Neuraxial techniques are the most effective form of labor analgesia, with significant rates of epidural (EPL) and combined spinal epidural (CSE) technique use in contemporary obstetric anesthesia practice. Although the medication combinations used in these techniques have evolved, the onset, block quality, and side effect profiles of both techniques warrant further optimization.
The EPL technique has minimal adverse effects but can be associated with slow onset and occasional, variable blockade qualities including inadequate sacral spread, unilateral or patchy sensory blockade, motor impairment, and epidural catheter failure.1,2 By contrast, the CSE technique provides rapid onset of uniform sensory blockade with excellent sacral coverage, but is associated with greater side effect profiles, including maternal pruritus, fetal bradycardia and delayed functional testing of the epidural catheter.3
The dural puncture epidural (DPE) technique is performed by creating a single dural perforation via a spinal needle placed through the shaft of an epidural needle, followed by placement of a catheter into the epidural space. However, unlike the CSE technique, where medications are directly administered through the spinal needle into the subarachnoid space, all medications for analgesia or anesthesia are introduced through the epidural catheter into the epidural space. The dural puncture creates a conduit for translocation of medications from the epidural to subarachnoid spaces, a process that is believed to be responsible for the unique characteristics that are observed with the DPE technique.4
When compared with the EPL technique, the DPE technique has been demonstrated to improve sacral block onset and spread of anesthesia and analgesia1,4; these properties are particularly advantageous in obstetric patients. In addition, the process of creating a dural puncture with a spinal needle through an epidural needle uses cerebrospinal fluid (CSF) return as a “confirmatory,” definitive end point for the likely positioning of the epidural needle tip within the epidural space. Furthermore, by avoiding direct intrathecal administration of medication, the DPE technique may have fewer adverse effects compared with the CSE technique.5,6 To date, all 3 techniques have not been compared.
We designed this prospective randomized, double-blind study to determine if the DPE technique would result in improved labor analgesia onset and block characteristics compared with the EPL technique, with fewer maternal and fetal side effects compared with the CSE technique. We hypothesized that the onset of labor analgesia would be most rapid with the CSE, followed by the DPE, and slowest with the EPL techniques. We also sought to determine if overall analgesia characteristics and side effects would favor the DPE technique.
This study was approved by the Institutional Research Ethics Board of Partners Healthcare, Brigham and Women’s Hospital, Boston, MA, and registered with the www.clinicaltrials.gov protocol registration system (NCT02008591, PI: Lawrence C. Tsen, first registered on November 26, 2013). All participants provided written informed consent. Healthy pregnant women with singleton, vertex presentation fetuses at 38 to 42 weeks’ gestation in active labor with cervical dilatation <5.0 cm and desiring epidural labor analgesia at Brigham and Women’s Hospital from January 2014 to November 2015 were eligible. We excluded subjects with diseases of pregnancy (eg, gestational hypertension, preeclampsia, or gestational diabetes), contraindications to neuraxial analgesia techniques, known fetal anomalies, or conditions associated with an increased risk of cesarean delivery (eg, vaginal birth after cesarean delivery, history of uterine rupture).
Randomization and Concealment of Group Assignments
On request for epidural labor analgesia, an attending or fellow anesthesiologist was notified to perform the neuraxial placement. Before entering the patient room, the anesthesiologist was informed of the study protocols, opened a sealed envelope containing a computer-generated (Microsoft Excel 2010, Microsoft, Redmond, WA) randomized assignment to 1 of 3 groups (ie, CSE, DPE, or EPL), and retrieved the appropriate medications. The anesthesiologist and bedside nurse were instructed not to reveal the randomization arm to the patient or study coinvestigators. A study coinvestigator waited outside the patient room and only entered upon notification from the nurse, at the request of the anesthesiologist following the completion of the neuraxial placement protocol.
Epidural Technique Placement Protocol
Before neuraxial placement, all subjects received 30 mL of sodium citrate orally and had an 18-G intravenous (IV) catheter placed with automated noninvasive blood pressure, pulse oximetry, and external tocodynamometry monitors applied. All subjects received a 500- to 1000-mL IV bolus of lactated Ringer’s solution over 15 minutes immediately before the initiation of neuraxial analgesia.
