Secondary Logo

Journal Logo

Original Research Articles: Original Clinical Research Report

Comparison of Programmed Intermittent Epidural Boluses With Continuous Epidural Infusion for the Maintenance of Labor Analgesia: A Randomized, Controlled, Double-Blind Study

Ojo, Oluremi A. BS; Mehdiratta, Jennifer E. MD; Gamez, Brock H. BS; Hunting, John MHS; Habib, Ashraf S. MBBCh, MSc, MHSc, FRCA

Author Information
doi: 10.1213/ANE.0000000000004104
  • Free

Abstract

KEY POINTS

  • Question: Does maintenance of labor epidural analgesia with programmed intermittent epidural boluses result in reduced patient-controlled epidural analgesia use compared to continuous epidural infusion in women receiving neuraxial labor analgesia using a commercially available pump?
  • Findings: There was no difference in patient-controlled epidural analgesia consumption per hour (primary outcome) or improvement in any of the secondary outcomes relating to labor analgesia or obstetric outcomes with programmed intermittent epidural boluses compared with continuous epidural infusion except for less motor block with programmed intermittent epidural boluses.
  • Meaning: Under the conditions of the study, outcomes were not improved with programmed intermittent epidural boluses compared to continuous epidural infusion except for less motor block with programmed intermittent epidural boluses.

Neuraxial techniques provide safe and effective pain relief and are widely used for labor analgesia. Historically, maintenance of epidural analgesia involved intermittent provider-administered boluses, patient-controlled epidural analgesia, and continuous epidural infusions with or without patient-controlled epidural analgesia.1,2 Continuous epidural infusion has been shown to be highly effective, providing consistent analgesia, improved patient satisfaction, and reduced workload for the anesthesia providers compared to provider-administered boluses.3,4 Continuous epidural infusion, however, is associated with greater local anesthetic consumption compared to intermittent boluses, which may increase motor block causing reduced mobility, decreased pelvic muscle tone, and impaired ability of the parturient to bear down during the second stage of labor.3 Motor blockade might also increase the risk of shoulder dystocia and instrumental deliveries.5

A newer approach to maintenance of labor epidural analgesia involves the administration of programmed intermittent epidural boluses with patient-controlled epidural analgesia dosing for breakthrough pain. Earlier studies have suggested that programmed intermittent epidural boluses may be superior to continuous epidural infusion for labor analgesia and may be associated with a reduction in local anesthetic requirement,6–8 decreased motor blockade,9 lower risk of instrumental deliveries,9 and improved patient satisfaction.6,7,10 This may be related to increased pressure generated during automated boluses, resulting in improved distribution of the local anesthetic in the epidural space.7 Notably, many of the earlier studies employed a 2-pump approach, one to administer programmed intermittent epidural boluses or continuous epidural infusion and another to administer patient-controlled epidural analgesia for breakthrough pain, or they involved research pumps not commercially available, which limits their application to everyday clinical practice.7,9,10

Pump technology capable of coadministering programmed boluses with patient-controlled epidural analgesia has now become available. A few studies compared programmed intermittent epidural boluses with continuous epidural infusion in the setting of commercially available pumps but reported conflicting results.1,11,12 Therefore, we performed this prospective, double-blind, randomized, controlled study employing a single-pump delivery system to compare programmed intermittent epidural boluses with continuous epidural infusion. We hypothesized that programmed intermittent epidural boluses would be associated with reduced patient-controlled epidural analgesia usage compared with continuous epidural infusion in women receiving labor epidural analgesia.

METHODS

This prospective, randomized, double-blind clinical trial was approved by the Duke University Institutional Review Board on November 1, 2016. The protocol was registered at ClinicalTrials.gov (NCT02949271; principal investigator, Ashraf S. Habib; registration date, October 31, 2016). Registration occurred before the start of the trial and any patient enrollment undertaken. This article adheres to the Consolidated Standards of Reporting Trials guidelines.

Following written informed consent, nulliparous or parous American Society of Anesthesiologists physical status II and III women >18 years of age, at gestational age >36 weeks, with singleton pregnancies, vertex presentation, in labor, desiring epidural labor analgesia at cervical dilation between 2 and 7 cm, and reporting a verbal pain score >5 (0–10 scale, 0 = no pain, 10 = worst possible pain) were enrolled in this study. The following changes were made to inclusion criteria following study registration, but before study initiation, to enhance enrollment and increase generalizability: inclusion of both nulliparous and parous women rather than only nulliparous women and increase cervical dilation at enrollment from 2 to 5 to 2 to 7 cm. Women with body mass index >50 kg/m2, history of IV drug use or opioid abuse, allergies to local anesthetics, and/or conditions requiring assisted second stage of labor were excluded from this study. After obtaining informed consent, subjects were allocated to a study arm (continuous epidural infusion or programmed intermittent epidural boluses) by computer-generated random assignment placed in sequentially numbered sealed opaque envelopes. The patient, anesthesia provider, and outcome assessor were blinded to the randomization assignment.

