Neuraxial analgesic techniques are the most effective form of labor analgesia.1 There are limited studies of the efficacy and difficulty of placement of neuraxial methods in parturients with previous spinal instrumentation for scoloisis correction. Beginning in the late 1980s and 1990s, small retrospective studies in the obstetric population (9–21 patients) demonstrated that successful neuraxial labor analgesia was achieved in only 50% of parturients with previous spine surgery.2–4 The only prospective study, conducted from 1997 to 2000, demonstrated 6 of 9 parturients with a history of spine instrumentation had successful epidural labor analgesia.5 Finally, between 44% and 92% of the parturients in these studies required multiple attempts (not well defined) to successfully insert the epidural catheter.2–5
Surgical techniques and hardware for scoliosis correction have evolved in an attempt to improve postoperative quality of life for these patients. The lower lumbar segments and epidural space are spared, and better derotation of the spine is achieved by the use of lateral implants placed outside the spinal canal. The epidural space may be more accessible via the midline approach, and scar tissue may be reduced within the epidural space compared with previously used surgical methods.6–8 Nonetheless, adhesions and scarring in the epidural space and surrounding nerves may affect local anesthetic spread and efficacy, whereas limited back mobility, increased degenerative disc and bone changes surrounding areas of hardware insertion, and underlying uncorrected, progressive, or secondary scoliosis or junctional deformities may contribute to difficulty in neuraxial placement.9–11 Data are sparse regarding the efficacy, difficulty of insertion, and complications of neuraxial labor analgesia in women with spine instrumentation for scoliosis repair.
The purpose of this prospective case-matched study was to compare efficacy of neuraxial labor analgesia in parturients with spinal instrumentation for scoliosis repair to a control group of women without instrumentation. Hourly bupivacaine consumption was used as a surrogate marker of efficacy, and time to epidural catheter placement was used as a surrogate of difficulty of placement. We hypothesized that women with spinal instrumentation would require more bupivacaine per hour to achieve adequate analgesia and longer times to placement of the epidural catheter compared with women with no previous back surgery. Secondary outcomes measured included rates of technique and analgesic failure and complications.
This study was approved by the IRB of Northwestern University (STU00018532). All participants provided informed written consent. The methods used in this study were identical to those used in a previous study of labor analgesia in women with previous discectomy.12 Women with a history of spinal instrumentation for scoliosis repair who requested neuraxial labor analgesia were recruited during preanesthetic consultation or after admission to the labor and delivery unit at Prentice Women’s Hospital of Northwestern University. Postoperative surgical radiographs and reports were reviewed if available before initiation of neuraxial analgesia; otherwise, surgical information was obtained by patient interview and physical examination alone.
When the anesthesia trainee who performed the neuraxial technique in the patient with previous back surgery was available, the next healthy parturient requesting neuraxial analgesia who had consented to participate in the study served as a control subject. Parturients with a history of uncorrected scoliosis, spondylolisthesis, chronic back pain outside of pregnancy, or other chronic pain syndromes (defined as pain of at least 6 months duration outside of pregnancy, requiring chronic pain medication or other pain interventions) were excluded from the control group. The neuraxial procedures in the spinal instrumentation for scoliosis and control subject were initially attempted by the same resident, fellow, or attending anesthesiologist to control for the skill level of the provider. All procedures were directly supervised by an attending anesthesiologist.
A baseline visual analog scale (VAS) score for pain (100-mm unmarked line with the end points labeled “no pain” and “worst pain imaginable”) was obtained at the time of request for analgesia. A combined spinal-epidural or epidural technique was chosen at the discretion of the attending anesthesiologist. The procedure was initiated with patients in the sitting position. In control subjects, the procedure was preferentially performed at the estimated L3 to L4 or L4 to L5 interspace. In subjects with spinal instrumentation for scoliosis repair, the interspace was selected at the discretion of the attending anesthesiologist based on the postsurgical x-rays, surgical history, or skin scarring of the parturient.