The epidural space was identified in seated position at the L2-L3 or L3-L4 interspace via the midline approach with a 17-G, 8.9-cm Weiss epidural needle using a loss of resistance to saline technique. In subjects randomly assigned to DPE and CSE techniques, a needle-through-needle technique was performed using a 25-G, 11.9-cm Whitacre spinal needle placed into the shaft of the previously sited epidural needle to create a single dural puncture with confirmation of free-flow CSF. The spinal needle protruded 1.2 cm beyond the epidural needle tip when fully inserted. A 19-G FlexTip Plus single open end catheter (Arrow International, Reading, PA) was placed 5 cm into the epidural space (Figure 1). After a negative aspiration for blood and CSF, all patients received initial dosing regimens as per study protocol based on current institutional practices. A lidocaine test dose was not used. Initial dosing for the EPL and DPE techniques consisted of 20 mL of 0.125% bupivacaine with fentanyl 2 μg/mL fractionated into four 5-mL boluses given over 5 minutes through the catheter; for the CSE technique, initial dosing consisted of bupivacaine 1.7 mg and fentanyl 17 μg (1 of 1.5 mL premixed solution of 0.25% bupivacaine 2.5 mg and fentanyl 25 μg). Upon completion of initial dosing in all groups, patient-controlled epidural analgesia (PCEA) was initiated immediately using the following parameters: bupivacaine 1.25 mg/mL with fentanyl 2 μg/mL, background infusion at 6 mL/h, demand dose of 6 mL, lockout interval of 15 minutes, and hourly limit of 20 mL.
Any procedural variance or complications that occurred during epidural technique placement or postpartum including no analgesia within the first 30 minutes, accidental dural puncture by the epidural needle, tactile dural puncture by the spinal needle without return of CSF, intravascular or intrathecal catheter insertion, delay or inability to thread catheter, postdural puncture headache, persistent paresthesia, or persistent back pain were documented by the anesthesiologist or nonstudy anesthesia provider in the study database. All study subjects were visited postpartum day 1 by an independent anesthesia provider not involved in the study who assessed the presence of headache, back pain, paresthesia, or other complications as per usual care using a structured follow-up form. Outcomes for subjects who experienced procedural variance or complications were recorded and analyzed according to the intention-to-treat principle.
Blinding and Outcome Assessments
Immediately after initial dosing was completed, while the patient was being assisted to the lateral recumbent position, the blinded study coinvestigator was summoned into the room and began outcome assessments (Figure 1). This was designated as time 0 (t = 0) and all outcomes were assessed at standardized times (ie, t = 0.5, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, and 30 minutes). Following the first 30 minutes, assessments continued at 90-minute intervals until delivery.
Analgesia was evaluated using the verbal numeric pain rating scale (NPRS) from 0 to 10 after every uterine contraction. Sensory blockade was evaluated using a nontraumatic pinprick stimulus starting at the S2 dermatome and moving in a caudad to cephalad direction. The lowest segment where the patient perceived the stimulus to be identical to the ipsilateral deltoid was recorded. In the lower extremity, dermatomal levels were assessed by stimulating the inguinal crease at the midclavicular line (L1), anteromedial thigh (L2), medial femoral condyle above the knee (L3), medial malleolus (L4), dorsum web between great and second toe (L5), lateral calcaneus (S1), and midpoint of the popliteal fossa (S2).7 On the torso, dermatomal levels were assessed in the midclavicular line. Motor strength was assessed with a modified Bromage score (0 = full flexion of knees and ankles, 1 = partial flexion of knees, full flexion of ankles, 2 = inability to flex knees and partial flexion of ankles, and 3 = inability to flex knees and ankles). Presence of motor blockade was defined as modified Bromage scale score ≥1; the highest Bromage score was recorded. Pruritus and nausea were evaluated by directly asking the subject for presence and severity graded on a scale ranging from 0 = none, 1 = mild, 2 = moderate, and 3 = severe. Temperature was taken orally using a standardized, calibrated, automated temperature probe permanently mounted to the wall of each labor room (WelchAllyn, Suretemp Plus 692, Skaneateles Falls, NY). Hypotension was defined as systolic blood pressure ≤20% from the admission blood pressure. In the presence of hypotension, the blood pressure monitor was reset to cycle every 1 minute, and treated with a single fluid bolus of 250 to 500 mL and IV phenylephrine 80 μg every 2 minutes as required; between doses, if hypotension persisted and the patient displayed symptoms (eg, nausea, light headedness), an additional “rescue” dose of IV phenylephrine 40 μg was administered. Asymmetric blockade was defined as difference in sensory blockade greater than 2 dermatomal levels between the left and right side of the patient.