Epidural catheters were placed in the sitting position at the L3/4 or L4/5 interspace with a 17-gauge Tuohy needle using loss of resistance to saline technique, and 4 cm of a multiport catheter was left in the epidural space. If the subject experienced an inadvertent dural puncture during epidural placement, she was removed from the study. Epidural analgesia was initiated and maintained with a solution of 0.1% ropivacaine with 2 µg/mL fentanyl. After the initial epidural loading dose of 20 mL was administered in an incremental fashion of 5 mL every 2 minutes, the CADD Solis Ambulatory Infusion Pump (Smiths Medical, Minneapolis, MN) was programmed to deliver either 6 mL programmed intermittent epidural boluses every 45 minutes with the first bolus administered 30 minutes after epidural initiation or continuous epidural infusion at 8 mL/h beginning immediately after the loading dose. Standard tubing for the CADD Solis Pump was used with a flow rate of 250 mL/h for the programmed intermittent epidural boluses, as well as the patient-controlled epidural analgesia boluses in both groups. The pump was programmed by an anesthesia provider not involved in data collection and covered by an opaque paper. All study participants were provided with patient-controlled epidural analgesia set to 8 mL boluses with a 10-minute lockout period and instructed not to use the patient-controlled epidural analgesia until 30 minutes after the loading dose. The maximum 1-hour limit was 45 mL. The lockout interval was also 10 minutes between the patient-controlled epidural analgesia and programmed intermittent epidural boluses. This means that a patient-controlled epidural analgesia request that fell within the 10-minute lockout of a programmed intermittent epidural bolus, or a patient-controlled epidural analgesia bolus was denied. Similarly, a delivered patient-controlled epidural analgesia dose delayed the beginning of the next programmed intermittent epidural bolus by the 10-minute patient-controlled epidural analgesia lockout time. A script was used to provide standardized explanation about the use of the patient-controlled epidural analgesia to each subject. As per standard of care at our institution, noninvasive blood pressure measurements were performed every 2 minutes for 15 minutes and then every 15 minutes afterward in addition to continuous heart rate and pulse oximetry. If a parturient did not achieve adequate analgesia at 30 minutes, defined as a pain score >4 with a need for catheter manipulation or replacement, the catheter was considered unsatisfactory and the subject was withdrawn from the study. At 30 minutes after administration of the epidural loading dose, we recorded verbal pain scores, sensory level to ice, and motor block with the modified Bromage scale, where a score of 1 = complete block (unable to dorsiflex the ankle), 2 = almost complete block (able to dorsiflex ankle only), 3 = partial block (able to move knees only), 4 = detectable weakness of hip flexion while supine (unable to raise extended leg), and 5 = no detectable weakness (able to raise extended leg).13 Sensory levels to ice were assessed on both sides starting from the sacral segments until the woman felt the cold sensation similar to the sensation of ice on her arm. Maternal pain scores, motor blockade, and occurrence of hypotension requiring vasopressor treatment were recorded every 2 hours. If the patient had inadequate analgesia, despite activating the patient-controlled epidural analgesia twice in the last 20 minutes, physician boluses were administered using 5 mL of 0.2% ropivacaine every 10 minutes. If the patient did not have adequate analgesia with bilateral sensory levels after 10 mL of 0.2% ropivacaine, the catheter was considered unsatisfactory and the patient was removed from the study. If there was bilateral sensory level to at least T10 after the ropivacaine boluses and the patient complained of pressure, 100 µg epidural fentanyl was administered. After delivery, information was collected from the pump about patient-controlled epidural analgesia usage. Each participant was evaluated during the day at least 24 hours postpartum to assess for adverse effects including neurological deficits or postdural puncture headache and for satisfaction with labor analgesia using 1–5 scale with 1 = completely dissatisfied, 2 = dissatisfied, 3 = neutral, 4 = satisfied, and 5 = completely satisfied.

The primary outcome of the study was the total volume of local anesthetic received through patient-controlled epidural analgesia per hour. Secondary outcomes included need for physician interventions per hour, volume of clinician boluses required per hour, patient-controlled epidural analgesia attempts per hour, ratio of patient-controlled epidural analgesia attempts/given per hour, motor blockade measured by the lowest recorded modified Bromage scale score, number of patients with hypotension, mode of delivery, labor pain scores, Apgar scores, and patient satisfaction with labor analgesia. The duration of labor analgesia was calculated as the time from pump initiation to delivery or transfer to the operating room for cesarean delivery.

