| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
*Department of Anaesthesia, Sunnybrook and Women’s College Health Sciences Centre, and the University of Toronto, Toronto, Ontario, Canada;
Department of Anesthesiology, Duke University Medical Center, Durham, North Carolina;
Department of Anesthesia, Royal University Hospital, College of Medicine, University of Saskatchewan, Saskatoon, Saskatchewan, and College of Medicine, University of Saskatchewan, Royal University Hospital, Saskatoon, Saskatchewan, Canada;
Department of Obstetrics and Gynaecology, University of Toronto, and Sunnybrook and Women’s College Health Sciences Centre, Toronto, Ontario, Canada;
||Department of Obstetrics and Gynaecology, University of British Columbia, British Columbia Women’s Hospital, Vancouver, British Columbia, Canada; and
¶Department of Mathematics & Statistics, Statistical Consulting Service, Dalhousie University, Halifax, Nova Scotia, Canada
Address correspondence and reprint requests to Stephen H. Halpern, MD, MSc, Department of Anesthesia, Sunnybrook and Women’s College Health Sciences Centre, 76 Grenville St., Toronto, Ontario, M5S 1B2, Canada. Address e-mail to stephen.halpern{at}sw.ca
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
IMPLICATIONS: Two-hundred forty-two healthy, term, nulliparous patients in spontaneous labor were randomized to receive either patient-controlled IV or patient-controlled epidural labor analgesia. No difference in the incidence of cesarean delivery was observed between groups. The data support the hypothesis that epidural analgesia does not increase the incidence of cesarean delivery compared with IV analgesia.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
This study was designed to determine whether epidural analgesia increases the incidence of cesarean delivery compared with parenteral opioids. The study was conducted in four centers across Canada to increase the applicability of the results to a wide population of parturients. We used patient-controlled techniques in an attempt to maximize analgesia while using as little drug as possible. Most importantly, a strict protocol was instituted for all obstetrical interventions.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Patients randomized to receive PCIA were given an initial IV dose of 100 µg of fentanyl incrementally over 1–5 min. If the pain was not adequately relieved, an additional 50 µg was given and repeated every 5 min until the patient reported adequate pain relief. The PCIA pump was initially set to give aliquots of 25 to 50 µg of fentanyl with a lockout interval of 10 min. The lockout period and bolus dose could be increased or decreased according to the discretion of the attending anesthesiologist. We assessed the patient for signs of excessive adverse effects, such as drowsiness and respiratory depression. If signs of excessive drowsiness or respiratory depression were present, the fentanyl was discontinued. If pain relief was inadequate with the PCIA fentanyl, patients were offered inhaled 50% nitrous oxide or the opportunity to receive epidural analgesia.
In patients randomized to receive PCIA, we identified the epidural space and inserted a multiport catheter at the L2-3 or L3-4 level. We initiated analgesia with 3 5-mL aliquots of 0.1% bupivacaine. Additional aliquots were given to a maximum of 25 mL with 100 µg of fentanyl. Failure of analgesia was treated with 0.25% bupivacaine, 2% lidocaine, and/or repositioning of the catheter.
Once adequate analgesia was established, we instructed the patient about the use of the PCEA pump and connected the device to the catheter. The pump provided 0.08% bupivacaine with fentanyl 1.6 µg/mL. The pump was set to give aliquots of 5 mL, with a lockout interval of 10 min, no maximum dose, and no continuous background infusion.
If the analgesia became incomplete during maintenance, the clinician gave an additional 5–10 mL of bupivacaine 0.125% with 50 µg of fentanyl. If pain relief was still inadequate and a bilateral sensory block higher than T10 was measurable, the clinician administered 5–10 mL of 2% lidocaine. If, despite the use of 2% lidocaine and an adequate sensory level, adequate analgesia was not achieved, the patient was considered an analgesic failure.
A diagnosis of dystocia in the first stage was considered if there was no progress in cervical dilation over a 2-h period or if in a 4-h period there was less than a 0.5-cm increase in cervical dilation. Cesarean delivery was not performed for dystocia without first giving a trial of oxytocin therapy of at least 2 h.
Patients were not encouraged to undertake expulsive efforts until the head was on the perineum or an overwhelming urge to push occurred. If there was no descent or rotation of the presenting part after 1 h of second stage, an oxytocin infusion was established. After 1 h of oxytocin infusion, the patient was assessed by both the delivering physician and the attending anesthesiologist. If the parturient had no urge to push and the head was not on the perineum, the analgesia requirements were assessed and adjustments were made, if necessary, to facilitate pushing. Oxytocin was continued for an additional hour of passive second stage. If the parturient had no urge to push but the head had descended to the perineum, she was encouraged to push. If 2 h of passive second stage with or without oxytocin infusion had elapsed and the mother still had no urge to push, pushing was encouraged for 90 min. If, after 90 min of pushing, delivery was still not achieved, an obstetrical assessment was performed, and there was an operative intervention if appropriate.
