Current Issue Previous Issues Published Ahead-of-Print CME Subjects Timely Topics Translations Podcasts For Authors Journal Info
Skip Navigation LinksHome > August 2002 - Volume 95 - Issue 2 > A Double-Blinded, Randomized Comparison of Intrathecal and E...
Anesthesia & Analgesia:
doi: 10.1213/00000539-200208000-00037

A Double-Blinded, Randomized Comparison of Intrathecal and Epidural Morphine for Elective Cesarean Delivery

Sarvela, J. MD, PhD; Halonen, P. MD; Soikkeli, A. MD; Korttila, K. MD, PhD, FRCA

Free Access
Article Outline
Collapse Box

Author Information

Department of Anaesthesia and Intensive Care, Helsinki University Central Hospital, Helsinki, Finland

April 5, 2002.

Address correspondence and reprint requests to Johanna Sarvela, MD, PhD, Department of Anaesthesia and Intensive Care, Helsinki University Central Hospital, PO Box 140, Haartmaninkatu 2, FIN-00029 HYKS, Finland. Address e-mail to

Collapse Box


We randomized 150 parturients into a double-blinded trial to receive intrathecal (IT) 100 μg (IT 100 group) or 200 μg (IT 200 group) or epidural 3 mg (Epidural group) of morphine for elective cesarean delivery with a combined spinal/epidural technique. The patients additionally received ketoprofen 300 mg/d. Postoperative pain relief and side effects were registered every 3 h up to 24 h, and all patients were interviewed on the first postoperative day. Pain control was equally good, but the parturients in the IT 100 group requested rescue analgesics more often compared with the other groups (P < 0.05). Itching was a common complaint and was reported by 74% of the parturients in the Epidural group and 65% and 91% in the IT 100 and IT 200 groups, respectively (P < 0.01). Medication for itching was requested by 44%, 24%, and 45% of the patients, respectively (P < 0.05). There was no difference in postoperative nausea or vomiting. The pain relief was perceived as good by >90% of the patients in all groups. In conclusion, because of the decreased incidence of and lesser requirements of medication for itching, IT morphine 100 μg with ketoprofen is recommended in cesarean deliveries. Rescue analgesics nevertheless need to be prescribed.

Spinal opioids are often used for pain relief after cesarean delivery. Neuraxial analgesia has proven to be an effective means of achieving postoperative pain control, with minimal exposure to opioids in lactating parturients. Morphine, which is the only opioid approved by the Food and Drug Administration for intrathecal (IT) use, has been used both epidurally and IT for this purpose. When administered IT, morphine may achieve a longer duration of analgesia compared with epidural administration (1). Both routes, however, are associated with a frequent incidence of side effects, including itching and postoperative nausea, which are the most common causes of dissatisfaction with postoperative analgesia (2). A study comparing two different doses of IT morphine (100 and 200 μg) and epidural morphine was performed to compare the level of analgesia and the profile of side effects. Combining nonsteroidal antiinflammatory analgesics with spinal opioids augments analgesia via a reduction of uterine cramping and an antiinflammatory effect (3,4). We tested the hypothesis that in patients receiving ketoprofen, IT morphine 100 μg provides equal analgesia but is associated with less itching and nausea or vomiting than IT morphine 200 μg or epidural morphine 3 mg.

Back to Top | Article Outline


After the approval of the institutional ethical committee was obtained, 162 elective parturients scheduled for elective cesarean delivery were enrolled in the study over a 1.5-yr period from the beginning of 2000. Written, informed consent was obtained from all patients. Women with a major systemic disease, those who had fetuses with known anomalies, or those who were allergic to drugs included in the study protocol were excluded. Randomization was stratified, with an equal number of multiparous and primiparous parturients in each group, by use of computer-generated random numbers.