Statistical Analysis

A statistician created a stratified mixed block randomization sequence with strata for nulliparous and parous women using nQuery 7.0 (Statistical Solutions Ltd, Boston, MA). Variables were assessed for general distribution, normality, and equality of variance with a Shapiro–Wilk test and F test, respectively. The primary outcome of the study, patient-controlled epidural analgesia consumption per hour, was analyzed using Wilcoxon rank sum test. Continuous secondary outcomes were analyzed using Wilcoxon rank sum test or t test as appropriate, depending on distributional assumptions. A per-protocol analysis was performed. Normally distributed data are reported as mean (SD) and nonnormally distributed data as median (interquartile range). Categorical secondary outcomes were analyzed using χ2 test or Fisher exact test in the presence of expected cell frequencies under 5. Pain and Bromage scores were assessed longitudinally between groups with linear mixed modeling while accounting for patient-level clustering (random intercept) under a compound symmetry correlation structure. The models consisted of main effects for treatment group and time. We tested for the presence of group-by-time interaction in the models but only retained the effect if found to be significant. When the interaction was statistically significant, we performed pairwise comparisons between groups at different time points and adjusted for multiple comparisons with the Dunnett method. A supplemental analysis was performed to investigate the potential association between parity and the primary outcome of patient-controlled epidural analgesia volume per hour while also accounting for the treatment arm. As patient-controlled epidural analgesia volume per hour was not normally distributed, it was log-transformed for analysis to fit an approximate normal distribution. A 2-way ANOVA was performed to model the main effects of parity and treatment arm on the dependent variable log of patient-controlled epidural analgesia volume per hour. All data were analyzed using JMP statistical software version 13.0 and SAS statistical software version 9.4 (SAS Institute Inc, Cary, NC), and P < .05 (2-sided) was considered statistically significant. Data from a pilot study including 18 patients suggested that the mean (SD) patient-controlled epidural analgesia volume used per hour was 6.3 mL (3.4 mL) with continuous epidural infusion and 3.9 mL (3.4 mL) with programmed intermittent epidural boluses. A sample size of 44 patients per group had 91% power at α = .05 in a 2-sided 2-sample t test to identify this difference. To account for dropouts and catheter failures, we aimed to enroll 120 patients. Sample size was calculated using nQuery software version 7. While the study was powered for assessment of the primary hypothesis and strictly controls the error rate at the .05 level for that comparison, statistical evaluations of the 16 secondary outcomes were not corrected for multiple comparisons except for the longitudinal analysis of Bromage scores.

RESULTS

The study was conducted at Duke University Medical Center from November 2016 to November 2017. The flow of patients in the study is summarized in Figure 1. A total of 120 patients were included in the final analysis (59 in the continuous epidural infusion and 61 in the programmed intermittent epidural boluses arm). Eighty-six patients were primiparous and 34 were parous. No clinically important differences were apparent in patient demographics or obstetrical characteristics between the groups (Table 1).

Table 1. - Patient and Obstetrical Characteristics
Programmed Intermittent Epidural Boluses (n = 61) Continuous Epidural Infusion (n = 59)
Maternal age (y) 29 (5) 30 (5)
Height (cm) 165 (6) 165 (8)
Weight (kg) 90 (19) 88 (21)
Body mass index (kg/m2) 32.9 (7.0) 32.6 (7.2)
Gestational age (wk) 39 (1) 39 (1)
Parity, n (%)
 Nulliparous 41 (67.2) 45 (76.3)
 Parous 20 (32.8) 14 (23.7)
ASA physical status, n (%)
 II 44 (72.1) 46 (78.0)
 III 17 (27.9) 13 (22.0)
Management of labor, n (%)
 Induction of labor 46 (75.4) 44 (74.6)
 Spontaneous rupture of membranes 15 (24.6) 15 (25.4)
Oxytocin usage for labor, n (%) 49 (80.3) 44 (74.6)
Pain scores at epidural placement 8 (2) 8 (1)
Cervical dilation at epidural placement (cm) 4 (1) 4 (2)
Duration of epidural analgesia (h) 9 (5.8–14) 7.9 (5.4–12.4)
Self-reported race/ethnicity, n (%)
 Black 18 (29.5) 20 (33.9)
 White 36 (59.0) 34 (57.6)
 Asian 6 (9.8) 1 (1.7)
 Latina 1 (1.6) 4 (6.8)
Data are represented as mean (SD), median (interquartile range), or n (%).
Abbreviation: ASA, American Society of Anesthesiologists.

F1
Figure 1.:
Consolidated Standards of Reporting Trials flow diagram. CEI indicates continuous epidural infusion; PIEB, programmed intermittent epidural bolus.

Primary and secondary outcomes are listed in Tables 2 and 3. The median (interquartile range) patient-controlled epidural analgesia volume consumed per hour was not significantly different between the groups: 4.5 mL/h(3.0–8.6 mL/h) for the continuous epidural infusion group and 4.0 mL/h(2.2–7.1 mL/h) for the programmed intermittent epidural boluses group (P = .17; Figure 2). The Hodges–Lehmann location shift estimate of the difference (95% CI) from the continuous epidural infusion to the programmed intermittent epidural boluses group is 0.9 mL/h (−0.4 to 2.2 mL/h). There were also no significant differences in median time to first patient-controlled epidural analgesia request or patient-controlled epidural analgesia attempts per hour, but the ratio of patient-controlled epidural analgesia attempts to boluses given per hour was higher in the programmed intermittent epidural boluses group than in the continuous epidural infusion group (P = .03). The percentage of patients who received clinician rescue boluses was 21.3% in the programmed intermittent epidural boluses group and 23.7% in the continuous epidural infusion group (P = .83). Similarly, the number of clinician boluses and the amount of ropivacaine and fentanyl administered during manual boluses did not differ between the groups.

