Background

The optimal timing of epidural analgesia has been a controversial issue, and how early women can benefit from epidural analgesia is still debated. The objective of this trial was to test the hypothesis that patient-controlled epidural analgesia given at cervical dilation of 1.0 cm or more does not increase the risk of prolonged labor or Cesarean delivery.

Methods

After institutional review board approval and patient consent, 12,793 nulliparous patients requesting neuraxial analgesia were enrolled and randomized to an early epidural (cervical dilation at least 1.0 cm) or delayed epidural (cervical dilation at least 4.0 cm) group. A 15-ml epidural analgesic mixture consisting of 0.125% (1.25 mg/ml) ropivacaine plus 0.3 microg/ml sufentanil was given in a single bolus, followed by patient-controlled pump with a 10-ml bolus without background infusion. Repeatable meperidine (25 mg) was prescribed as being the rescue analgesic to patients in the delayed epidural group. The primary outcome was the rate of Cesarean section.

Results

The median diameters of cervical dilation were 1.6 cm and 5.1 cm in the early and delayed epidural groups, respectively (P < 0.0001). The duration of labor from analgesia request to vaginal delivery was equal in both groups (11.3 +/- 4.5 h for early epidural and 11.8 +/- 4.9 h for delayed epidural group women, P = 0.90). No statistically significant difference in the rate of Cesarean section was observed between the two groups on the intention-to-treat analysis (23.2% vs. 22.8% in the early and delayed epidural groups, respectively; P = 0.51).

Conclusions

Epidural analgesia in the latent phase of labor at cervical dilation of 1.0 cm or more does not prolong the progression of labor and does not increase the rate of Cesarean in nulliparous women compared with the delayed analgesia at the cervical dilation of 4.0 cm or more.

NEURAXIAL analgesia is the most effective treatment available for pain control during labor and delivery, and it is a preferable method because it can provide more effective pain relief compared with nonneuraxial pharmacological analgesia.1However, optimal timing of epidural analgesia (EA) has been a controversial issue and how early in the latent phase of labor can women benefit from epidural is still debated. Before 2002, clinical guidelines recommended that the administration of EA in nulliparous women should be delayed until the cervical dilation reaches at least 4.0 to 5.0 cm and that other forms of analgesia should be used until that time.2In 2005, Wong et al.  published a paper clarifying that pain relief early in labor with neuraxial analgesia at the cervix dilated 2.0 cm or more does not increase the risk of Cesarean delivery3; this combined with Ohel’s report4contributed to the change in recommendation on EA in labor pain control from the American College of Obstetricians and Gynecologists in June 2006.5 

The current best available evidence in nulliparous women at term with singleton fetus in vertex presentation supports that EA is safe in laboring women with cervix dilated 2 cm or more,3,4and systematic reviews on this topic suggest improved definition of dystocia and nonreassuring fetal status diagnoses with precise and repeatable criteria.1,6,7The National Institute for Health and Clinical Excellence guidelines8,9suggest that “women in labor who desire regional analgesia should not be denied it, including women in severe pain in the latent first stage of labor – a period of time begins from painful contractions and some cervical change, including cervical effacement and dilation up to 4 cm. However, data in its finding-review section do not address the indication when cervical dilation is less than 2.0 cm.

The current randomized controlled trial was designed to test the hypothesis that patient-controlled epidural analgesia in women requesting and receiving early labor analgesia in the latent phase of labor (at least 1.0 cm cervical dilation) do not have an increased risk of prolonged labor or Cesarean section (CS) compared to women who are assigned to wait for a cervical dilation of at least 4.0 cm.

Participants and Ethics

After approval by the institution’s Ethics Examining Committee of Human Research (the Affiliated Nanjing Maternity and Child Health Care Hospital, Nanjing Medical University, Nanjing, Jiangsu, China), all healthy nulliparous women who spontaneously went into labor at term requesting labor analgesia were screened for eligibility. All included participants were asked to participate in the study soon after admission to the delivery suite and signed an informed consent before randomization. If the woman did not want an epidural, we reassigned her to receive alternative methods for labor pain control. A full explanation was given to those willing to accept epidural analgesia with respect to epidural puncture and catheterization, the opioid and local anesthetics used in this study, and the linear Visual Analog Scale (VAS) of pain and satisfaction with analgesia. The data were collected between January 2003 and December 2007 at the Nanjing Medical University–affiliated tertiary teaching hospital in China. Hospital teaching status was ascertained from the Council of Teaching Hospitals of Chinese Medical Colleges. During the whole recruitment course, the labor epidural rate in our institution is 80% on average in nulliparous women.

Randomization

When patients requested analgesia, the cervical examination was performed by obstetricians. If the cervical dilation was 1.0 cm or greater in diameter, the subjects were assigned by a completely random allocation in a single block to either the latent phase analgesia with cervix at least 1.0 cm or the active phase analgesia with cervix at least 4.0 cm. The random number list was generated by means of the QuickCalcs#(GraphPad Software Inc., La Jolla, San Diego, CA). The group assignment numbers were sealed in an envelope and kept by the study supervisor. After the written consent was signed and a cervical dilation was examined to be 1.0 cm or greater, then the opaque envelope was unsealed to determine which analgesic technique would be performed. The healthcare providers, data-collecting members, and parturients were not masked to the group allocation except for data-analyzing members.