The epidural space was located with a 17G Tuohy epidural needle using a loss-of-resistance to air or saline technique. Epidural analgesia was initiated as follows: the epidural catheter (Arrow Flextip Plus, Reading, PA) was threaded 4 to 5 cm in the epidural space, and a 3-mL test dose of lidocaine 15 mg/mL with epinephrine 5 μg/mL was administered through the epidural catheter. An initial loading dose of fentanyl 100 μg and bupivacaine 1.25 mg/mL in increments of 5 mL up to 10 mL was injected until the patient was comfortable. The combined spinal-epidural analgesia procedure mimicked the epidural procedure, except that the dura was punctured with a 27G Sprotte needle (Pajunk Sprotte, Norcross, GA) before inserting the epidural catheter. The intrathecal dose consisted of fentanyl 15 μg combined with bupivacaine 2.5 or 1.25 mg, or fentanyl 25 μg without bupivacaine, at the anesthesiologist’s discretion based on the labor progress in the parturient. Typically, intrathecal fentanyl without bupivacaine was used early in labor. An epidural test dose was administered as described. The VAS score was assessed 30 minutes after the test dose in all subjects.
Data regarding the neuraxial technique were collected by a research nurse and included the number of interspaces attempted (defined as a new puncture at a different lumbar interspace) and the time to epidural catheter placement (defined as interval from local anesthetic injection skin wheal injection to test dose administration). Complications of the procedure (e.g., unintentional dural puncture with the epidural needle) were recorded by the research nurse. Neuraxial analgesia failure was defined as the inability to initiate satisfactory neuraxial analgesia or any epidural catheter requiring replacement for inadequate analgesia at any time during labor or operative delivery after appropriate administration of bupivacaine through the epidural catheter.
The management of maintenance analgesia and breakthrough pain was standardized. Analgesia was maintained with patient-controlled epidural anesthesia (PCEA) with bupivacaine 0.625 mg/mL and fentanyl 1.95 μg/mL (background infusion 15 mL/h, patient-controlled bolus dose 5 mL, lockout period 10 minutes, and hourly maximum 30 mL/h). At the request for additional analgesia, the VAS score and sensory level of the patient were assessed. If a manual epidural bolus (top-up) was considered appropriate, the first epidural manual rebolus dose consisted of bupivacaine 1.25 mg/mL administered in 5-mL aliquots to a total of 10 to 15 mL; the concentration of the PCEA epidural solution was then increased to bupivacaine 1.1 mg/mL with fentanyl 1.7 μg/mL. Twenty minutes after the manual bolus, the VAS score and sensory level were reassessed to ensure adequate pain relief. If there was a second request for analgesia, the same manual bolus was administered, and the background epidural infusion rate was increased to 20 mL/h. VAS scores and sensory level of the subjects were reassessed 20 minutes later, and if the rebolus failed to provide pain relief, the epidural catheter was deemed a neuraxial analgesia failure. Typically, the attending anesthesiologist would consider replacement of the epidural catheter or initiation of some other form of analgesia, depending on the degree of difficulty during the initial neuraxial procedure and the labor progression in the patient.
Analgesic data collected included total bupivacaine consumption and number of episodes of breakthrough pain requiring a manual rebolus. Bupivacaine consumption was normalized by dividing the total bupivacaine dose by the time interval to delivery (interval from initiation of PCEA labor analgesia until 1 hour after vaginal delivery or decision for cesarean delivery). The initial intrathecal dose and epidural loading dose of bupivacaine were not included in the total bupivacaine consumption. Patients were visited on postpartum day 1, and complaints of worsening or new-onset neurologic symptoms were documented and followed up per routine.
The sample size estimated for this study was based on the bupivacaine consumption per hour of labor analgesia. In a previous study using a similar labor analgesic technique, mean (±SD) bupivacaine consumption per hour of labor was 12.3 ± 2.5 mg.13 Assuming that a difference of 2.2 mg per unit hour would represent a significant and clinically important increase in bupivacaine consumption, a sample size of 41 subjects with spinal instrumentation and 41 control subjects would be required to detect this difference with a power of 0.95 at P ≤ 0.05 using a 2-sample 2-sided Wilcoxon rank sum test. These results are based on 5000 Monte Carlo samples from the null distributions. The sample size estimate was based on grouped, and not paired, samples because pairs of subjects were matched for anesthesia provider initiating neuraxial analgesia and not on characteristics associated with analgesic use. This sample would also achieve 0.91 power to detect a difference of 1.5 minutes in time to place the epidural catheter, assuming an average time of 5 minutes and an SD of 2 minutes.12 Sample size calculations were performed using PASS 5 (Release date April 23, 2007; NCSS LLC, Kaysville, UT).