Table 2. - Patterns of Epidural Analgesia Use
Programmed Intermittent Epidural Boluses (n = 61) Continuous Epidural Infusion (n = 59) P
Patient-controlled epidural analgesia, vol/h (mL/h) 4.0 (2.2–7.1) 4.5 (3.0–8.6) .17
Time to first patient-controlled epidural analgesia request (min) 72 (47–236) 93 (53–185) .97
Patient-controlled epidural analgesia attempts/h 0.75 (0.33–1.84) 0.63 (0.42–1.35) .53
Patient-controlled epidural analgesia attempts/patient-controlled epidural analgesia given/h 0.17 (0.10–0.30) 0.12 (0.08–0.18) .03
Total local anesthetic received from pump (mL/h) 11.5 (9.2–14.8) 12.4 (10.5–17.0) .18
Number of clinician boluses 1 (1–2) 1 (1–1) .52
Requirement of clinician boluses 13 (21.3) 14 (23.7) .83
Clinician bolus (0.2% ropivacaine, mL/h) 1.3 (0.7) 1.0 (0.5) .22
Clinician bolus (fentanyl, µg/h) 6.4 (2.9–7.4) 11.5 (7.9–32.5) .14
Bromage score <5Table 2. 14 (27.5) 26 (50) .03
Minimum Bromage scoreTable 2. 5 (4–5) 4.5 (4–5) .03
Maximum pain score 3 (0–6) 2 (0–4) .27
Patients with hypotensionTable 2. 8 (13.1) 3 (5.1) .21
Data are median (interquartile range) or n (%), and P values are from χ2 tests or Wilcoxon rank sum tests unless otherwise noted.
aMissing for 7 continuous epidural infusion and 10 programmed intermittent epidural bolus patients, and minimum Bromage scores are excluding the 30-min time point.
bFisher exact test used.

Table 3. - Delivery Outcomes and Patient Satisfaction
Programmed Intermittent Epidural Boluses (n = 61) Continuous Epidural Infusion (n = 59) P
Delivery mode
 Spontaneous vaginal delivery 41 (67.2) 37 (62.7) .85
 Assisted vaginal delivery 5 (8.2) 5 (8.5)
 Cesarean 15 (24.6) 17 (28.8)
Duration of second stage of labor in those with spontaneous vaginal delivery (min) 44 (15–119) 63 (27–130) .47
Apgar score (min)
 1 8 (7–8) 8 (8–9) .12
 5 9 (9–9) 9 (9–9) .57
Patient satisfactionTable 3. .08
 Very satisfied 35 (70.0) 39 (81.25)
 Satisfied 9 (18.0) 3 (6.25)
 Neutral 3 (6.0) 5 (10.4)
 Dissatisfied 3 (6.0) 0 (0.0)
 Very dissatisfied 0 (0.0) 1 (2.1)
Data are median (interquartile range) or n (%), and P values are from Fisher exact tests or Wilcoxon rank sum tests.
aPatient satisfaction data are missing for 11 patients in each of the treatment groups.

F2
Figure 2.:
Patient-controlled epidural analgesia volume consumed per hour. The length of the box represents the interquartile range, with the top and bottom at the 25th and 75th percentiles, respectively. The horizontal line represents the median. The vertical lines extend to the farthest value observed within 1.5 times the interquartile range. Values outside of this range are identified by a circle. CEI indicates continuous epidural infusion; PCEA, patient-controlled epidural analgesia; PIEB, programmed intermittent epidural bolus.
Table 4. - Least-Squares Mean Estimates for Group Differences in Bromage Score From Linear Mixed Model
Hours After Epidural Placement Observations (n) Mean Difference (95% CI) Adjusted P
0.5 110 −0.023 (−0.161 to 0.114) >.99
2 87 0.120 (0.008–0.232) .215
4 73 0.264 (0.081–0.446) .030
6 53 0.407 (0.122–0.691) .033
8 41 0.550 (0.156–0.945) .040
10 27 0.694 (0.186–1.201) .047
Difference is from programmed intermittent bolus to continuous epidural infusion group mean, and multiple comparison correction was performed by Dunnett method and reported in the adjusted P column.