Demographic Characteristics

The following data were collected as demographic characteristics of the subjects: age at delivery, weight, height, gestational age of fetus, current status of smoking, whether spontaneous rupture of the membrane more than 12 h before oxytocin infusion, and VAS rating of pain intensity before any analgesia.

Exclusion Criteria

Parturients were excluded from the study if one or more of the following criteria were met: (1) allergy to opioids, a history of the use of centrally–acting drugs of any sort, chronic pain, and psychiatric diseases records; (2) participants younger than 18 yr or older than 45 yr; (3) those who were not willing to or could not finish the whole study; (4) alcohol- or opioid-dependent patients were excluded for their influence on the analgesic efficacy of the epidural analgesics; (5) subjects with a nonvertex presentation or scheduled induction of labor; (6) diagnosed diabetes mellitus and pregnancy-induced hypertension; (7) twin gestation and breech presentation.

Subjects who did not request any analgesic procedures (see Table 1, Supplemental Digital Content 1, which is a table listing all possible reasons for nonanalgesia request in Chinese population, https://links.lww.com/ALN/A543) were not enrolled into the study, and they were still treated with standard obstetric care procedures according to clinical indications (see Table 2, Supplemental Digital Content 2, which is a table listing the procedures of standard obstetric care, https://links.lww.com/ALN/A544). Even so, the systemic or epidural analgesia was still available to them if they wanted an expostanalgesia based on the individual request. In case the standard epidural analgesia was used, the technique was performed by nonstudy members, and the corresponding data were not analyzed in this trial.

Analgesia Design

Cervix of parturient admitted into the delivery room was examined after analgesia request. If the cervical dilation was less than 1.0 cm, 25 mg of meperidine was prescribed intramuscularly as being the rescue analgesic until cervical dilation of at least 1.0 cm, and repeatable 25-mg meperidine could be given intramuscularly if the pain control was inadequate at the patient’s request; if the cervical dilation was at least 1.0 cm, the parturients were randomly assigned into one of two groups by means of a random-number list: group of latent phase analgesia (cervical dilation of at least 1.0 cm) and group of active phase analgesia (cervical dilation of at least 4.0 cm) (see flowchart in fig. 1). Definitions of the first and second stages of labor are adapted from the National Institute for Health and Clinical Excellence guidelines8and reports elsewhere.10If a patient with the cervical dilation less than 4.0 cm was randomized into active group, 25 mg of meperidine was prescribed intramuscularly as being the rescue analgesic until the dilation reached 4.0 cm or more, and repeatable 25-mg meperidine could be given intramuscularly if the pain control was inadequate at the patient’s request. Epidural puncture and catheterization was performed to all participants after cervical examination by the obstetricians or midwives. The test dose of 3.0 ml of lidocaine 1.5% (45 mg) plus 5 μg/ml epinephrine was given to both groups. After delivering a test dose, all participants received a 15-ml epidural analgesic mixture in a single bolus of 0.125% (1.25 mg/ml) ropivacaine with 0.3 μg/ml sufentanil followed by patient-controlled analgesic pump with a 10-ml patient-controlled bolus without background infusion, a lockout interval of 15 min, and hourly limit of 30 ml. Cervical examination was performed by the obstetric staff who were not involved in this study in the delivery room hourly until cervix dilation reached 4.0 cm or more.

Fig. 1. Selection flow of the study participants. The major reasons for exclusion and lost follow–up were presented in the result part in the text. * See reasons for nonanalgesia request in the table, Supplemental Digital Content 1; † see standard obstetric care procedures in the table, Supplemental Digital Content 2; ‡ see indications of cesarean section in the table, Supplemental Digital Content 2; § denotes protocol violation occurred of these patients,  i.e., subjects with cervical dilation ≥ 4.0 cm were randomly assigned into the latent phase analgesia group. i.m. = intramuscularly. 

Fig. 1. Selection flow of the study participants. The major reasons for exclusion and lost follow–up were presented in the result part in the text. * See reasons for nonanalgesia request in the table, Supplemental Digital Content 1; † see standard obstetric care procedures in the table, Supplemental Digital Content 2; ‡ see indications of cesarean section in the table, Supplemental Digital Content 2; § denotes protocol violation occurred of these patients,  i.e., subjects with cervical dilation ≥ 4.0 cm were randomly assigned into the latent phase analgesia group. i.m. = intramuscularly. 

Close modal

If an inadvertent dural puncture with the epidural needle during epidural catheterization was encountered, therapeutic procedures were followed (see Table 3, Supplemental Digital Content 3, which is a table listing possible therapeutic maneuvers after accidental dural puncture and the accidental rate in the current study, https://links.lww.com/ALN/A545). This population was treated on an intention-to-treat analysis.

Peripartum Management and Monitoring

A catheter was inserted in a right or left antecubital vein for fluid and drug administration. Ringer’s solution (8 ml/kg) was titrated 15 min before initiation of the epidural analgesia. Intrapartum fluid management included replacement of preexisting fluid deficits, normal losses (maintenance requirements), quantification and measurement of urine, hemodynamic variables, and hemoglobin concentration.

The maternal parameters monitored during the whole study from before the analgesic procedures to the end of the labor included the heart rate by three-lead electrocardiograph, respiratory rate, noninvasive systolic and diastolic blood pressure, mean arterial pressure, and fingertip pulse oximetry.