Compliance with both the anesthetic and obstetrical protocol was monitored throughout the study. A physician not involved with patient care adjudicated cases of breach of protocol. Those monitoring the adherence to the obstetrical protocol were blinded to analgesic treatment group.
The primary outcome was the incidence of cesarean delivery in each treatment group. Secondary obstetrical outcomes included instrumental vaginal delivery, spontaneous vaginal delivery, and duration of the second stage of labor. We measured maternal pain by using a visual analog scale every 2 h during the first stage of labor and every half hour during the second stage of labor. Concurrently, we assessed motor block of the lower extremities by using a six-point modified Bromage scale (1 = no motor power; 6 = able to get out of bed and do a deep knee bend) (12). We measured maternal satisfaction with analgesia by using a visual analog scale after delivery. We recorded the incidence of adverse effects—such as drowsiness, respiratory depression (respiratory rate <10 breaths/min), maternal fever (>38.0°C), and the need for medication for nausea and vomiting—throughout labor and delivery. The incidence of epidural analgesia in the PCIA group (crossover) was also recorded. Finally, we noted the total doses of bupivacaine and fentanyl for each group. Neonatal outcomes included Apgar scores at 1 and 5 min, umbilical artery cord pH, umbilical artery cord PCO2, umbilical artery cord base excess, use of naloxone, use of resuscitation with oxygen or positive-pressure ventilation, and the incidence of neonatal fever, as defined by a temperature of >38°C within 24 h of birth.
The data were analyzed according to group assignment (intention to treat) by using MINITAB (Version 13; State College, PA). Demographics were analyzed with means and standard deviations or medians and interquartile ranges, as appropriate. Categorical data, including the primary outcome, were analyzed with Fisher’s exact test. The median score for maternal pain for each time interval (before analgesia, first stage of labor, and second stage of labor) was calculated for each patient. These medians were compared by using a Mann-Whitney test. The most profound motor block experienced by each parturient was taken to be the Bromage score for that patient for the first and second stages of labor. Drowsiness was measured with a four-point scale (0 = no sedation; 1 = subjective sedation; 2 = sleeping between contractions, but awake during contractions; and 3 = sleeping during contractions). Other data were analyzed with unpaired Student’s t-tests or Mann-Whitney tests as appropriate. A P value of <0.05 was considered statistically significant.
At the time of the study, the incidence of cesarean delivery in this population of healthy, nulliparous parturients was approximately 11% in all 4 centers. The sample size was based on a 5% (absolute) reduction in cesarean delivery rate to 6% in the IV opioid group and a power of 0.8. Approximately 485 patients per group were needed. A priori, we decided to inspect the neonatal data after enrolling 200 patients to ensure neonatal safety. We also decided to stop the study after 2 yr of enrollment, regardless of the number of patients.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The maternal and neonatal demographics were similar between groups (Table 1). There was no difference in the incidence of cesarean delivery between groups (Table 2). The other maternal and neonatal outcomes are shown in Tables 2 and 3, respectively. The use of medications for both groups is shown in Table 4.
|
|
|
|
Maternal adverse effects were less frequent in the PCEA group compared with PCIA. Fewer patients required antiemetic therapy, and fewer patients were drowsy (Table 2). Maternal satisfaction scores were higher and pain scores lower in the PCEA group (Table 2, Fig. 1). There was no significant difference in the incidence of maternal fever.
|
The neonatal outcomes are shown in Table 3. More neonates received naloxone (20 of 118 versus 4 of 124; P < 0.001) and required active resuscitation (61 of 118 versus 38 of 124; P = 0.001) in the PCIA group. Both of these were statistically significant. The incidence of an Apgar score of <7 at 1 min reflects this difference (33 of 118 versus 21 of 121; P = 0.04), but by 5 min there were very few infants with low scores in either group (5 of 118 versus 4 of 123; P = 0.68). There was no difference in the umbilical artery cord gases, nor was the incidence of neonatal fever different between groups.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Unfortunately, we were unable to recruit the full sample size to confirm with more confidence that there was no statistically significant difference in the cesarean delivery rate between groups. This was because of difficulty in recruitment over the course of the clinical trial caused by an increase in the rate of induced labor. However, the cesarean delivery rate in both groups (
10%) was very similar to the historical values (11% in the year 1996) for institutions participating in the study; this lends credibility to the claim that epidural analgesia did not cause an increase in rate. This finding contrasts with a previous study that did find a significant difference (14). In that study, cesarean delivery was significantly more frequent after labor epidural analgesia (25%) and was significantly less frequent after parenteral opioid analgesia (2%) than the historical rate for that institution (15%) (14). Therefore, there may have been other causes, apart from the mode of analgesia, for the differences seen.