Hourly bupivacaine consumption was compared between parturients with a history of spinal instrumentation for scoliosis and those with no history of back surgery using the Wilcoxon rank sum test. The median shift and 95% confidence interval (CI) of the shift were calculated using the Hodges-Lehmann estimator. Subject characteristics and labor and analgesic outcomes were compared between groups using the Wilcoxon rank sum test or the Fisher exact test. The proportion of subjects in each group with neuraxial analgesia failure was compared between groups using a 2-sample test for equality of proportions with continuity correction. The mean time required to place the epidural catheter was compared between groups by using the generalized pivotal CI method with 100,000 simulations following lognormal transformation.14 Unbiased median and interquartile range for the placement time in the spinal instrumentation and control subjects and for the difference between spinal and control times were determined by bootstrapping (replications = 10,000). Other characteristics of epidural placement were evaluated using McNemar test for binominal data or the sign test for ordinal data. All reported P values are 2 sided, and CIs are 95%. Data were analyzed using R version 3.1.2 (release date October, 31, 2014; R Foundation for Statistical Computing, Vienna, Austria).
The study was conducted between July 2007 and April 2014. Ninety-one women were approached for participation in this study (Fig. 1). Data from 41 women with spinal instrumentation for scoliosis repair and 41 women with no history of previous back surgery who requested epidural labor analgesia were included for analysis. One subject had an anterior correction, and the other 40 subjects had posterior corrections. The surgical details of the scoliosis repair group are presented in Table 1. Eleven parturients (27%) presented in labor without previous preanesthetic consultation. Seventy-three percent of women had surgical reports and/or radiographs available for review by the anesthesia team.
The obstetric characteristics of study participants are shown in Table 2. The groups were similar in age, gestational age, parity, and body mass index. Labor and infant outcomes (VAS score at request for analgesia, mode of delivery, time to delivery, and fetal weight) were not different between groups.
Median (interquartile range) hourly bupivacaine consumption was 15.2 mg/h (12.5–18.7) in the spinal instrumentation group and 14.2 mg/h (11.8–16.0) in the control group; the difference in medians was 1 mg/h (95% CI, −1.3 to 3.0; P = 0.38) (Fig. 2). Other analgesic outcomes are presented in Table 3. The VAS score at 30 minutes after initiation of analgesia was significantly greater in the spinal instrumentation group. The total bupivacaine consumption, number of manual reboluses, and the number of subjects requiring an increase in bupivacaine concentration to achieve effective analgesia did not differ between groups.
The evaluation of neuraxial procedure in the paired spinal instrumentation and control subjects is shown in Table 4. More parturients in the spinal instrumentation group (39%) received a traditional epidural procedure than in the control group (5%). The mean time required to complete the neuraxial technique was 6.5 ± 2.0 minutes in the spinal instrumentation group and 4.6 ± 1.8 in the matched control subjects. The mean difference in time to place the catheter was 41% (95% CI, 7%–108%; P = 0.01) longer in the spinal instrumentation group. More than 1 interspace was attempted in 24% of those with spinal instrumentation compared with 5% of the control group. Although there was no difference in the spinal level of the initial attempt, the final level of placement was more likely to be caudad in the spinal instrumentation group than the control group. A more experienced provider was needed to successfully complete the neuraxial procedure in 7 (17.1%) of the women with spinal instrumentation compared with none of the women in the control group.
Neuraxial analgesia failure occurred in 5 (12%) of women in the spinal instrumentation group but in none of the control patients (difference [95% CI], 12% [−0.3% to 25%]; P = 0.06). The details of the failed neuraxial analgesia and complications are shown in Table 5.