If necessary, the intrauterine pressure sensor was placed to show the intensity of uterine contraction. Oxytocin was infused by the nursing staff set by the obstetricians according to the clinical guidelines. A decision whether an operative delivery was necessary was made by the obstetrical team that was not involved in the study depending on maternal and fetal indications. The treatment of obstetric staff was standardized according to the standard care procedures with or without EA.

The patient-derived VAS scores of pain with the 100-mm gauge (based on a 0–100 linear VAS: 0 = no pain, 100 = worst pain imaginable) were recorded at analgesia request, and were then measured hourly from the initiation of epidural analgesia to the completion of the labor. The average scoring of pain was calculated to each woman at the end of the labor. In addition, the maternal satisfaction with analgesia was assessed with the VAS system (a 0–100 linear VAS: 0 = dissatisfactory, 100 = very satisfactory). Finally, the incidence of the side effects, such as nausea, vomiting, pruritus, shivering, and urinary incontinence or retention throughout the study were recorded by the follow-up physicians according to the maternal reports; maternal oral temperature, hypotension and tachycardia were measured by caregiver intermittently (see Supplemental Digital Content 4, which is a table listing all side effects encountered after epidural analgesia, https://links.lww.com/ALN/A546).

Fetal and Neonatal Management

A continuous external electronic fetal heart-rate monitoring and tocodynamometry were made. Apgar scores were rated by the pediatric personnel according to the standard assessment. Umbilical-cord blood gas analysis was performed by the investigators.

Primary Outcome

The method of operative delivery, i.e. , the rate of CS was selected as the primary outcome of the different analgesic procedures.

Secondary Outcome

The following measures were selected as the secondary outcomes: the indication for Cesarean delivery; rate of instrument-assisted delivery; the time to vaginal delivery, latent phase, and active phase of the first stage and the second stage; the duration of epidural analgesia; the verbal ratings of VAS pain at the time of analgesic request and epidural placement; the subjects were also asked to rate their contraction pain at the endings of different phases of labor determined by cervical examination; satisfaction with analgesia after completion of vaginal delivery; oxytocin requirement; the maternal oral temperature; the incidence of side effects from epidural puncture and drug delivery.

In addition, we investigated the association between the maternal demographic variables and the rates of Cesarean delivery. Each one unit increase in age (Δ= 1 yr), weight (Δ= 1 kg), height (Δ= 1 cm), and oxytocin use (Δ= 1 mU/min) are assessed by using multivariate regression modeling.

Infant outcomes include the body weight, Apgar scorings, umbilical-cord blood gas measurement, antibiotic treatment, and sepsis assessment. Neonatal sepsis evaluation was performed as previously reported.11In brief, a blood culture and a complete blood count of the neonate suspected for sepsis possibility performed and followed by a lumbar puncture if one major or two minor criteria presented at the time of the study. Major criteria included rupture of membranes for more than 24 h or a sustained fetal heart rate of more than 160 beats per minute 15 min or longer. Minor criteria included a low-grade maternal temperature of 37.5°C to 38°C, rupture of membranes for 12 to 24 h, maternal admission white blood cell count of more than 15,000 cells/ml3, or 5-min Apgar score less than 7.

Sample Size

A prestudy power table where Δ (the mean difference in Cesarean delivery recorded in a pilot study, i.e. , a difference of 2.3% in the CS rate) equaled 19.1% in the latent phase analgesia group and 16.8% in the active phase analgesia group, two-sided α= 0.05, one-sided β= 0.10, and the power of test = 0.90. Therefore, a minimal sample size of 5,840 subjects per group was needed to detect this difference. We increased the sample size to 7,500 in each group to account for potential missing data and dropouts during the 5-yr study course. The approximately 20% increase in sample size was mainly based on an institutional estimate that around 14% (median; interquartile range 8% to 22%) of patients dropped out or had missing data during the follow-up period in our previous clinical trials (the major reason for such a high dropout rate was because Chinese people had a different philosophical view about human research that differs from the Occidental population). In addition, given the long-lasting period of this study, during which many factors would have influenced subject enrollment, such as changes in patient’s attitude and knowledge about epidural and clinical study, and increment of physician’s experience, and improvement of medical facilities, etc., we increased the sample size to 7,500 per group using the upper limit 22%.

Statistical Analysis

All our data analyses, were first performed for patients in both of the intervention groups to which they were allocated –i.e. , on an intention-to-treat basis. Excluded and withdrawn subjects after randomization because of protocol violation and patient-controlled analgesia device were considered as the cases that encountered CS and analyzed in the group to which they were randomized according to the data that were available. In addition, the per protocol population analysis included all subjects who underwent the study intervention and completed the study. Excluded and withdrawn subjects, including protocol violation and patient-controlled analgesia device cases were excluded in this population. The primary analysis was based on the CS rate of both intention-to-treat and per-protocol populations.