Of note, none of the previously reported studies were of sufficient size to answer the question of whether epidural analgesia increases the risk of cesarean delivery. However, there are a number of well conducted, relatively large clinical trials published on this topic. The results from these studies are sufficiently homogeneous to justify combining them statistically in a meta-analysis. When this was done for all studies published before March 2001, there were almost 4300 patients available for analysis. In these studies, there was no difference in the cesarean delivery rate (odds ratio, 1.00; 95% confidence interval, 0.77 to 1.28) (15).
We also found no difference between groups in the incidence of forceps delivery or spontaneous vaginal delivery. A meta-analysis of other, similar, clinical trials showed a statistically significant increase in the instrumental vaginal delivery rate in patients who received epidural analgesia compared with those who received opioids (15). However, at least in one circumstance, the authors clearly stated that there were more forceps deliveries in the epidural group to facilitate training of residents (1). This may have been true for other studies as well. When only instrumental vaginal deliveries for dystocia were included, the meta-analysis showed no difference between groups, and the results were similar to those of this study (15).
There was approximately a 23-minute increase in the duration of the second stage of labor in the PCEA group (Table 2). This is similar to findings of other studies. It should be noted that, whereas patients who had a cesarean delivery in the first stage of labor were excluded from this outcome, those who had an operative delivery in the second stage were not. Further, 12 patients in the PCIA group received epidural analgesia in the second stage of labor. These cases were analyzed in the PCIA group (intent to treat), and the anesthetic intervention may have shortened the second stage. Therefore, the difference between groups is difficult to interpret, and whether the difference is clinically significant in a particular setting will depend on the obstetrical criteria for intervention.
In contrast to other studies (9,14), we did not observe an increased incidence of the need for oxytocin after initiation of labor analgesia (Table 2). This may have been due to obstetrical compliance to the protocol for initiation of oxytocin or because of different criteria for its use.
Women in the PCEA group were significantly more satisfied with their analgesia compared with those in the PCIA group. Better analgesia, a reduced incidence of nausea (and requests for antiemetic therapy), and a reduced incidence of drowsiness are the likely contributing factors. Inadequate analgesia was the main reason that 39 of the patients in the PCIA group requested epidural analgesia (12 requests were in the second stage to facilitate operative delivery). This occurred despite the large doses of IV fentanyl that the patients administered to themselves (Table 4). In contrast, few patients in the PCEA group had bothersome motor block of the lower extremities, because of the small concentration of bupivacaine used.
The condition of the neonates was good in both groups. There was no difference between groups in the umbilical artery blood gases. Although there were more recorded Apgar scores of <7 at one minute in the PCIA group, there was no difference between groups at five minutes. In addition, more neonates in the PCIA group required active resuscitation with oxygen and bag and mask, but this was short-lived. As in previous studies, more neonates in the PCIA group received naloxone. This may be a reflection of the large amount of fentanyl patients required to obtain sufficient comfort or may reflect a bias on the part of the physicians caring for the neonate, who were not blinded to the method of intrapartum analgesia. The incidence of neonatal fever was very small in this study and was similar for both groups (Table 3).
This investigation had a number of deficiencies that may have reduced the reliability of the data. First, because epidural analgesia gives superior pain relief to IV opioids, it was impossible to blind the clinician or the patient to the treatment group. However, there was good compliance with the anesthetic and obstetrical protocols in both therapy arms. Second, many of the women who received PCIA did not obtain adequate analgesia and sought other means of pain relief. Although a number of statistical methods have been proposed to deal with this situation specifically, none is completely satisfactory. Finally, we did not enroll the number of patients we had originally intended. This has occurred in other clinical trials and can lead to bias if stopping rules are not clearly established a priori (14). We had decided on a time limit of two years at each institution and therefore stopped the clinical trial at that time. Therefore, although our ideal sample size was not achieved, we would not expect the results to provide biased information. Nevertheless, this investigation alone lacked the power to demonstrate small changes in rates of cesarean delivery between groups.
In conclusion, PCEA does not cause an increased incidence of cesarean delivery or instrumental vaginal delivery when compared with fentanyl PCIA in healthy nulliparous women in spontaneous labor. However, maternal satisfaction, because of reduced nausea and drowsiness and better pain relief, is better with PCEA. Further, neonatal depression (as measured by the Apgar score at one minute), the need for naloxone, and the need for resuscitation is less frequent in patients who receive PCEA. These results are consistent with other similar studies that have been published over the last 15 years. Clinicians should be aware of this information when discussing pain-relief options with their patients.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
1 Muir HA, Shukla R, Liston R, Writer D. Randomized trial of labour analgesia: a pilot study to compare patient-controlled intravenous analgesia with patient-controlled epidural analgesia to determine if analgesic method affects delivery outcome. Can J Anaesth 1996;43:A60. 
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. McBain, E. G. Lemire, and D. C. Campbell Epidural labour analgesia in a parturient with Noonan syndrome: a case report: [Analgesie peridurale chez une parturiente atteinte du syndrome de Noonan : une presentation de cas]. Can J Anesth, March 1, 2006; 53(3): 274 - 278. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