Inadequate Analgesia Protocol
If analgesia was inadequate (defined as patient request for supplemental analgesia beyond self-administered PCEA boluses) at 30 minutes following epidural technique placement, an anesthesia provider blinded to group assignment assessed and, if warranted, administered top-up doses according to the following physician top-up bolus protocol. The initial intervention consisted of 6 mL of bupivacaine 1.25 mg/mL with fentanyl 2 μg/mL via the PCEA infusion pump. Inadequate analgesia after 10 minutes resulted in the catheter being pulled back 1 cm and administration of 10 mL of bupivacaine 1.25 mg/mL with fentanyl 2 μg/mL as a manual bolus over a period of 5 minutes. Inadequate analgesia after another 10 minutes resulted in 10 mL of bupivacaine 2.5 mg/mL being given as a manual bolus over a period of 5 minutes. In the event of failed block, initially or at a later time (defined as no change in NPRS response after interventions), catheter replacement would be discussed with the patient. All catheter adjustments and boluses were recorded.
Analysis of Uterine Contractions and Fetal Heart Tracings
Continuous uterine contractions and fetal heart tracings were stored on the hospital patient electronic system and retrospectively interpreted by 2 independent obstetricians who were not involved in the clinical care of study participants and blinded to group assignments and study outcomes. Disagreements in their evaluations were resolved with review of the tracings and discussion until consensus was obtained.
Using an approach previously described,8 uterine contraction and fetal heart rate (FHR) monitoring patterns were extracted in 10-minute epochs, starting 1 hour before and up to 1 hour after completion of initial dosing of epidural analgesia. Baselines were mean values of the six 10-minute epochs before and after epidural placement.
All classifications of uterine contraction and FHR frequencies and abnormalities were based on definitions developed by the National Institute of Child Health and Human Development (NICHD).9 Quantitative assessment of uterine contractions included frequency, presence of uterine tachysystole, and hypertonus. Uterine tachysystole was defined as >5 contractions in 10 minutes, averaged over a 30-minute window. Uterine hypertonus was defined as a single contraction lasting longer than 2 minutes.9,10 Presence of uterine tachysystole and hypertonus were further qualified as to the presence or absence of associated FHR decelerations.
Quantitative assessment of fetal heart tracings included baseline heart rate, variability, accelerations, and decelerations. Each type of deceleration (early, late, or variable) was described and listed as the mean number present. Finally, the obstetricians assigned a category to the fetal heart tracings before and after the epidural placement based on the 3-tier NICHD system.
The primary outcome of the study was time to NPRS ≤ 1 between the DPE and EPL groups. The choice of this outcome was based on a previous study by Beilin et al11 demonstrating that very few patients with NPRS 0 or 1 would desire more medication for labor epidural analgesia compared with those with NPRS > 1. We estimated 95% and 75% of subjects in DPE and EPL, respectively, would achieve numeric pain rating scale ≤ 1 at 30 minutes using data from our previous study. We assumed a constant proportional hazard between DPE and EPL throughout the 30-minute follow-up time. We determined a constant hazard ratio of 2.16 (ln[0.05]/ln[0.25]) between the 2 groups. Based on these, we would need 53 events from approximately 63 subjects in total (32 per group) to have 80% power (2-tailed) to detect a difference in survival curves between groups using a Cox-proportional hazard model with α = .05. In keeping with the total number of subjects enrolled in our previous study, and to account for any potential study issues (eg, dropouts), the sample size was arbitrarily increased to 40 per group (80 in total). The total study population (n = 120) included a same-sized CSE group (n = 40), given the clinical acknowledgment of significantly faster onset of CSE analgesia relative to the other techniques, but allowing for an analysis of secondary outcomes.