Values are expressed as the mean, SD, median, interquartile range, or numbers. All categorical data were analyzed with a chi-square test to indicate the trend. The difference in parametric data, including the demographic data and background characteristics, were compared with Student t  test. Mann–Whitney U test was used in analyzing nonnormally distributed variables and presented as the medians and interquartile ranges, including the effects of the epidural analgesia on patient’s self-rated VAS scorings of pain and satisfaction, and in the duration of labor and each stage of delivery, oxytocin use, gestational age of fetus, and the highest sensory block level. Cumulative-event curves indicating different time lengths of labor were estimated with the Kaplan-Meier method, and the groups were compared using the log-rank test used elsewhere.12 

Multivariable logistic regression analyses were performed to assess the adjusted association between maternal baseline variables and CS. First, a bivariate analysis was performed by using a liberal significance level of P < 0.25 to identify the baseline variables that were independently associated with the primary outcome for inclusion in the modeling after controlling for the other variables.13Modeling, was then performed by stepwise forward method with a criterion of P < 0.05 for variable retention in the models. In the logistic regression model, the independent variables assessed included age, height, weight, smoking status, and oxytocin infusion. Smoking status (0 = No; 1 = Yes) was entered as dichotomous variable for the logistic regression analysis. A Pearson correlation matrix of variables was used to identify collinear predictive variables.13Furthermore, Hosmer-Lemeshow test was used to assess the models’ fit (larger P  value means better fit or calibration), and predictive accuracy was assessed by the c-index as recommended elsewhere.13,14The odds ratio (OR) and its 95% confidence interval (95% CI) were calculated. The statistical significance was accepted at the level of P ≤ 0.05.

Analyses were performed by using GraphPad Prism version 5.0 (GraphPad Software Inc., San Diego, CA) or SPSS version 13.0 (SPSS Inc., Chicago, IL).

Fifteen thousand gravidas were screened (fig. 1). We excluded 2,207 subjects during the screening period because of the following reasons: 732 (33.2%) were without vertex presentation; 667 (30.2%) were diagnosed with diabetes and pregnancy-induced hypertension; 567 (25.7%) had twin gestation and breech presentation; 113 (5.1%) were not in the inclusion range of age (18 to 45 yr) at delivery; 87 (3.9%) were not willing to participate in the study; 30 (1.4%) were alcoholic; 11 (0.5%) had history of the use of centrally acting drugs. Finally, a total of 12,793 subjects were randomly assigned to the two groups and followed-up. After randomization, 164 subjects were excluded, and 6,274 subjects in the latent phase analgesia group and 6,355 subjects in the active phase analgesia group completed the whole study. Of the 164 excluded patients, 143 (87.2%) were excluded because of violation of the study protocol during data collection, and 21 subjects were excluded because of issues of the patient-controlled analgesic device and accidental dural puncture. The reasons for the protocol violation for the excluded 143 subjects was as follows: in 102 subjects whose cervix were over 4.0 cm were randomly assigned into the latent phase analgesia group; the cervical diameter was less than 4.0 cm or without cervical examination and then epidural analgesia was initiated in the active phase group (28 women), and 13 parturients whose cervix were less than 4.0 cm refused meperidine and received epidural analgesia in this group. All these excluded patients after randomization were treated as the intention-to-treat analysis.

Table 1summarizes the demographic, background characteristics, and baseline vital signs (all were within the physiologic ranges). No significant difference was observed between the two groups.

Table 1. Demographic and Background Characteristics of the Subjects 

Table 1. Demographic and Background Characteristics of the Subjects 
Table 1. Demographic and Background Characteristics of the Subjects 

The rate of Cesarean delivery in both groups displayed no statistically significant difference, P = 0.51. In addition, no significant differences were observed between the two groups in the indications of Cesarean delivery, the percentages of subjects who were treated with oxytocin infusion, or the maximal oxytocin dose. There was also no statistical significance in the rate of instrument-assisted delivery, P = 0.10 (table 2). Besides, similar results received when analyzing the primary outcome on the per protocol population basis, and there were no significant differences in the CS rate and instrument-assisted deliveries, P = 0.86 and 0.91, respectively.

Table 2. Major Maternal Outcomes after Epidural Analgesia 

Table 2. Major Maternal Outcomes after Epidural Analgesia 
Table 2. Major Maternal Outcomes after Epidural Analgesia 

There were no significant differences in the length of labor before vaginal delivery, the time to latent and active phases in the first stage, and the duration of the second stage of the labor between the two groups (table 2). Figure 2presented the time durations of different labor stages in Kaplan-Meier curves, and there was no difference in the median time from the request for analgesia and complete cervical dilation (562 min vs.  554 min in the latent and active phase groups, respectively; P = 0.42; fig. 2A), the time to the vaginal delivery (627 min vs.  661 min in the latent and active phase groups, respectively; P = 0.08; fig. 2B), the time to latent and active phases of first stage (fig. 2C and D), the time to second stage (fig. 2E), and the time to Cesarean and instrumental deliveries (fig. 2F and G).

Fig. 2. Kaplan-Meier curves for the length of labor.  Arrows indicate median values of time. (  A ) Time to successful vaginal delivery. The difference in both of the median time to delivery was –34 min (95% CI, –58 to –12 min,  P = 0.35). (  B ) The merged time to latent and active phase of the first stage of labor. The difference in both of the median time was –8 min (95% CI, –13 to –3 min,  P = 0.84). (  C ) Time to latent phase of first stage of labor. The difference in both of the median time was –14 min (95% CI, –22 to –4 min,  P = 0.67). (  D ) Time to active phase of first stage of labor. The difference in both of the median time was –19 min (95% CI, –37 to –8 min,  P = 0.39). (  A to D ) Time 0 is the time of the analgesia request. (  E ) Time to the second stage of labor. The difference in the median time was –8 min (95% CI, –18 to –2 min,  P = 0.92). (  F ) Time to delivery by Cesarean section. The difference in the median time was –35 min (95% CI, –55 to –20 min,  P = 0.44). (  G ) Time to instrument-assisted delivery. The difference in the median time was –35 min (95% CI, –46 to –23 min,  P = 0.53). 