There were no statistically significant differences between the groups in verbal rating pain scores nor evidence of a group-by-time interaction through hour 10 of labor in the linear mixed-effects model analysis (P = .85; mean difference [95% CI], 0.05 [−0.39 to 0.48]). Although there was no difference in Bromage score between groups at 30 minutes after the loading dose (P = .74), there was a significant group-by-time interaction (P = .02) through hour 10 of labor, and the linear mixed-effect model indicated that the continuous epidural infusion group had increasingly lower Bromage scores over time than the programmed intermittent epidural boluses group. The estimated mean differences in Bromage score by time are listed in Table 4. Maximum pain scores were not different between the groups, but there were significantly more patients with Bromage score <5 in the continuous epidural infusion group compared with those in the programmed intermittent epidural boluses group (P = .03), and the minimum Bromage score was also lower in the continuous epidural infusion compared with that in the programmed intermittent epidural boluses group (P = .03). Patient satisfaction with labor analgesia was high and not different between the groups. Regarding obstetrical outcomes, there were no differences in duration of second stage of labor, Apgar scores, or mode of delivery between the 2 groups. In the supplemental post hoc analysis, a 2-way ANOVA was used to model the main effects of parity and treatment arm with the dependent variable log of patient-controlled epidural analgesia volume per hour. Both parity and treatment arm showed insignificant associations with the log of patient-controlled epidural analgesia volume per hour (P = .33 and .23, respectively).

DISCUSSION

In this study, we found no significant difference in local anesthetic consumption or pattern of patient-controlled epidural analgesia use when epidural analgesia was maintained with programmed intermittent epidural boluses compared with continuous epidural infusion of 0.1% ropivacaine with 2 µg/mL fentanyl, except for a higher ratio of patient-controlled epidural analgesia attempts/given per hour and less motor block in the programmed intermittent epidural boluses group.

It has been suggested that, when a bolus is administered through a multiport epidural catheter, the local anesthetic solution exits through all catheter orifices leading to a wider sensory block and more uniform spread throughout the epidural space compared to that with continuous epidural infusion, where the solution primarily exits at the most proximal end of the catheter, limiting the spread of the anesthetic.7 This improved local anesthetic spread with bolus dosing has been attributed to generating higher pressures compared with continuous epidural infusion.14 In fact, increased peak pressures have been reported with increasing delivery speed, being higher with multiport compared with single-port epidural catheters with programmed intermittent epidural boluses.15

In 2013, George et al6 published a meta-analysis of 9 low risk of bias randomized controlled trials comparing programmed intermittent epidural boluses with continuous epidural infusion with or without patient-controlled epidural analgesia for labor analgesia. Programmed intermittent epidural boluses were associated with a reduction in local anesthetic consumption, higher maternal satisfaction, and a reduction in the duration of second stage of labor (12 minutes shorter in the programmed intermittent epidural boluses group) when compared to the continuous epidural infusion group. However, mode of delivery, total duration of labor, and need for physician top-up doses were not significantly different between the groups.6 A more recent Cochrane review included 3 additional studies and reported a reduction in breakthrough pain, a decrease in local anesthetic consumption, and an improved maternal satisfaction with programmed intermittent epidural boluses compared to continuous epidural infusion, with no difference in any other outcomes.16 With the exception of 1 study included in the Cochrane review,17 the studies included in those meta-analyses were conducted before pumps capable of administering programmed intermittent epidural boluses with patient-controlled epidural analgesia were commercially available; therefore, the majority of the earlier studies required 2 separate pumps, 1 to administer programmed intermittent epidural boluses or continuous epidural infusion and 1 to administer patient-controlled epidural analgesia doses. Therefore, flow rates and interaction between patient-controlled epidural analgesia and the programmed intermittent boluses might differ from commercially available pumps. Specifically, when 2 pumps are used, lockout between the programmed intermittent epidural bolus and patient-controlled epidural analgesia bolus is not a factor, since the 2 pumps operate independently, whereas a lockout between the programmed intermittent epidural bolus and patient-controlled epidural analgesia bolus is present with the commercially available pumps. Other earlier studies by the Singapore group utilized research pumps that are not universally commercially available.18,19

More recently, there have been an increasing number of studies using commercially available single-pump technology to administer programmed intermittent epidural boluses or continuous epidural infusion with patient-controlled epidural analgesia for the maintenance of labor analgesia. Some of those studies have also suggested that programmed intermittent epidural boluses may be a favorable technique to continuous epidural infusion, although the results were inconsistent. A retrospective study by Tien et al1 compared programmed intermittent epidural boluses with continuous epidural infusion (0.125% bupivacaine + 2 µg/mL fentanyl) using the CADD-Solis v3.0 pump with the same patient-controlled epidural analgesia settings in all groups. Programmed intermittent epidural boluses delivered as 3 mL every 30 minutes were associated with significantly decreased patient-controlled epidural analgesia usage in comparison to continuous epidural infusion delivered at 5 mL/h and programmed intermittent epidural boluses administered as 5 mL every 60 minutes. However, local anesthetic consumption, motor blockade, and mode of delivery were not significantly different among the groups. Of note, the group with lower patient-controlled epidural analgesia demands received more programmed local anesthetic per hour than the other 2 groups. Furthermore, similar to our current study, the ratio of patient-controlled epidural analgesia attempts/given was significantly higher in both programmed intermittent epidural boluses groups compared with the continuous epidural infusion group.1 This might be a surrogate for patient discomfort. As patients become increasingly uncomfortable, they may increase their patient-controlled epidural analgesia attempts for pain control, but as a result of the lockout period and hourly maximum local anesthetic allowed, there may be an imbalance between the patient-controlled epidural analgesia attempts and delivered boluses.1 Since the patient-controlled epidural analgesia attempts, patient satisfaction scores and pain scores did not differ significantly among the groups in our study, it is possible that this outcome reflects the fact that programmed intermittent epidural bolus recipients were locked out from receiving patient-controlled epidural analgesia boluses more frequently than continuous epidural infusion recipients.