We predetermined a subset of secondary outcomes to be analyzed; specifically, these secondary outcomes included the incidence of bilateral S2 blockade, physician top-up intervention, asymmetric block, nausea, pruritus, hypotension, presence of motor block, combined uterine tachysystole and hypertonus, fetal bradycardia, and NICHD fetal heart tracing category progression.
Kaplan-Meier curves and Cox proportional hazard model were used to analyze the primary outcomes. For the secondary outcomes, risk ratios, 95% confidence intervals, and P values based on χ2 tests were reported. When the expected number in one or more cells in the tabulated tables of secondary outcomes was ≤5, we further applied Yates correction to preserve the statistical validity.
For primary outcomes, the significance threshold was set to be P < .025, because 2 comparisons were performed (DPE versus EPL and DPE versus CSE). For the secondary outcomes, the significance threshold was set to be P < .05. All analyses were performed with statistical software R version 3.1.2 (R Foundation, Vienna, Austria).
A total of 120 patients were recruited. Participant flow and baseline characteristics are summarized in Figure 2 and Table 1, respectively. The groups were similar at baseline with no clinically important differences. Two patients in the DPE group did not have CSF return on dural puncture. There were no other procedural complications.
There was no significant difference in the time to NPRS ≤ 1 between DPE and EPL (hazard ratio 1.4, 95% confidence interval [CI] 0.83–2.4, P = .21). DPE achieved NPRS ≤ 1 significantly slower than CSE (hazard ratio 0.36, 95% CI 0.22–0.59, P = .0001). The median times (interquartile range) to NPRS ≤ 1 were 2 (0.5–6) minutes for CSE, 11 (4–120) minutes for DPE, and 18 (10–120) minutes for EPL (Figure 3 and Table 2).
Block characteristics and maternal adverse effects, uterine contraction and fetal heart tracing assessment are summarized in Tables 3–5.
Compared with EPL, DPE had significantly greater incidence of bilateral S2 blockade at 10 minutes (risk ratio [RR] 2.13; 95% CI 1.39–3.28; P < .001), 20 minutes (RR 1.60; 95% CI 1.26–2.03; P < .001), and 30 minutes (RR 1.18; 95% CI 1.01–1.30; P = .034); lower incidence of asymmetric block after 30 minutes (RR 0.19; 95% CI 0.07–0.51; P < .001); and lower incidence of physician top-up bolus intervention (RR 0.45; 95% CI 0.23–0.86; P = .011; Table 6). The EPL technique appeared to have the greatest presence of motor blockade, followed by the DPE and CSE techniques (Table 3); however, after adjusting for duration of epidural infusion, we found no difference in the presence of motor block among the 3 techniques (Table 6).
Compared with CSE, DPE had significantly lower incidence of physician top-up bolus intervention (RR 0.45; 95% CI 0.23–0.86; P = .012), hypotension (RR 0.38; 95% CI 0.15–0.98; P = .032), pruritus (RR 0.15; 95% CI 0.06–0.38; P < .001), postneuraxial placement combined uterine tachysystole and hypertonus (RR 0.22; 95% CI 0.08–0.60; P < .001), and NICHD category I to II conversion in FHR tracing following neuraxial placement (RR 0.38; 95% CI 0.15–0.98; P = .032; Table 6).