Fig. 2. Kaplan-Meier curves for the length of labor.  Arrows indicate median values of time. (  A ) Time to successful vaginal delivery. The difference in both of the median time to delivery was –34 min (95% CI, –58 to –12 min,  P = 0.35). (  B ) The merged time to latent and active phase of the first stage of labor. The difference in both of the median time was –8 min (95% CI, –13 to –3 min,  P = 0.84). (  C ) Time to latent phase of first stage of labor. The difference in both of the median time was –14 min (95% CI, –22 to –4 min,  P = 0.67). (  D ) Time to active phase of first stage of labor. The difference in both of the median time was –19 min (95% CI, –37 to –8 min,  P = 0.39). (  A to D ) Time 0 is the time of the analgesia request. (  E ) Time to the second stage of labor. The difference in the median time was –8 min (95% CI, –18 to –2 min,  P = 0.92). (  F ) Time to delivery by Cesarean section. The difference in the median time was –35 min (95% CI, –55 to –20 min,  P = 0.44). (  G ) Time to instrument-assisted delivery. The difference in the median time was –35 min (95% CI, –46 to –23 min,  P = 0.53). 

Close modal

The average VAS ratings of pain at analgesia request were similar in both groups (table 1). There were no statistically significant differences in the VAS pain scorings during the latent and active phases and the second stage of labor (table 2). The duration of epidural analgesia in both groups was calculated from the bolus injection of the drug mixture into epidural space to the disappearance of the sensory block. The analgesic time epidurally in the latent phase group was much longer (median 12.6 h) versus  the active phase group (median 4.8 h), P = 0.02.

The intrapartum nausea and vomiting had a significantly lower incidence in the latent phase analgesia group than the active phase analgesia group. Fewer patients experienced pruritus in the active phase analgesia group. No significant difference was observed in the other side effects (high temperature greater than 38°C, urinary incontinence and retention, shivering, and hypotension) between the two groups (see Table 4, Supplemental Digital Content 4, which is a table listing all side effects encountered after epidural analgesia, https://links.lww.com/ALN/A546). The breastfeeding success at 6 weeks after delivery was less in the latent phase analgesia group compared with the active phase analgesia one, P < 0.0001 (table 2). There was no statistically significant difference in the neonatal outcomes between groups (table 3).

Table 3. Infant Outcomes after Epidural Analgesia 

Table 3. Infant Outcomes after Epidural Analgesia 
Table 3. Infant Outcomes after Epidural Analgesia 

Table 4summarizes the association between the maternal demographic variables and the rates of Cesarean delivery analyzed with a multivariate regression. Each one unit increase in age (yr), weight (kg), height (cm), and oxytocin use (mU/min) are associated with Cesarean delivery (the adjusted OR were 1.46, 1.67, 1.25, and 2.20, respectively, all P < 0.05), but current smoking status was not.

Table 4. Logistic Regression to Evaluate the Association between Maternal Demographic Variables and Cesarean Delivery 

Table 4. Logistic Regression to Evaluate the Association between Maternal Demographic Variables and Cesarean Delivery 
Table 4. Logistic Regression to Evaluate the Association between Maternal Demographic Variables and Cesarean Delivery 

The data from this 5-yr study demonstrate that providing epidural analgesia in the latent phase of spontaneous labor at term, compared with the active phase, did not increase the rate of intrapartum Cesarean delivery. In addition, no association was found between early epidural analgesic intervention and the rate of operative vaginal delivery.

These results have extended the findings from other studies. Wong et al.  have compared the risk of Cesarean delivery after early neuraxial analgesia (combined spinal-epidural analgesia, but not EA) and systemic analgesia when the cervix was between 2 and 4 cm, in which the cervical dilation reached at least 2.0 cm.3Chestnut’s studies evaluated early epidural analgesia at the cervical dilation was at least 3.0 cm or greater.15,16Ohel reported an approximately 10% CS rate in both groups when an epidural was placed at a mean cervical dilation approximately 2.4 cm versus  4.6 cm, which was enviable but may lack generalizability to other clinical environments.4In the current study, we performed epidural analgesia in the latent phase of the first stage of labor at the cervical dilation of at least 1.0 cm versus  the active phase analgesia at 4.0 cm or more, and no statistical difference in CS rate was found. As such, the interventional window of labor analgesia can be enlarged from 4.0 cm to 1.0 cm of cervical dilation.

Wong et al.  found that intrathecal (spinal) opioid use significantly shortened the first stage of labor compared with the systemic opioid administration.3In the current trial, we detected a slightly longer but not statistically significant prolonged duration of the first stage of labor in women who received epidurals in the latent phase. As such, we can still conclude that the earlier initiation of epidural analgesia in nulliparous women does not prolong the progress of labor.