A combined spinal/epidural (CSE) needle-through-needle technique with a 16-gauge Tuohy epidural needle and a Whitacre 26-gauge spinal needle was used at the L2-3 or L3-4 interspace. After prehydration with 1 L of acetated Ringer’s solution, the parturients received 8–9 mg of plain bupivacaine 0.5% (1.6–1.8 mL), fentanyl 15 μg (0.3 mL), and either 100 μg (Group IT 100) or 200 μg (Group IT 200) of preservative-free morphine in saline or saline (Group epidural) for spinal anesthesia. The volume of the study solution was 0.25 mL, and the final volume of spinal solution was 2.15–2.35 mL. The study drugs were prepared by an anesthesiologist who was not participating in the care or evaluation of the patients. After injection of the spinal anesthetic, an epidural catheter was inserted. Initially the patient was positioned on her right side with a pillow under her head, and she was then turned to a supine position. The operating table was immediately tilted at least 20° to the left, and a urinary catheter was inserted. If the anesthesia level did not reach T5 at 15 min, the epidural catheter was tested with lidocaine 2% with epinephrine, and epidural top-ups with the same local anesthetic were administered as needed. Oxygen 4 L/min was administered through a face mask until delivery. IV boluses of 5–10 mg of ephedrine and additional IV fluids were administered to treat hypotension, which was defined as systolic blood pressure <95 mm Hg or a decrease in systolic pressure of >20% from the baseline value. The surgical technique was uniform for all patients and included exteriorization of the uterus. At delivery, blood samples were collected from the umbilical vein for blood gas analysis. Five international units of oxytocin were administered IV after delivery. Apgar scores were assessed at 1 and 5 min.

Maternal electrocardiogram and oxygen saturation were continuously monitored, and noninvasive blood pressure was measured from the right arm every 1 or 2 min until the birth of the child and thereafter every 5 min until the patient was moved to the postanesthesia care unit (PACU). Perioperative pain was estimated with a visual analog score (VAS) from 0 (no pain) to 10 (worst possible pain).

Ninety minutes after the induction of spinal anesthesia, an epidural test dose of 3 mL of 2% lidocaine with epinephrine 5 μg/mL was administered, followed by either 10 mL of saline (for Groups IT 100 and IT 200) or 3 mg of morphine in 10 mL of saline (Group epidural) in the PACU. The study drugs were prepared by an anesthesiologist not participating in the care or evaluation of the patient. At the same time, 100 mg of ketoprofen in saline was given IV and thereafter every 8 h orally or rectally. Fentanyl 50 μg IV was used during surgery and in the PACU, and oxycodone 0.1 mg/kg IM was used on the ward as a rescue analgesic if it was requested and the VAS score for pain was >3.

Nurses assessed the patients up to 24 h after the induction of spinal anesthesia—every 3 h during their stay in the PACU and on the ward. Respiratory frequency and the adequacy of respiration and sedation (0, none; 1, mild sedation; 2, barely arousable) were assessed hourly for 18 h. Pain and possible side effects were recorded every 3 h as follows. Pain was estimated (at rest during deep breathing) as described previously; itching (0, none; 1, some pruritus; 2, disturbing pruritus) and nausea and vomiting (0, none; 1, nausea; 2, severe nausea or vomiting) were evaluated unless the patient was asleep. The investigators were blinded to the study group.

The nature of the side effects was explained to the patients, and they were told about the treatment. Side effects were treated on request only. Nausea and vomiting were treated with IV droperidol 0.5 mg or IV metoclopramide 10 mg, and itching was treated with hydroxyzine hydrochloride 25 mg and, if this was inadequate, with naloxone 0.04 mg repeatedly IV or 0.4 mg IM as necessary.

The patients were visited during the first postoperative day, at which time the postoperative pain was rated as none, some, or considerable. Maximal VAS score while moving during the first 24 h after surgery was registered. The efficacy of pain relief was categorized into three groups (unsuccessful, satisfactory, and good), as was the severity of the side effects (none, present but acceptable, and unbearable). The patients were also asked whether they would like to have similar pain treatment for future cesarean delivery.