In this study of 3 distinct labor analgesia techniques, analgesia onset was most rapid for the CSE technique with no difference between DPE and EPL techniques. Compared with the EPL technique, the DPE technique had an earlier and greater incidence of sacral coverage and a lesser incidence of asymmetric blockade and physician top-up interventions. Compared with the CSE technique, the DPE technique had a lesser incidence of maternal hypotension, pruritus, physician top-up interventions, uterine tachysystole with hypertonus, and FHR conversion from NICHD category I to II tracings. In parturients in early labor requesting neuraxial analgesia, the DPE technique provides the sacral and symmetric block characteristics of the CSE technique with the lower maternal and fetal side effect profile of the EPL technique.
The DPE technique differs from the EPL technique by the presence of a dural puncture conduit, which enables solutions administered into the epidural space to gain access to the subarachnoid space. Leach et al12 observed the translocation of epidural contrast dye into the subarachnoid space through an inadvertent dural puncture with a Tuohy needle (gauge not documented). An in vitro study demonstrated notable lidocaine flux with an 18-G Tuohy and 24-G Sprotte needle punctures, but no flux with a 27-G Whitacre needle puncture.13 Consistent with this observation, Thomas et al5 found that the DPE technique with a 27-G Whitacre did not improve labor analgesia quality compared with an EPL technique. By contrast, Suzuki et al,4 in patients undergoing surgical anesthesia with the DPE technique using a 26-G Whitacre, observed an earlier and greater sacral spread compared with the EPL technique. In our earlier labor analgesia study,1 which derived the DPE acronym, and our current investigation, we also observed improved sacral spread with the DPE technique using a 25-G Whitacre compared with the EPL technique. Sacral coverage is an important and often incomplete feature of analgesia, particularly during the second stage of labor and with instrumented deliveries.14 As importantly, the highest thoracic distribution was no different among the CSE, DPE, and EPL techniques.
The rates of bilateral block symmetry between the CSE and DPE techniques were no different, but substantively greater than with the EPL technique before and after the first 30 minutes. These findings are consistent with large retrospective15,16 and prospective studies17; a recent meta-analysis of 10 RCTs in 1722 parturients found a significant reduction in the relative risk of unilateral block for CSE versus EPL techniques.18 With an incidence of up to 8%,19,20 asymmetry of neuraxial labor analgesia blockade represents an important reason for catheter manipulation and replacement,11 and a source of patient discomfort, dissatisfaction, and potential morbidity or mortality if alternate analgesic or anesthetic interventions are required.21
A novel finding in our study was fewer physician top-up interventions with the DPE technique, when compared with both CSE and EPL techniques. Although Goodman et al22 found no difference in top-up interventions between CSE (27-G Whitacre) and EPL groups, this may have been due to the dural puncture size. A meta-analysis of CSE versus EPL labor analgesia by Heesen et al18 also found no difference in top-up interventions, but identified marked between-study heterogeneity. In our study, the lower top-up interventions with DPE compared with EPL techniques are likely related to improved block quality (eg, symmetric sensory blockade with sacral presence). By contrast, the difference between DPE and CSE techniques illustrates 2 analgesic elements that warrant further investigation. First, the notable CSE transition from initial spinal analgesia to ongoing epidural analgesia often provokes a request for physician top-up interventions. Although our findings are underpowered to robustly evaluate this outcome, we observed earlier requests for top-up interventions with the CSE technique. Second, when administered in early labor, CSE analgesia has been associated with greater uterine contractility and more rapid cervical dilation23,24; these alterations could contribute to greater analgesia requirements and fetal consequences.
The incidence in fetal bradycardia following the CSE technique has been associated with the dose of intrathecal opioids and alterations in uterine tone.6,25,26 In our study, the CSE opioid dose selected is used in our current clinical practice and acknowledges the ED95 (1.66 mg) of intrathecal plain bupivacaine with fentanyl 15 μg in active labor.27 Our dose is consistent with that used by Abrão et al26 (ie, intrathecal sufentanil 2.5 μg = fentanyl 11 μg using a relative potency ratio of 4.4:128), which observed a greater incidence of uterine tone and FHR abnormalities compared with the EPL technique. Although Abrão et al26 used an intrauterine pressure catheter, FHR tracings were only assessed for 15 minutes pre- and postanalgesia initiation; our study extended the assessment period to 60 minutes pre- and postanalgesia initiation in an attempt to capture the effect of peak analgesia following neuraxial techniques. Our findings indicate that transient uterine tachysystole and hypertonus can occur even in the absence of FHR alterations; these effects may become even more relevant in situations when fetal tolerance of transient alterations in uterine tone is limited or compromised.