The rate of instrument-assisted delivery might be increased if the epidural infusion of local anesthetics lasted longer,17,18but our data did not support such demonstration. In our study, the median time of instrumental delivery was similar in both of the groups (548 min vs.  513 min in the latent and active phase analgesia women, respectively). Meanwhile, the percentages of instrumental delivery in both groups were nearly same (approximately 12.0%). Therefore, epidural analgesia in early labor does not also increase the rate of instrument-assisted delivery.

Gambling et al.  reported an association between epidural analgesia and intrapartum oxytocin infusion,19and other studies have found that epidural analgesia combined with low-dose oxytocin infusion would increase the rate of Cesarean delivery.20In contrast to these studies, we did not find significant association among epidural analgesia, oxytocin requirement, and maternal delivering outcomes, which was in agreement with the results of Wong et al.  3 

In our study, we used meperidine as the rescue analgesic until the cervix dilated adequately to perform epidural analgesia in both groups. We did not find significant difference in the VAS pain ratings in both groups during latent labor, although epidural analgesia was given in the latent phase group, but meperidine in the active phase women. Jain et al.  reported that women who received meperidine still felt moderate pain (VAS 40–69 mm),21and the median VAS score in another study lowered significantly 30 min after the meperidine to 54 mm compared with the saline control at 78 mm.22Our finding was consistent with these data that meperidine during latent labor in the active phase analgesia group produced moderate pain-relieving effect with the median VAS score 48 mm.

We measured the oral temperature during epidural analgesia recommended by Banerjee et al.  that oral temperature is a preferred routine detection of maternal pyrexia in labor for its positive correlation with intrauterine temperature.23However, the 0.2ºC difference in oral temperature that we detected was so small as to be irrelevant in the clinical environment despite the statistical difference that was detected. We did not detect a significant increase in temperature of the magnitude that was shown by Camann et al.  24 

Nausea and vomiting are common side effects induced by pregnancy.25In addition, pain is a major cause of nausea.26We found that effective epidural analgesia in early labor can decrease intrapartum nausea and vomiting more significantly than the later labor analgesia. Also, opioids are a common cause of nausea, so the fact that our active labor patients got more meperidine likely played a role. Our data evidenced that the longer the time to epidural analgesia, the more the incidence of pruritus during labor. This is mainly because longer epidural analgesia means a relatively larger dose of lipophilic opioid was absorbed into blood, which produced more incidence of itching.

We have found potentially controversial results with respect to the influence of intrapartum labor pain control on postpartum breastfeeding initiation. Chang and Heaman found in their cohort study that epidural analgesia did not inhibit effective breastfeeding.27However, the Academy of Breastfeeding Medicine Protocol Committee considered that epidural analgesia with narcotic during labor has adverse effect on breastfeeding by delaying early skin-to-skin contact between mother and baby.28In addition, Volmanen’s questionnaire survey found the problem of “not having enough milk” was more often reported by those who had received epidural analgesia.29In our study, more than 70% of women in both epidural groups had successful breastfeeding, but the active phase analgesia group had more successful cases in breastfeeding than the latent phase analgesia group, which is consistent with previous reports that epidural analgesia is associated with impaired spontaneous breastfeeding, including breastfeeding at discharge from the hospital.30–32This difference is likely to be associated with a relatively longer duration of analgesia in the latent phase group compared with the active phase analgesia women, although the precise mechanism underlying such phenomenon is yet to be guaranteed in the future research.

The 1-min and 5-min Apgar scores were similar in both of the groups. This expressed early epidural analgesia does not exert significant influence on neonatal Apgar ratings. The results of umbilical-cord blood gases measurement were not different in agreement with the data from Wong et al.  3Goetzl et al.  found that epidural analgesia could increase the rate of neonatal sepsis work-ups in afebrile women, in which the rate was 20.4%.11Nonetheless, we did not find convincing evidence that epidural analgesia performed early or late in spontaneous labor increases the rate of neonatal sepsis. The 0.2ºC difference in maternal oral temperature in our study was not merely statistically nonsignificant, but it was not associated with an increase in neonatal sepsis work-ups. According to the report of Fisler et al. , neonates received more diagnostic tests and therapeutic interventions such as antibiotic administration in those whose mothers received epidural analgesia for pain relief during labor than infants born after an elective Cesarean section.33The neonatal antibiotic treatment in the current study was identical (approximately 13%) in both groups and in agreement with the aforementioned report (approximately 10–14%).

We also analyzed the association between maternal demographic variables and CS rate. In agreement with other previous studies,3,34our results showed that each one-unit increase in age, weight, height, and oxytocin infusion are significantly associated with the increase of Cesarean delivery. Nevertheless, the current smoking status did not increase CS rate. These indicate that maternal baseline characteristics are relatively more important predictors for Cesarean delivery than the intrapartum epidural analgesia intervention.

This study was subject to several limitations. First, this study lasted 5 yr; for such a long enrollment period, any changes in healthcare made may have influenced the results. The experience and knowledge of caregivers and the medical instruments were improved, and the guidelines for parturient care were perfected, and the increasing recognition for better medical service of parturients all were potential confounders for the difference detection of CS rate. In addition, we merely investigated the nulliparous women with single and vertex presentation, but whether such results could be applied to other populations is not known. Another question is the difficulty in blinding the study treatment groups from the obstetricians who ultimately made the decision for Cesarean delivery. Finally, we monitored the fetal heart-rate as a possible indicator for emergency Cesarean delivery, although the association among epidural analgesia, fetal heart-rate variability, and Cesarean delivery was not analyzed. In our study, the occurrence of an unfavorable fetal heart rate in both of the groups are similar (data are not shown).