McKenzie et al11 performed a before/after retrospective impact study and reported fewer physician rescue doses in patients who received 9 mL programmed intermittent epidural boluses every 45 minutes when compared to a continuous epidural infusion rate of 12 mL/h of 0.0625% bupivacaine and 0.4 µg/mL sufentanil. However, there were no differences with regard to time to first rescue bolus, instrumental delivery rate, or pain scores.11 Patient-controlled epidural analgesia settings were not standardized among programmed intermittent epidural boluses and continuous epidural infusion recipients; continuous epidural infusion patients had their patient-controlled epidural analgesia set to 12 mL every 15 minutes, whereas programmed intermittent epidural bolus patients had their patient-controlled epidural analgesia set to 10 mL with a lockout of 10 minutes, raising the possibility that more patient-controlled epidural analgesia consumption might have contributed to the reduction in the need for rescue physician boluses. An hourly maximum local anesthetic dose was not specified, and pattern of patient-controlled epidural analgesia use was not reported in this study.

Delgado et al20 performed a retrospective study comparing continuous epidural infusion administered as 10 mL/h with programmed intermittent epidural boluses administered as 10 mL every 60 minutes, 10 mL every 45 minutes, or 10 mL every 45 minutes at a high flow rate (500 mL/h), with all groups using 0.0625% bupivacaine with 2 µg/mL fentanyl. Need for physician top ups was lowest in the programmed intermittent epidural boluses every 45 minutes and programmed intermittent epidural boluses every 45 minutes with the high flow rate settings groups, although no significant differences in other outcomes, including time to first physician top-up dose, time to delivery, and obstetric outcomes, were observed. It is important to note that the advantages of programmed intermittent epidural boluses over continuous epidural infusion were not observed when the same hourly dose of local anesthetic was administered to both groups. However, when the programmed intermittent epidural bolus interval was shortened to 45 minutes, patients were actually receiving 12.5 mL every hour when compared with continuous epidural infusion 10 mL/h, so the advantages of the programmed intermittent epidural boluses over continuous epidural infusion observed in this study may actually reflect the higher amount of local anesthetic administered to programmed intermittent epidural bolus patients. This study also did not comment on patterns of patient-controlled epidural analgesia use. Therefore, in all those recent retrospective studies performed with commercially available pumps in North America, benefits from programmed intermittent epidural boluses were modest and might have resulted from delivering more programmed local anesthetic compared to continuous epidural infusion.

A recent prospective study from Brazil by Nunes et al12 compared continuous epidural infusion administered as 0.15% ropivacaine and 0.2 µg/mL sufentanil at 5 mL/h with programmed intermittent epidural boluses using 0.1% ropivacaine and 0.2 µg/mL sufentanil delivered as 10 mL every 60 minutes or 0.15% ropivacaine and 0.2 µg/mL sufentanil at 10 mL every 60 minutes. Programmed intermittent epidural boluses were associated with reduced cesarean delivery rate, but maternal satisfaction, motor blockade, and instrumental delivery rates were not different among the groups. However, local anesthetic hourly dose was different between the groups, and the management of breakthrough pain was not standardized, with only continuous epidural infusion recipients having access to patient-controlled epidural analgesia. Programmed intermittent epidural bolus patients were encouraged to request manual boluses from nursing staff for breakthrough pain, and there was no information regarding the need for additional rescue analgesia.

Another prospective, double-blind study by Ferrer et al21 comparing continuous epidural infusion 10 mL/h and programmed intermittent epidural boluses 10 mL every 60 minutes of 0.1% bupivacaine with 2 µg/mL fentanyl found that programmed intermittent epidural boluses were associated with lower local anesthetic consumption. However, no differences were found regarding pain control, characteristics of the block, hemodynamics, side effects, or neonatal outcomes. Breakthrough pain was not managed with patient-controlled epidural analgesia but was instead managed with rescue boluses administered by a nurse who was not blinded to the patient’s group assignment, which may have introduced bias. Conversely, Lin et al17 also performed a prospective, randomized controlled study comparing a continuous epidural infusion regimen of 5 mL/h with programmed intermittent epidural boluses of 5 mL every 60 minutes of 0.1% ropivacaine plus 0.3 µg/mL sufentanil with access to patient-controlled epidural analgesia for breakthrough pain. Programmed intermittent epidural boluses were associated with lower pain scores and lower local anesthetic consumption when compared with continuous epidural infusion.