The primary finding from this study is that women with spinal instrumentation for scoliosis repair require similar amounts of epidural bupivacaine per hour of labor as women who have had no previous back surgery. No differences in number of manual doses or need to switch to a greater local anesthetic maintenance infusion were observed. We did observe greater pain scores at 30 minutes in the spinal instrumentation group, but this difference could be explained by the greater number of epidural compared with combined spinal-epidural procedures in the control group.15,16 Successful analgesia was achieved in 88% of women with surgical correction for scoliosis on the first attempt. This rate is much greater than reported in previous studies in this population.2–5 Most surgical repairs in our patients (71%) were conducted subsequent to the surgeries examined in the most recent publication evaluating neuraxial labor analgesia in these parturients.5
Despite a high rate of successful analgesia, greater difficulty was encountered in epidural catheter insertion, as reflected by longer procedure time, more needle redirections, greater number of attempted interspaces, and greater likelihood of switching to a more experienced provider. Our initial failure rate of 12% in the spinal instrumentation group was greater than in the matched controls but fell within the reported failure rates for labor analgesia in women without back surgeries cited in the literature. For example, neuraxial analgesia failure rates of 15.4%17 and 13.1%18 were reported in 2 retrospective studies from academic institutions similar to our own.
The final spinal level of the neuraxial technique was lower in the spinal instrumentation group, likely in an attempt to avoid scar tissue and hardware. Although attempts were made to avoid the surgical area, uncorrected, progressive, or secondary scoliosis, junctional deformities, loss of lordosis in the lower lumbar region, and increased degenerative disc and bone changes surrounding areas of hardware may have contributed to difficulty in neuraxial placement.9,10
The specific neuraxial technique chosen for individual patients was at the discretion of the attending anesthesiologist, but our typical institutional practice is to initiate neuraxial labor analgesia with a combined spinal-epidural technique. Despite this institutional preference, more patients in the spinal instrumentation group had analgesia initiated using a traditional epidural technique. This preference may be due to the ability to immediately assess epidural catheter function without the delay imposed by the initial spinal dose. Studies are mixed regarding whether dural puncture in the setting of epidural labor analgesia reduces the incidence of breakthrough pain or need for epidural catheter replacement.19,20 Studies do suggest that epidural catheters fail at a lower rate when inserted as part of a combined spinal-epidural technique compared with a traditional epidural technique.21,22 Indeed, we observed a greater incidence of epidural catheter failure in the spinal instrumentation group, but the epidural catheter replacements occurred in patients who received both combined spinal-epidural and traditional epidural analgesia.
Most of the epidural catheters in this study were used for neuraxial labor analgesia (88%) and not for surgical anesthesia. Thus, we cannot estimate the failure rate for conversion from epidural labor analgesia to surgical anesthesia. The conversion failure rate may be greater in women with spine instrumentation, given the requirement of more extensive dermatomal spread and denser anesthesia for cesarean delivery compared with analgesia, but this possibility remains unproven.
Our study is only valid if interpreted in the context of its limitations. The study was underpowered to find a small difference in hourly bupivacaine consumption in the spinal instrumentation group. Our study lacked many details regarding precise surgical techniques, instrumentation types, and spinal levels of surgery, which were often obtained by patient history and not from operative reports. Thus, we cannot draw conclusions regarding whether differences in surgical procedures might affect neuraxial analgesia outcomes.
We did not use ultrasonography to assist with identification of anatomical landmarks, as this technique was not common at our institution as the time of study initiation. Ultrasonography of the spine before initiation of neuraxial procedures has been shown to reduce attempts and the need for epidural catheter replacement in parturients without back surgery.23,24 Case reports and case series suggest that identification of the epidural space and catheter insertion may be facilitated by ultrasonography in patients with a history of severe scoliosis or scoliosis repair.25–28 Future studies are needed to determine whether this patient population may benefit from the use of ultrasonography to identify interspinous spaces and assist with proper angulation of the Tuohy needle and identification of the epidural space.
In summary, the findings of this investigation suggest that previous surgery for scoliosis repair does not affect the efficacy but may reduce the overall success of neuraxial labor analgesia. Previous scoliosis repair surgery will likely increase the time and difficulty in initiating neuraxial labor analgesia. We suggest the anesthesiologist attempt neuraxial placement in the lower lumbar segments, as radiographs and surgical reports may not always be readily available in laboring parturients, though they may assist in better understanding the anatomical variations. Neuraxial labor analgesia should be offered to parturients with previous surgery for scoliosis repair although informed consent should include a discussion of the possibility of technical difficulties and surgical anesthesia failure.