For statistical analysis, the Kruskal-Wallis test and χ2 test (cross-tabs) for quantitative and categorical data, respectively, were used with SPSS Version 9.0 (SPSS, Inc., Chicago, IL). When the results were statistically significant, an intergroup analysis with the Mann-Whitney U-test and χ2 tests was applied. P < 0.05 was considered significant. The demographic data are expressed as mean ± sd. The sample size was determined to be 48 patients per group in this study to detect a decrease of itching from 80% to 50% at the 0.05 significance level with 90% power.

Back to Top | Article Outline


The groups were comparable with each other (Table 1). Four patients (three in the IT 200 group and one in the IT 100 group) were excluded from the study because more than 6 mL of epidural supplementation was needed to achieve adequate anesthesia. No patient required general anesthesia because of failed block. There was no difference in the requirements for ephedrine or fentanyl during surgery. One patient in each group reported a VAS of >3 during the surgery. One patient in the IT 200 group needed fentanyl in the PACU because of suspicion of uterine atony and concomitant painful manual pressure exerted on the uterus (Table 2). Postoperative bleeding requiring reoperation did not occur in any of the patients.

Table 1
Table 1
Image Tools
Table 2
Table 2
Image Tools

At 3 h after injection of spinal anesthesia (in the PACU), less itching was reported by patients in the Epidural group than in either IT group (P < 0.05 versus IT 100 and P < 0.01 versus IT 200) (Table 2). During the postoperative follow-up on the ward up to 24 h after surgery, there was a statistically significant difference among the three groups in itching (P < 0.01) and in the need for medication for itching (P < 0.05). When the two IT groups were compared with each other, there were fewer complaints of itching (P < 0.01) and less medication was requested for itching (P < 0.05) in the IT 100 group. In intergroup comparisons between the Epidural and IT 200 groups, more itching occurred in the latter group (P < 0.05), but there was no difference in the medical treatment of itching. The Epidural and IT 100 groups did not differ statistically in the incidence of itching (Fig. 1).

Figure 1
Figure 1
Image Tools

Postoperative nausea or vomiting (PONV) occurred in 16% (IT 100), 26% (Epidural), and 28% (IT 200) of patients (not significant). There was no difference in the incidence of request for medication for PONV (Fig. 1). No patient exhibited respiratory depression (respiratory rate <10 breaths/min) or had a sedation score of 2 (barely arousable).

There was no difference in the highest VAS score for pain at rest during the first 21 h or during coughing or moving in the first 24 h after surgery. The only difference in pain sensations was at 24 h, when better pain control was achieved in the Epidural group than in either IT group (P < 0.05;Fig. 2). However, during the stay in the ward, there was a significant difference in the need for rescue analgesics among the three groups (P < 0.05). In intergroup comparisons, the parturients in the IT 100 group required rescue analgesics more often than those in the IT 200 group (P < 0.05), but the difference between the IT 100 group and the Epidural group was not statistically significant. The mean time to first request for rescue analgesia was 15 h (n = 5), 13 h (n = 10), and 5 h (n = 2) in the Epidural, IT 100, and IT 200 groups, respectively (not significant).

Figure 2
Figure 2
Image Tools

Satisfaction with pain control was good in more than 90% of the parturients in all groups. None of the patients reported inadequate pain relief. Similarly, the side effects were experienced either as none or moderate by more than 90% of the patients. Intolerable side effects were reported by one patient in the IT 200 group (itching and PONV) and by three in the Epidural group (two had intolerable itching, and one had PONV). One patient each in the Epidural (PONV) and the IT 200 (PONV and itching) groups and two in the IT 100 group (one wished for better pain control and one experienced PONV) wanted to try some alternative pain relief in the future (Table 3).

Table 3
Table 3
Image Tools
Back to Top | Article Outline


CSE anesthesia was used because of its reliability and flexibility (5), enabling the use of small amounts of local anesthetics for spinal anesthesia and potential supplementation of spinal anesthesia with epidural top-ups. Four patients were excluded from the IT groups when more than 6 mL of epidural supplementation was required to achieve an adequate level of anesthesia (T5 dermatome) (6). This was done because it was assumed that with large epidural anesthetic requirements, all of the IT morphine may not have reached the IT space. In the Epidural group, however, when this happened there was no doubt that the dose of morphine was given epidurally because the catheter was tested with supplemental anesthesia.