In a similar parturient population, the use of an identical DPE technique (25-G Whitacre) with different EPL dosing regimens in our current (ie, 0.125% bupivacaine 20 mL) and earlier (ie, 0.25% bupivacaine 12 mL) studies allows for some indirect observations. First, the use of a more dilute, higher-volume initial EPL dose in our current study, consistent with contemporary EPL dosing regimens, was associated with a thoracic sensory blockade of more rapid onset and greater median cephalad spread (new, T4 [T2-T8]; old, T10 [T9-10]). It also yielded a sacral sensory blockade with more rapid S2 onset and greater frequency (100% vs 90% within 20 minutes). These findings are consistent with the findings of Christiaens et al,29 which demonstrated optimal epidural analgesia quality and duration when bupivacaine 20 mg was diluted into a volume of 20 mL and sequentially diminished with smaller total volumes of 10 mL and 4 mL. Their study also found greater motor blockade with greater volumes, although this was not observed in our study. The greater sacral sensory blockade found with the larger initial EPL dosing volumes is of particular value for the second stage of labor, yet often insufficient or lacking given the distance from the epidural catheter, the larger-diameter nerve fibers with thicker dura mater, and the tendency for medications introduced into the lumbar epidural space to move predominantly in a cephalad direction.30
We acknowledge several limitations in our study. First, despite every effort to maintain blinding among all participants, the rapid onset of analgesia or side effects (eg, pruritus) associated with the CSE technique and the protocol for initial drug administration (Figure 1) may have made it possible to ascertain subject assignment.22 We minimized this possibility by (1) having the coinvestigator conducting the outcome assessments enter the room only after completion of the initial dosing for all neuraxial techniques, (2) using objective patient-based outcomes at standardized time points, and (3) having obstetricians blinded to the study interventions and outcomes assess the FHR and tocodynamometer data. Second, NPRS scores for discomfort could only be obtained following uterine contractions; given the variable frequency of contractions during labor, it is possible that the NPRS ≤ 1 threshold was reached more quickly or slowly. In addition, for the small number of patients where NPRS ≤ 1 was not obtained in the first 30 minutes, the next assessment was at t = 120 minutes. However, all subjects were in active labor with a similar number on oxytocin infusions; moreover, survival analysis allowed us to examine all subject data even if the NPRS end point was not obtained within the first 30 minutes. Third, although the Bonferroni correction with reference to the total number of comparisons is occasionally applied to the consideration of secondary outcomes, we elected to retain the significance threshold for secondary outcomes as P < .05, with the suggestion that these findings be carefully considered and confirmed as primary outcomes in larger studies.31 Finally, the adjustment (5%–10%) and replacement rates (0%) of our neuraxial techniques were lower than reported in other studies;15,16,22,32–34 this was likely due to our use of only experienced (attending and fellow) anesthesiologists who were facile with all 3 techniques. Although the absence of inadvertent dural puncture and other complications are likely attributable to this study feature, they may also limit the external validity of our findings.
In summary, analgesia onset was most rapid with the CSE technique, with no difference between the DPE and EPL techniques; however, the DPE technique has fewer maternal and fetal side effects compared with the CSE technique, and improved block quality when compared with the EPL technique. We conclude that the DPE appears to offer a favorable risk-benefit ratio for initiating and maintaining analgesia in laboring parturients.
This study was supported in part by the University of British Columbia Clinician Investigator Program and the Harvard Catalyst Clinical and Translational Science Center. The authors are indebted to their obstetrics, nursing, midwifery, and anesthesia colleagues for their support and assistance during this study.
Name: Anthony Chau, MD, MMSc, FRCPC.
Contribution: This author helped design the study, conduct the study, review analysis, write and approve final manuscript.
