.2, StataCorp, College Station, TX).

Back to Top | Article Outline


After the successful initiation of a labor epidural analgesia, 1110 women were approached and given information regarding the study between March 2002 and November 2004. Consent to participate in the study was obtained before delivery. Figure 1 provides the CONSORT flow diagram on the participation of women approached for the study and reasons for nonparticipation in the study.6 Less than 50% of the women gave consent to participate in the study if they delivered vaginally. The majority of nonparticipants declined participation because of reticence in receiving an opioid prophylactically. Two hundred twenty-eight women who delivered vaginally were randomized to receive epidural morphine (n = 113) or saline (n = 115). There was no difference in the demographic variables between groups (Table 1). Women participating in the study were primarily Caucasian primiparous patients, with an average maternal age of >30 yr. One-third of study participants had an operative vaginal delivery and 169 of 228 (74%) women had major perineal trauma. The mean maternal VAS pain score just before administration of the study drug was <2, indicative of residual epidural analgesia after delivery.

Figure 1.

Figure 1.

Table 1

Table 1

The number of women who required additional analgesics in the first 24 h after vaginal delivery was 8 of 113 in the epidural morphine group compared with 37 of 115 in the epidural saline group (P < 0.001). This is a 78% reduction in the risk of opioid analgesia requirement among women who received epidural morphine compared with saline (RR 0.22, 95% CI: 0.12–0.41). This risk reduction with the administration of epidural morphine was consistent among both primiparous (RR 0.25, 95% CI: 0.13–0.49) and multiparous women (RR 0.12, 95% CI: 0.02–0.63) (Table 2). Among the women with differing degrees of perineal trauma, the effect of analgesic sparing seemed to remain consistent. For women with minor perineal trauma, the RR reduction in the requirement for opioid analgesics was 85% (RR 0.15, 95% CI: 0.03–0.86), and among those with major trauma, the risk reduction was 73% (RR 0.27, 95% CI: 0.14–0.50). RR estimates for reduction of epidural morphine in analgesic requirements remained unchanged when adjusted for degree of perineal trauma and parity (adjusted odds ratio 0.16, 95% CI: 0.07–0.36).

Table 2

Table 2

The VAS pain scores at time of first request for supplemental opioid were similar between the 2 groups. The time to first request for an additional opioid analgesia was prolonged among women receiving epidural morphine by 4 h (P < 0.001). The average time to request supplemental opioid analgesia among women who requested additional analgesia was 8 h after delivery. Estimation of maternal satisfaction with perineal pain management was completed at 24 h postpartum by 209 of the 218 women. Maternal satisfaction was not different between the groups.

The incidence of side effects did not differ between the groups. The most common side effect was urinary retention, among 14% of the study population. The proportion of women reporting pruritus to their nurses was 9% overall. Nausea, defined as the number of women receiving antiemetics in the first 24 h, occurred in 7% of all women. Only 1% of the study population met the definition of drowsiness.

During the first week postpartum, analgesic use was evaluated among 203 women. Among women receiving epidural morphine, 81% (79 of 98) used analgesics, including opioids and nonopioids, compared with 69% (72 of 105) of women receiving epidural saline (P = 0.06).

Back to Top | Article Outline


Women receiving 2.5 mg of epidural morphine after a vaginal delivery reduced their use of opioid analgesics by 78% compared with women receiving placebo in the first 24 h postpartum. We believe that the benefit of epidural morphine was underestimated because our control group received improved postpartum perineal pain management compared with what occurs normally on our postpartum wards. The entire study population received scheduled doses of acetaminophen and ibuprofen during the first 24 h, which has not been routine in the past. Before the study, patients received a bedside pack of acetaminophen and ibuprofen for self-administration. Therefore, if we had used our standard of postpartum pain care for the control group, the risk reduction with morphine use would likely have been even greater.

Although vaginal deliveries seem inherently to be associated with less pain than cesarean delivery, recent evidence has contradicted this belief. Eisenach et al.7 have demonstrated in a prospective study of 1288 women that acute postpartum pain is associated with development of persistent pain at 8 weeks' postpartum, regardless of the mode of delivery. The authors identified an independent risk (RR 2.5) for development of persistent postpartum pain; if severe, acute postpartum pain was present. It seems reasonable to apply principles of analgesic management to all women who deliver, because the individual's pain response to trauma injury may not be correlated with degree of trauma. It remains to be proven whether limiting or eliminating severe, acute postpartum pain can reduce the development of chronic pain syndromes. Beyond providing excellent intrapartum pain relief with epidural analgesia, the availability of excellent postpartum analgesia may be an additional benefit available to women who choose epidural labor analgesia.