With multimodal pain control, a relatively small dose of epidural morphine (3 mg) is needed (7). In our previous study, the small dose of bupivacaine used (8–9 mg) was shown to be adequate as part of the CSE technique (6). In that study, the failure rate was 5%.

Adequate pain relief was obtained for the first 24 hours with both methods and with three different doses of intraspinal morphine. However, rescue analgesics were needed more often in the IT 100 group, in which every fifth patient needed rescue medication even when ketoprofen was administered regularly during the study period. Therefore, in this study, analgesia with IT morphine was dose related. Others have not found this (8–10). Although good pain relief has been achieved in many studies with much smaller amounts of morphine (4,11), wide variation in analgesic requirements among parturients undergoing cesarean delivery has also been noted. In many of those studies, the group sizes have been significantly smaller than in our study. It is also possible that there is wider variation in the need for analgesics after cesarean delivery than after other surgical procedures because of the exceptionally positive nature of the operation. In addition, some patients tend to be highly concerned about taking rescue analgesics because of potential accumulation in breast milk, and they may thus try to tolerate pain more than after other operations.

The latency time to rescue analgesics did not differ among the groups, unlike in another study in which 0.1 and 0.25 mg of IT morphine were compared (12). In this study, there was wide variation in the time interval in the groups other than the IT 200 group, in which only two patients required oxycodone. The latency time ranged from 4 to 24 hours in the IT 100 group and from 6 to 24 hours in the Epidural group, whereas the two patients in the IT 200 group needed rescue analgesia within 5 hours. At the last time point of VAS observations (at 24 hours after spinal anesthesia) the scores in the Epidural group were statistically significantly lower than in the IT groups. This finding was, however, clinically insignificant because the median values ranged from 0.7 to 1.1 and were probably due to the study protocol, in which epidural morphine was given 90 minutes later than both the IT doses. For the same reason, three hours after the administration of spinal anesthesia, the incidence of itching in PACU was less among the parturients receiving epidural morphine compared with the IT groups.

The results of analgesia and the incidence of itching in this study strongly support the proposed epidural/IT dose ratio of 20:1 (13). The least amount of itching was noted in the IT 100 group, but additional analgesia was required more often in this group than in the IT 200 group. The results of the Epidural group were between those of the two IT groups.

The overall incidence of itching—77% during the first postoperative 24 hours—was rather frequent compared with the observations of Uchiyama et al. (9) but was comparable with the results of other studies (4,14,15). This may at least partly depend on the patients’ awareness of possible side effects and regular questioning concerning side effects. The incidence of itching was least in the IT 100 group. The overall incidence for the medical treatment of itching was 37%. This side effect was dose related in our study. The difference between the groups was significant only between the two IT Morphine groups, although medication was needed almost as often in the Epidural group. The fact that both itching and analgesia develop concurrently is supported by the 50% and 95% effective dose (ED50 and ED95, respectively) in labor analgesia studies. In a study with fentanyl, the occurrence of pruritus expressed with ED50 and ED95 closely paralleled the analgesic ED50 and ED95 responses (16).

The incidence of PONV was 23%; this is less than reported in other studies (9,14) and did not differ among the groups. It did not appear to be dose related in this study, unlike in some other studies (9,10,15). None of the parturients had significant sedation or respiratory depression.

The results of this study may lead to different conclusions, depending on the value given to the need for rescue analgesics and the need to treat side effects. According to some studies, parturients appreciate steady and good analgesia regardless of side effects, such as itching (17,18). In this study, patient satisfaction was high in all the study groups, despite the frequent incidence of itching. Only 2.7% of the patients were not willing to have the same method of analgesia for a future cesarean delivery. We conclude that after cesarean delivery in women who are receiving ketoprofen, although rescue analgesics may be required, IT morphine 100 μg is superior for postoperative analgesia.