While we did not find any analgesic benefits of programmed intermittent epidural boluses in our study, there was less motor block in the programmed intermittent epidural boluses group despite the lack of difference in local anesthetic consumption. The higher risk of motor block with continuous epidural infusion has been attributed to the presence of consistently higher concentration of local anesthetic in the extraneural space than the intraneural space, increasing the concentration inside the nerve and leading to motor block, which is less likely to occur when low concentrations of local anesthetics are given intermittently in the programmed intermittent epidural boluses mode.22

The findings of our study, coupled with the modest and inconsistent benefits reported in recent studies using commercially available pumps, suggest that programmed intermittent epidural bolus settings might need to be optimized to consistently improve outcomes, since there are several variables that could be manipulated. Recent studies have investigated the optimum interval between programmed intermittent epidural bolus doses,23 bolus volume,24 and bolus flow rate.25 A larger programmed intermittent epidural bolus volume could have resulted in improved analgesic outcomes without an increase in motor block, since we saw less motor block in the programmed intermittent epidural bolus group in our study compared to the continuous epidural infusion group, but this needs to be investigated in future studies.

This study must be interpreted within the context of its limitations. We included both parous and nulliparous women in our study, but randomization was stratified by parity to ensure adequate balance between the groups. A post hoc analysis also did not suggest that parity affected the primary outcomes. Another limitation of our study is the wider range of cervical dilation that we allowed compared to that in the previous studies. However, most patients (104/120 [87%]) received epidural analgesia with cervical dilations between 2 and 5 cm. We also did not correct for the multiple secondary outcomes included in our study; therefore, the findings of higher patient-controlled epidural analgesia demands/given per hour and less motor block in the programmed intermittent epidural boluses group must be interpreted with caution. Further, while the study was powered for the primary outcome, it might not have been adequately powered for our secondary outcomes. Finally, this study was well powered to detect a difference of 2.4 mL/h in the primary outcome, but there may be smaller differences, like that of the upper bound of treatment difference of 2.2 mL/h, which could be deemed clinically meaningful.

In summary, under the conditions of our study, we did not find significantly improved outcomes with programmed intermittent epidural boluses compared to continuous epidural infusion except for less motor block with programmed intermittent epidural boluses. Future well-designed prospective studies should assess whether smaller but clinically important differences exist, to further investigate possible differences in motor block, and evaluate different parameters of programmed intermittent epidural boluses to optimize analgesia and outcomes.

DISCLOSURES

Name: Oluremi A. Ojo, BS.

Contribution: This author helped collect the data, analyze the data, prepare the manuscript, and finally approve the version to be published.

Conflicts of Interest: None.

Name: Jennifer E. Mehdiratta, MD.

Contribution: This author helped design the study, collect the data, prepare the manuscript, and finally approve the version to be published.

Conflicts of Interest: None.

Name: Brock H. Gamez, BS.

Contribution: This author helped collect the data, prepare the manuscript, and finally approve the version to be published.

Conflicts of Interest: None.

Name: John Hunting, MHS.

Contribution: This author helped analyze the data, prepare the manuscript, and finally approve the version to be published.

Conflicts of Interest: None.

Name: Ashraf S. Habib, MBBCh, MSc, MHSc, FRCA.

Contribution: This author helped design the study, collect the data, analyze the data, prepare the manuscript, and finally approve the version to be published.

Conflicts of Interest: A. S. Habib is a senior editor for Anesthesia & Analgesia

This manuscript was handled by: Jill M. Mhyre, MD.