Name: Carolina Bibbo, MD.
Contribution: This author conducted the study and approved final manuscript.
Name: Chuan-Chin Huang, ScD.
Contribution: This author helped design the study, analyze the results, and approve the final manuscript.
Name: Kelly G. Elterman, MD.
Contribution: This author helped design the study, conduct the study, and approve the final manuscript.
Name: Eric C. Cappiello, MD.
Contribution: This author conducted the study and approved the final manuscript.
Name: Julian N. Robinson, MD.
Contribution: This author conducted the study and approved the final manuscript.
Name: Lawrence C. Tsen, MD.
Contribution: This author conceived the study, designed the study, conducted the study, reviewed analysis, wrote and approved final manuscript.
This manuscript was handled by: Jill M. Mhyre, MD.
1. Cappiello E, O’Rourke N, Segal S, Tsen LC. A randomized trial of dural puncture epidural technique compared with the standard epidural technique for labor analgesia. Anesth Analg. 2008;107:1646–1651.
2. Norris MC, Fogel ST, Conway-Long C. Combined spinal-epidural versus epidural labor analgesia. Anesthesiology 2001;95:913–920.
3. Collis RE, Davies DW, Aveling W. Randomised comparison of combined spinal-epidural and standard epidural analgesia in labour. Lancet 1995;345:1413–1416.
4. Suzuki N, Koganemaru M, Onizuka S, Takasaki M. Dural puncture with a 26-gauge spinal needle affects spread of epidural anesthesia. Anesth Analg. 1996;82:1040–1042.
5. Thomas JA, Pan PH, Harris LC, Owen MD, D’Angelo R. Dural puncture with a 27-gauge Whitacre needle as part of a combined spinal-epidural technique does not improve labor epidural catheter function. Anesthesiology. 2005;103:1046–1051.
6. Van de Velde M, Teunkens A, Hanssens M, Vandermeersch E, Verhaeghe J. Intrathecal sufentanil and fetal heart rate abnormalities: a double-blind, double placebo-controlled trial comparing two forms of combined spinal epidural analgesia with epidural analgesia in labor. Anesth Analg. 2004;98:1153–1159.
7. Kirshblum SC, Burns SP, Biering-Sorensen F, et al. International standards for neurological classification of spinal cord injury (revised 2011). J Spinal Cord Med. 2011;34:535–546.
8. Stewart AM, Macones GA, Odibo AO, Colvin R, Cahill AG. Changes in fetal heart tracing characteristics after magnesium exposure. Am J Perinatol. 2014;31:869–874.
9. Macones GA, Hankins GD, Spong CY, Hauth J, Moore T. The 2008 National Institute of Child Health and Human Development workshop report on electronic fetal monitoring: update on definitions, interpretation, and research guidelines. Obstet Gynecol. 2008;112:661–666.
10. Bleich AT, Villano KS, Lo JY, et al. Oral misoprostol for labor augmentation: a randomized controlled trial. Obstet Gynecol. 2011;118:1255–1260.
11. Beilin Y, Zahn J, Bernstein HH, Zucker-Pinchoff B, Zenzen WJ, Andres LA. Treatment of incomplete analgesia after placement of an epidural catheter and administration of local anesthetic for women in labor. Anesthesiology. 1998;88:1502–1506.
12. Leach A, Smith GB. Subarachnoid spread of epidural local anaesthetic following dural puncture. Anaesthesia. 1988;43:671–674.
13. Bernards CM, Kopacz DJ, Michel MZ. Effect of needle puncture on morphine and lidocaine flux through the spinal meninges of the monkey in vitro
. Implications for combined spinal-epidural anesthesia. Anesthesiology. 1994;80:853–858.
14. Le Coq G, Ducot B, Benhamou D. Risk factors of inadequate pain relief during epidural analgesia for labour and delivery. Can J Anaesth. 1998;45:719–723.
15. Eappen S, Blinn A, Segal S. Incidence of epidural catheter replacement in parturients: a retrospective chart review. Int J Obstet Anesth. 1998;7:220–225.