To our knowledge, there are 4 other randomized trials evaluating epidural morphine for postvaginal delivery pain,8,9*† 2 of which are only reported as abstracts.*†Table 3 outlines the characteristics and results of 4 studies, along with this study. Women were randomized to receive doses of 1–4 mg of epidural morphine or epidural saline. In our study, the choice of a 2.5-mg dose of epidural morphine was made based on evidence documenting the effectiveness of this dose for postcesarean analgesia10 and the early work by Macdonald and Smith8 in postepisiotomy pain. The study of Macdonald and Smith8 evaluated only women with episiotomies because the investigators thought that this degree of perineal trauma necessitated additional analgesia. The 2 largest studies8,9 and our study had the highest quality (≥4/5 Jadad score11) and found a similar drug-sparing effect in the epidural morphine compared with saline groups. The 2 small studies*† reported only as abstracts did not find a reduction in maternal VAS pain scores during the first 24 h postpartum. This may have occurred as a result of insufficient sample size or differing supplemental analgesic regimes.

Table 3

Table 3

The strengths of our study include the large sample size (other studies enrolled <125 patients) and stratification of parity to ensure adequate distribution in both groups. Our earlier work demonstrated that between postpartum Day 1 and Day 7, multiparous women experienced 10%–30% less perineal discomfort than that of primiparous women.1 This discrepancy could potentially explain the lack of efficacy of 2-mg epidural morphine in the study of Nunlee et al.† In their study, there was an unequal distribution of primiparous patients in the morphine group (56%) compared with the control group (8%). Primiparous women who received epidural morphine had similar supplemental analgesic and pain scores compared with the multiparous women who received epidural saline. We demonstrated that women with varying degrees of perineal trauma still benefit from the use of epidural morphine; however, only 26% of our population had minor trauma. Five patients in the minor perineal trauma group (n = 59) required additional analgesics in the first 24 h postpartum; however, all of the patients were in the saline group. It remains to be confirmed by a further larger study whether patients with minor perineal trauma benefit as much as patients with more extensive trauma.

Our study would have been improved by evaluating functional activities of daily life, such as ability to sit and urinate in the first day. Qualitative assessment of the patient's perineal pain may have revealed further differences in the first day experiences. However, many patients delivered during evening and nighttime hours and were discharged from the hospital within 24–48 h postpartum. Ensuring time available to question patients in the first 24 h postpartum can be challenging among those delivering vaginally. Maternal satisfaction did not seem to be different between the 2 groups; however, the tool used to assess satisfaction with analgesia was crude. Perhaps a multidimensional satisfaction scale or assessing satisfaction over a longer period of the postpartum recovery would have identified differences between the groups, if there were any.

Generalizing our findings to other labor and delivery populations requires a comment. Although only 446 women of 1110 approached agreed to participate in the study, the remaining 58% declined for reasons that would be unlikely to diminish the effect of epidural morphine, should it have been given. Women who declined to participate did so because of a concern over receiving additional medication, particularly an opioid. We do not have a biological reason why the epidural morphine would have worked less effectively in this group of nonparticipants. Our study also had a higher rate of operative vaginal delivery (32%) than the general Ontario delivery population, which has been estimated at 15%.12 However, our study population is different than the general delivery population, in that they had selected epidural analgesia for labor. Whether causal or not, epidural analgesia has been associated with an increased risk of operative delivery. Other factors associated with increased incidence of an operative vaginal delivery were present in our study population including older age participants, who were more likely to be primiparous and Caucasian.13 Our overall institutional cesarean delivery rate during the study period was approximately 34%. We await investigations by other labor and delivery units on this analgesic modality to make a final assessment of the generalizability of our study.

The use of epidural morphine in the postpartum period requires an increased period of maternal monitoring for potential respiratory depression. The incidence of respiratory depression with the use of epidural or intrathecal morphine is lower in obstetric patients compared with nonobstetric patients and has been estimated between 0.07% and 0.9%.14 Our current institutional practice requires monitoring of the mother's sedation level and respiratory rate for 24 h after epidural or intrathecal morphine administration. Other side effects of epidural or intrathecal morphine include pruritus, nausea and vomiting, and urinary retention. We may not have detected a difference because the incidence of all side effects has decreased with reduction of the epidural dose administered in the past 15 yr. Original administration of epidural morphine began with doses between 3 and 5 mg, and the incidence of pruritus was between 40% and 80% and nausea/vomiting between 20% and 60%.15 Although our study did not demonstrate an increased incidence of side effects, this is likely attributable to an inadequate sample size to assess for differences among infrequently occurring events. Further study is required to assess whether the analgesic-sparing effect of epidural morphine is worth any increased side-effect risk such as bladder catheterization or prolonged hospitalization secondary to urinary retention.