In summary, this study provides robust evidence that epidural analgesia in the latent phase (at the cervical dilation of at least 1.0 cm) of the first stage of spontaneous labor at term does not prolong the progression of labor and not increase the rate of Cesarean section in nulliparous women compared with the active phase analgesia at the cervical dilation 4.0 cm or greater.

The authors thank all obstetricians, midwives, pediatricians, nurses, and anesthesiologists who participated in collecting the data in our hospital, and they are grateful for the generous help from all statisticians at the Nanjing Medical University, Nanjing, China. They also thank YuFeng Wang, M.A., Senior Librarian and Staff Literature Archiving, Library of Southeast University Medical School, Nanjing, China, for her help in performing literature search.

1.
Anim-Somuah M, Smyth R, Howell C: Epidural versus  non-epidural or no analgesia in labor. Cochrane Database Syst Rev 2005; 4:CD000331
2.
Goetzl LM: ACOG Committee on Practice Bulletins-Obstetrics: ACOG practice bulletin. Clinical management guidelines for obstetrician-gynecologists number 36, July 2002. Obstetric analgesia and anesthesia. Obstet Gynecol 2002; 100:177–91
3.
Wong CA, Scavone BM, Peaceman AM, McCarthy RJ, Sullivan JT, Diaz NT, Yaghmour E, Marcus RJ, Sherwani SS, Sproviero MT, Yilmaz M, Patel R, Robles C, Grouper S: The risk of cesarean delivery with neuraxial analgesia given early versus  late in labor. N Engl J Med 2005; 352:655–65
4.
Ohel G, Gonen R, Vaida S, Barak S, Gaitini L: Early versus  late initiation of epidural analgesia in labor: Does it increase the risk of cesarean section? A randomized trial. Am J Obstet Gynecol 2006; 194:600–5
5.
American College of Obstetricians and Gynecologists Committee on Obstetric Practice: ACOG committee opinion. No. 339: Analgesia and cesarean delivery rates. Obstet Gynecol 2006;107:1487–8
American College of Obstetricians and Gynecologists Committee on Obstetric Practice
6.
Liu EHC, Sia ATH: Rates of caesarean section and instrumental vaginal delivery in nulliparous women after low concentration epidural infusions or opioid analgesia: Systematic review. BMJ 2004; 328:1410–20
7.
Marucci M, Cinnella G, Perchiazzi G, Brienza N, Fiore T: Patient-requested neuraxial analgesia for labor: Impact on rates of cesarean and instrumental vaginal delivery. Anesthesiology 2007; 106:1035–45
8.
National Collaborating Centre for Women’s and Children’s Health: Intrapartum care-care of healthy women and their babies during childbirth. NICE Clinical Guideline 55, Royal College of Obstetricians and Gynaecologists (RCOG) Press, 2007, pp 109–73
National Collaborating Centre for Women’s and Children’s Health
NA
,
Royal College of Obstetricians and Gynaecologists (RCOG) Press
9.
National Collaborating Centre for Women’s and Children’s Health: Induction of labour. NICE Clinical Guideline 70, Royal College of Obstetricians and Gynaecologists (RCOG) Press 2008, pp 72–5
National Collaborating Centre for Women’s and Children’s Health
NA
,
Royal College of Obstetricians and Gynaecologists (RCOG) Press
10.
Pernoll ML: Course and conduct of labor and delivery, Benson & Pernoll’s Handbook of Obstetrics & Gynecology. New York, McGraw-Hill, 2001, pp 156–8Course and conduct of labor and delivery,
New York
,
McGraw-Hill
11.
Goetzl L, Cohen A, Frigoletto F Jr, Ringer SA, Lang JM, Lieberman E: Maternal epidural use and neonatal sepsis evaluation in afebrile mothers. Pediatrics 2001; 108:1099–102
12.
Zhang J, Yancey MK, Klebanoff MA, Schwarz J, Schweitzer D: Does epidural analgesia prolong labor and increase risk of cesarean delivery? A natural experiment. Am J Obstet Gynecol 2001; 185:128–34
13.
Hosmer DW, Lemeshow S: Applied Logistic Regression, 2nd edition. New York, John Wiley & Sons, 2000, pp 1–347
New York
,
John Wiley & Sons
14.
Katz MH: Multivariable analysis: A primer for readers of medical research. Ann Intern Med 2003; 138:644–50
15.
Chestnut DH, Vincent RD Jr, McGrath JM, Choi WW, Bates JN: Does early administration of epidural analgesia affect obstetric outcome in nulliparous women who are receiving intravenous oxytocin? Anesthesiology 1994; 80:1193–200
16.
Chestnut DH, McGrath JM, Vincent RD Jr, Penning DH, Choi WW, Bates JN, McFarlane C: Does early administration of epidural analgesia affect obstetric outcome in nulliparous women who are in spontaneous labor? Anesthesiology 1994; 80:1201–8
17.