REFERENCES

1. Tien M, Allen TK, Mauritz A, Habib AS. A retrospective comparison of programmed intermittent epidural bolus with continuous epidural infusion for maintenance of labor analgesia. Curr Med Res Opin. 2016;32:1435–1440.
2. Sng BL, Sia ATH. Maintenance of epidural labour analgesia: the old, the new and the future. Best Pract Res Clin Anaesthesiol. 2017;31:15–22.
3. Boutros A, Blary S, Bronchard R, Bonnet F. Comparison of intermittent epidural bolus, continuous epidural infusion and patient controlled-epidural analgesia during labor. Int J Obstet Anesth. 1999;8:236–241.
4. Lamont RF, Pinney D, Rodgers P, Bryant TN. Continuous versus intermittent epidural analgesia: a randomised trial to observe obstetric outcome. Anaesthesia. 1989;44:893–896.
5. Thornton JG, Capogna G. Reducing likelihood of instrumental delivery with epidural anaesthesia. Lancet. 2001;358:2.
6. George RB, Allen TK, Habib AS. Intermittent epidural bolus compared with continuous epidural infusions for labor analgesia: a systematic review and meta-analysis. Anesth Analg. 2013;116:133–144.
7. Wong CA, Ratliff JT, Sullivan JT, Scavone BM, Toledo P, McCarthy RJ. A randomized comparison of programmed intermittent epidural bolus with continuous epidural infusion for labor analgesia. Anesth Analg. 2006;102:904–909.
8. Fettes PD, Moore CS, Whiteside JB, McLeod GA, Wildsmith JA. Intermittent vs continuous administration of epidural ropivacaine with fentanyl for analgesia during labour. Br J Anaesth. 2006;97:359–364.
9. Capogna G, Camorcia M, Stirparo S, Farcomeni A. Programmed intermittent epidural bolus versus continuous epidural infusion for labor analgesia: the effects on maternal motor function and labor outcome: a randomized double-blind study in nulliparous women. Anesth Analg. 2011;113:826–831.
10. Leo S, Ocampo CE, Lim Y, Sia AT. A randomized comparison of automated intermittent mandatory boluses with a basal infusion in combination with patient-controlled epidural analgesia for labor and delivery. Int J Obstet Anesth. 2010;19:357–364.
11. McKenzie CP, Cobb B, Riley ET, Carvalho B. Programmed intermittent epidural boluses for maintenance of labor analgesia: an impact study. Int J Obstet Anesth. 2016;26:32–38.
12. Nunes J, Nunes S, Veiga M, Cortez M, Seifert I. A prospective, randomized, blinded-endpoint, controlled study—continuous epidural infusion versus programmed intermittent epidural bolus in labor analgesia [in Portuguese]. Rev Bras Anestesiol. 2016;66:439–444.
13. Breen TW, Shapiro T, Glass B, Foster-Payne D, Oriol NE. Epidural anesthesia for labor in an ambulatory patient. Anesth Analg. 1993;77:919–924.
14. Carvalho B, George RB, Cobb B, McKenzie C, Riley ET. Implementation of programmed intermittent epidural bolus for the maintenance of labor analgesia. Anesth Analg. 2016;123:965–971.
15. Klumpner TT, Lange EM, Ahmed HS, Fitzgerald PC, Wong CA, Toledo P. An in vitro evaluation of the pressure generated during programmed intermittent epidural bolus injection at varying infusion delivery speeds. J Clin Anesth. 2016;34:632–637.
16. Sng BL, Zeng Y, de Souza NNA, et al. Automated mandatory bolus versus basal infusion for maintenance of epidural analgesia in labour. Cochrane Database Syst Rev. 2018;5:CD011344.
17. Lin Y, Li Q, Liu J, Yang R, Liu J. Comparison of continuous epidural infusion and programmed intermittent epidural bolus in labor analgesia. Ther Clin Risk Manag. 2016;12:1107–1112.
18. Sia AT, Lim Y, Ocampo C. A comparison of a basal infusion with automated mandatory boluses in parturient-controlled epidural analgesia during labor. Anesth Analg. 2007;104:673–678.
19. Sia AT, Leo S, Ocampo CE. A randomised comparison of variable-frequency automated mandatory boluses with a basal infusion for patient-controlled epidural analgesia during labour and delivery. Anaesthesia. 2013;68:267–275.
20. Delgado C, Ciliberto C, Bollag L, Sedensky M, Landau R. Continuous epidural infusion versus programmed intermittent epidural bolus for labor analgesia: optimal configuration of parameters to reduce physician-administered top-ups. Curr Med Res Opin. 2018;34:649–656.
21. Ferrer LE, Romero DJ, Vásquez OI, Matute EC, Van de Velde M. Effect of programmed intermittent epidural boluses and continuous epidural infusion on labor analgesia and obstetric outcomes: a randomized controlled trial. Arch Gynecol Obstet. 2017;296:915–922.
22. Capogna G, Stirparo S. Techniques for the maintenance of epidural labor analgesia. Curr Opin Anaesthesiol. 2013;26:261–267.
23. Epsztein Kanczuk M, Barrett NM, Arzola C, et al. Programmed intermittent epidural bolus for labor analgesia during first stage of labor: a biased-coin up-and-down sequential allocation trial to determine the optimum interval time between boluses of a fixed volume of 10 ml of bupivacaine 0.0625% with fentanyl 2 mug/ml. Anesth Analg 2017;124:537–541.
24. Zakus P, Arzola C, Bittencourt R, Downey K, Ye XY, Carvalho JC. Determination of the optimal programmed intermittent epidural bolus volume of bupivacaine 0.0625% with fentanyl 2 μg.ml-1 at a fixed interval of forty minutes: a biased coin up-and-down sequential allocation trial. Anaesthesia. 2018;73:459–465.
25. Lange EMS, Wong CA, Fitzgerald PC, et al. Effect of epidural infusion bolus delivery rate on the duration of labor analgesia: a randomized clinical trial. Anesthesiology. 2018;128:745–753.
Copyright © 2019 International Anesthesia Research Society