A new sustained-release morphine formulation of morphine has become available in the United States for postoperative analgesia.16 This product uses proprietary-protected lipid technology that surrounds conventional morphine sulfate, enabling a sustained release of morphine for up to 48 h. The reports on use of this drug have been limited to fewer than 200 patients after cesarean delivery; however, it seems to offer improvements after the first 12 h over conventional epidural morphine. This product has not been evaluated in postvaginal delivery patients and may be hindered by the requirement for monitoring side effects for 48 h after administration. Discharge from hospital often occurs for postvaginal delivery patients between 36 and 48 h postpartum.

As a result of our study findings, women at our institution who deliver vaginally now receive scheduled doses of acetaminophen and ibuprofen until discharge, and women with epidural labor analgesia are offered a single dose of postpartum epidural morphine. We have implemented a standard order set for these patients with monitoring variables, side-effect treatment, and a link to obstetrician postpartum orders when the 24-h monitoring period ends. The ideal dose for postvaginal analgesia has yet to be evaluated, and the confirmation of these findings in populations with younger patients and fewer operative deliveries has yet to be examined.

Back to Top | Article Outline


1. Macarthur A, Macarthur C. Incidence, severity and determinants of postpartum perineal pain: a prospective cohort study. Am J Obstet Gynecol 2004;191:1199–204
2. Pan PH. Post cesarean delivery pain management: multimodal approach. Int J Obstet Anesth 2006;15:185–8
3. Lavand'homme P. Postcesarean analgesia: effective strategies and association with chronic pain. Curr Opin Anaesthesiol 2006;19:244–8
4. Cade L, Ashley J, Ross W. Comparison of epidural and intravenous opioid analgesia after elective cesarean section. Anaesth Intensive Care 1992;20:41–5
5. de Leon-Casasola OA, Lema MJ, Karabella D, Harrison P. Postoperative myocardial ischemia: epidural versus intravenous patient-controlled analgesia. A pilot project. Reg Anesth 1995;20:105–12
6. Moher D, Schulz KF, Altman D. The CONSORT statement: revised recommendations for improving the quality of reports of parallel-group randomized trials. JAMA 2001;285:1987–91
7. Eisenach JC, Pan PH, Smiley R, Lavand'homme P, Landau R, Houle TT. Severity of acute pain after childbirth, but not type of delivery, predicts persistent pain and postpartum depression. Pain 2008;140:87–94
8. Macdonald R, Smith PJ. Extradural morphine and pain relief following episiotomy. Br J Anaesth 1984;56:1201–5
9. Goodman SR, Drachenberg AM, Johnson SA, Negron MA, Kim-Lo SH, Smiley RM. Decreased postpartum use of oral pain medication after a single dose of epidural morphine. Reg Anesth Pain Med 2005;30:134–9
10. Palmer CM, Mogaimi WM, Van Maren G, Alves DM. Postcesarean epidural morphine: a dose-response study. Anesth Analg 2000;90:887–91
11. Jadad AR, Moore RA, Carroll D, Jenkinson C, Reynolds DJM, Gavaghan DJ, McQuay HJ. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials 1996;17:1–12
12. Public Health Agency of Canada. Canadian Perinatal Health Report, 2008 Edition. Ottawa: Public Health Agency of Canada, 2008:83
13. O'Leary CM, de Klerk N, Keogh J, Pennell C, de Groot J, York L, Mulroy S, Stanley FJ. Trends in mode of delivery during 1984–2003: can they be explained by pregnancy and delivery complications? BJOG 2007;114:855–64
14. Carvalho B. Respiratory depression after neuraxial opioids in the obstetric setting. Anesth Analg 2008;107:956–61
15. Fuller JG, McMorland GH, Douglas MJ, Palmer I. Epidural morphine for analgesia after caesarean section: a report of 4880 patients. Can J Anaesth 1990;37:636–40
16. Carvalho B, Roland LM, Chu LF, Campitelli VA, Riley ET. Single-dose, extended-release epidural morphine (DepoDur) compared to conventional epidural morphine for post-cesarean pain. Anesth Analg 2007;105:176–83

*Ascanio RS, Mahaney G, Sarna MC, Soni AK, Oriol NE. A single dose of epidural morphine alleviates post-partum perineal pain for at least one week. SOAP 1997:A153.
Cited Here...

†Nunlee WY, Perry PM, Lawal AH, Ivankovich AD. Does a single dose of epidural morphine provide extended analgesia after vaginal delivery? Anesthesiology 2000:A160.
Cited Here...

© 2010 International Anesthesia Research Society