Comparative Obstetric Mobile Epidural Trial (COMET) Study Group UK. Effect of low-dose mobile versus  traditional epidural techniques on mode of delivery: A randomised controlled trial. Lancet 2001; 358:19–23
Comparative Obstetric Mobile Epidural Trial (COMET) Study Group UK
18.
Wilson MJ, Cooper G, MacArthur C, Shennan A, Comparative Obstetric Mobile Epidural Trial (COMET) Study Group UK: Randomized controlled trial comparing traditional with two “mobile” epidural techniques: Anesthetic and analgesic efficacy. Anesthesiology 2002; 97:1567–75
Comparative Obstetric Mobile Epidural Trial (COMET) Study Group UK
19.
Gambling DR, Sharma SK, Ramin SM, Lucas MJ, Leveno KJ, Wiley J, Sidawi JE: A randomized study of combined spinal-epidural analgesia versus  intravenous meperidine during labor: Impact on cesarean delivery rate. Anesthesiology 1998; 89:1336–44
20.
Kotaska AJ, Klein MC, Liston RM: Epidural analgesia associated with low-dose oxytocin augmentation increases cesarean births: A critical look at the external validity of randomized trials. Am J Obstet Gynecol 2006; 194:809–14
21.
Jain S, Arya VK, Gopalan S, Jain V: Analgesic efficacy of intramuscular opioids versus  epidural analgesia in labor. Int J Gynaecol Obstet 2003; 83:19–27
22.
Tsui MH, Ngan Kee WD, Ng FF, Lau TK: A double blinded randomised placebo-controlled study of intramuscular pethidine for pain relief in the first stage of labour. BJOG 2004; 111:648–55
23.
Banerjee S, Cashman P, Yentis SM, Steer PJ: Maternal temperature monitoring during labor: Concordance and variability among monitoring sites. Obstet Gynecol 2004; 103:287–93
24.
Camann WR, Hortvet LA, Hughes N, Bader AM, Datta S: Maternal temperature regulation during extradural analgesia for labour. Br J Anaesth 1991; 67:565–8
25.
Einarson A, Maltepe C, Boskovic R, Koren G: Treatment of nausea and vomiting in pregnancy: An updated algorithm. Can Fam Physician 2007; 53:2109–11
26.
Andersen R, Krohg K: Pain as a major cause of postoperative nausea. Can Anaesth Soc J 1976; 23:366–9
27.
Chang ZM, Heaman MI: Epidural analgesia during labor and delivery: Effects on the initiation and continuation of effective breastfeeding. J Hum Lact 2005; 21:305–14
28.
Montgomery A, Hale TW: Academy Of Breastfeeding Medicine Protocol Committee: ABM clinical protocol #15: Analgesia and anesthesia for the breastfeeding mother. Breastfeed Med 2006; 1:271–7
29.
Volmanen P, Valanne J, Alahuhta S: Breast-feeding problems after epidural analgesia for labour: A retrospective cohort study of pain, obstetrical procedures and breast-feeding practices. Int J Obstet Anesth 2004; 13:25–9
30.
Jordan S: Infant feeding and analgesia in labour: The evidence is accumulating. Int Breastfeed J 2006; 1:25
31.
Wiklund I, Norman M, Uvnäs-Moberg K, Ransjö-Arvidson AB, Andolf E: Epidural analgesia: Breast-feeding success and related factors. Midwifery 2009; 25:e31–8
32.
Beilin Y, Bodian CA, Weiser J, Hossain S, Arnold I, Feierman DE, Martin G, Holzman I: Effect of labor epidural analgesia with and without fentanyl on infant breast-feeding: A prospective, randomized, double-blind study. Anesthesiology 2005; 103:1211–7
33.
Fisler RE, Cohen A, Ringer SA, Lieberman E: Neonatal outcome after trial of labor compared with elective repeat cesarean section. Birth 2003; 30:83–8
34.
Smith GC, Celik E, To M, Khouri O, Nicolaides KH: Fetal Medicine Foundation Second Trimester Screening Group: Cervical length at mid-pregnancy and the risk of primary cesarean delivery. N Engl J Med 2008; 358:1346–53
Fetal Medicine Foundation Second Trimester Screening Group

Appendix 1: Members Of The Labor Analgesia Examining Group (LAEG)

ShiQin Xu, M.Sc., Associate Professor, YuSheng Liu, M.D., Lecturer, QingSong Zhao, M.Sc., Staff Anesthesiologist, ShanWu Feng, M.D., Research Assistant, Staff Anesthesiologist, Li Ma, M.Sc., Staff Anesthesiologist, XianQiang Yao, M.Sc., Lecturer, HongMei Yuan, B.S., Staff Anesthesiologist, XiaoHong Li, B.S., Research Assistant, LiPing Zhao, M.Sc., Associate Professor, WanGen Wang, B.S., Staff Anesthesiologist, the Department of Anesthesiology; Ru Liu, M.D., Lecturer, the Institute of Pediatrics; ManHong Cai, M.D., Associate Professor, Rong Shen, M.D., Professor, the Department of Obstetrics and Gynecology; DongYing Fu, M.Sc., Senior Administrating Nurse, the Operating Theater – all in the Affiliated Nanjing Maternity and Child Health Care Hospital, Nanjing Medical University, Nanjing, Jiangsu, China.