16. Norris MC. Are combined spinal-epidural catheters reliable? Int J Obstet Anesth. 2000;9:3–6.
17. Vernis L, Dualé C, Storme B, Mission JP, Rol B, Schoeffler P. Perispinal analgesia for labour followed by patient-controlled infusion with bupivacaine and sufentanil: combined spinal-epidural vs. epidural analgesia alone. Eur J Anaesthesiol. 2004;21:186–192.
18. Heesen M, Van de Velde M, Klöhr S, Lehberger J, Rossaint R, Straube S. Meta-analysis of the success of block following combined spinal-epidural vs. epidural analgesia during labour. Anaesthesia. 2014;69:64–71.
19. Ducrow M. The occurrence of unblocked segments during continuous lumbar epidural analgesia for pain relief in labour. Br J Anaesth. 1971;43:1172–1174.
20. Collier CB. Why obstetric epidurals fail: a study of epidurograms. Int J Obstet Anesth. 1996;5:19–31.
21. Bauer ME, Kountanis JA, Tsen LC, Greenfield ML, Mhyre JM. Risk factors for failed conversion of labor epidural analgesia to cesarean delivery anesthesia: a systematic review and meta-analysis of observational trials. Int J Obstet Anesth. 2012;21:294–309.
22. Goodman SR, Smiley RM, Negron MA, Freedman PA, Landau R. A randomized trial of breakthrough pain during combined spinal-epidural versus epidural labor analgesia in parous women. Anesth Analg. 2009;108:246–251.
23. Tsen LC, Thue B, Datta S, Segal S. Is combined spinal-epidural analgesia associated with more rapid cervical dilation in nulliparous patients when compared with conventional epidural analgesia? Anesthesiology. 1999;91:920–925.
24. Wong CA, Scavone BM, Peaceman AM, et al. The risk of cesarean delivery with neuraxial analgesia given early versus late in labor. N Engl J Med. 2005;352:655–665.
25. Mardirosoff C, Dumont L, Boulvain M, Tramèr MR. Fetal bradycardia due to intrathecal opioids for labour analgesia: a systematic review. BJOG. 2002;109:274–281.
26. Abrão KC, Francisco RP, Miyadahira S, Cicarelli DD, Zugaib M. Elevation of uterine basal tone and fetal heart rate abnormalities after labor analgesia: a randomized controlled trial. Obstet Gynecol. 2009;113:41–47.
27. Whitty R, Goldszmidt E, Parkes RK, Carvalho JC. Determination of the ED95 for intrathecal plain bupivacaine combined with fentanyl in active labor. Int J Obstet Anesth. 2007;16:341–345.
28. Palmer CM, Cork RC, Hays R, Van Maren G, Alves D. The dose-response relation of intrathecal fentanyl for labor analgesia. Anesthesiology 1998;88:355–361.
29. Christiaens F, Verborgh C, Dierick A, Camu F. Effects of diluent volume of a single dose of epidural bupivacaine in parturients during the first stage of labor. Reg Anesth Pain Med. 1998;23:134–141.
30. Arendt K, Segal S. Why epidurals do not always work. Rev Obstet Gynecol. 2008;1:49–55.
31. Rothman KJ. No adjustments are needed for multiple comparisons. Epidemiology 1990;1:43–46.
32. van de Velde M, Teunkens A, Hanssens M, van Assche FA, Vandermeersch E. Post dural puncture headache following combined spinal epidural or epidural anaesthesia in obstetric patients. Anaesth Intensive Care. 2001;29:595–599.
33. Pan PH, Bogard TD, Owen MD. Incidence and characteristics of failures in obstetric neuraxial analgesia and anesthesia: a retrospective analysis of 19,259 deliveries. Int J Obstet Anesth. 2004;13:227–233.
© 2017 International Anesthesia Research Society
34. Groden J, Gonzalez-Fiol A, Aaronson J, Sachs A, Smiley R. Catheter failure rates and time course with epidural versus combined spinal-epidural analgesia in labor. Int J Obstet Anesth. 2016;26:4–7.