Back to Top | Article Outline


1. Chadwick HS, Ready LB. Intrathecal and epidural morphine sulfate for postcesarean analgesia: a clinical comparison. Anesthesiology 1988; 68: 925–9.

2. Sinatra RS, Ayoub CM. Postoperative analgesia: epidural and spinal techniques. In: Chestnut DH, ed. Obstetric anesthesia: principles and practice. 2nd ed. St. Louis: Mosby, 1999: 536–42.

3. Pavy TJ, Gambling DR, Merrick PM, Douglas MJ. Rectal indomethacin potentiates spinal morphine analgesia after caesarean delivery. Anaesth Intensive Care 1995; 23: 555–9.

4. Cardoso MM, Carvalho JC, Amaro AR, et al. Small doses of intrathecal morphine combined with systemic diclofenac for postoperative pain control after cesarean delivery. Anesth Analg 1998; 86: 538–41.

5. Choi DH, Kim JA, Chung IS. Comparison of combined spinal epidural anesthesia and epidural for cesarean section. Acta Anaesthesiol Scand 2000; 44: 214–9.

6. Sarvela PJ, Halonen PM, Korttila KT. Comparison of 9 mg of intrathecal plain and hyperbaric bupivacaine both with fentanyl for cesarean delivery. Anesth Analg 1999; 89: 1257–62.

7. Sun HL, Wu CC, Lin MS, Chang CF. Effects of epidural morphine and intramuscular diclofenac combination in postcesarean analgesia: a dose-range study. Anesth Analg 1993; 76: 284–8.

8. Palmer CM, Emerson S, Volgoropolous D, Alves D. Dose-response relationship of intrathecal morphine for postcesarean analgesia. Anesthesiology 1999; 90: 437–44.

9. Uchiyama A, Ueyama H, Nakano S, et al. Low dose intrathecal morphine and pain relief following caesarean section. Int J Obstet Anesth 1994; 3: 87–91.

10. Yang T, Breen TW, Archer D, Fick G. Comparison of 0.25 mg and 0.1 mg intrathecal morphine for analgesia after cesarean section. Can J Anaesth 1999; 46: 856–60.

11. Gerancher JC, Floyd H, Eisenach J. Determination of an effective dose of intrathecal morphine for pain relief after cesarean delivery. Anesth Analg 1999; 88: 346–51.

12. Abboud TK, Dorr A, Mosaad P, et al. Minidose intrathecal morphine for the relief of postcesarean section pain: safety, efficacy, and ventilatory responses to carbon dioxide. Anesth Analg 1988; 67: 137–41.

13. Sinatra RS, Ayoub CM. Postoperative analgesia: epidural and spinal techniques. In: Chestnut DH, ed. Obstetric anesthesia: principles and practice. 2nd ed. St. Louis: Mosby, 1999: 521–36.

14. Rosaeg OP, Lindsay MP. Epidural opioid analgesia after caesarean section: a comparison of patient-controlled analgesia with meperidine and single bolus injection of morphine. Can J Anaesth 1994; 41: 1063–8.

15. Milner AR, Bogod DG, Harwood RJ. Intrathecal administration of morphine for elective Caesarean section: a comparison between 0.1 mg and 0.2 mg. Anaesthesia 1996; 51: 871–3.

16. Herman NL, Choi KC, Affleck PJ, et al. Analgesia, pruritus and ventilation exhibit a dose-response relationship in parturients receiving intrathecal fentanyl during labor. Anesth Analg 1999; 89: 378–83.

17. Swart M, Sewell J, Thomas D. Intrathecal morphine for caesarean section: an assessment of pain relief, satisfaction and side-effects. Anaesthesia 1997; 52: 373–7.

18. Cade L, Ashley J, Ross W. Comparison of epidural and intravenous opioid analgesia after elective caesarean section. Anesth Intensive Care 1992; 20: 41–5.

© 2002 International Anesthesia Research Society


Become a Society Member