Dr. Cynthia Wong is the Section Editor for Obstetric Anesthesiology for Anesthesia & Analgesia. This manuscript was handled by Dr. Steven L. Shafer, Editor-in-Chief, and Dr. Wong was not involved in any way with the editorial process or decision.
Name: Jeanette R. Bauchat, MD.
Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.
Attestation: Jeanette R. Bauchat has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.
Name: Robert J. McCarthy, PharmD.
Contribution: This author helped design the study, analyze the data, and write the manuscript.
Attestation: Robert J. McCarthy has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
Name: Tyler R. Koski, MD.
Contribution: This author helped write the manuscript
Attestation: Tyler R. Koski approved the final manuscript
Name: Cynthia A. Wong, MD.
Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.
Attestation: Cynthia A. Wong has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.
1. Anim-Somuah M, Smyth RM, Jones L. Epidural versus non-epidural or no analgesia in labour. Cochrane Database Syst Rev. 2011:CD000331
2. Daley MD, Rolbin SH, Hew EM, Morningstar BA, Stewart JA. Epidural anesthesia for obstetrics after spinal surgery. Reg Anesth. 1990;15:280–4
3. Crosby ET, Halpern SH. Obstetric epidural anaesthesia in patients with Harrington instrumentation. Can J Anesth. 1989;36:693–6
4. Hubbert CH. Epidural anesthesia in patients with spinal fusion. Anesth Analg. 1985;64:843
5. Villevieille T, Mercier FJ, Benhamou D. Is obstetric epidural anaesthesia technically possible after spinal surgery and does it work? [in French]. Ann Fr Anesth Reanim. 2003;22:91–5
6. Desai SK, Brayton A, Chua VB, Luerssen TG, Jea A. The lasting legacy of Paul Randall Harrington to pediatric spine surgery: historical vignette. J Neurosurg Spine. 2013;18:170–7
7. Hasler CC. A brief overview of 100 years of history of surgical treatment for adolescent idiopathic scoliosis. J Child Orthop. 2013;7:57–62
8. McCormick J, Aebi M, Toby D, Arlet V. Pedicle screw instrumentation and spinal deformities: have we gone too far? Eur Spine J. 2013;22(Suppl 2):S216–24
9. Aebi M. The adult scoliosis. Eur Spine J. 2005;14:925–48
10. Arlet V, Aebi M. Junctional spinal disorders in operated adult spinal deformities: present understanding and future perspectives. Eur Spine J. 2013;22(Suppl 2):S276–95
11. Benzon HT, Braunschweig R, Molloy RE. Delayed onset of epidural anesthesia in patients with back pain. Anesth Analg. 1981;60:874–7
12. Bauchat JR, McCarthy RJ, Koski TR, Cambic CR, Lee AI, Wong CA. Prior lumbar discectomy surgery does not alter the efficacy of neuraxial labor analgesia. Anesth Analg. 2012;115:348–53
13. 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;102:904–9
14. Ledolter J, Dexter F, Epstein RH. Analysis of variance of communication latencies in anesthesia: comparing means of multiple log-normal distributions. Anesth Analg. 2011;113:888–96
15. Simmons SW, Taghizadeh N, Dennis AT, Hughes D, Cyna AM. Combined spinal-epidural versus epidural analgesia in labour. Cochrane Database Syst Rev. 2012;10:CD003401
16. Niesen AD, Jacob AK. Combined spinal-epidural versus epidural analgesia for labor and delivery. Clin Perinatol. 2013;40:373–84
17. 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;13:227–33
18. Eappen S, Blinn A, Segal S. Incidence of epidural catheter replacement in parturients: a retrospective chart review. Int J Obstet Anesth. 1998;7:220–5
19. 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;107:1646–51
20. 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;103:1046–51
21. 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;108:246–51
22. Gómez P, Echevarría M, Calderón J, Caba F, Martínez A, Rodríguez R. The efficacy and safety of continuous epidural analgesia versus intradural-epidural analgesia during labor [in Spanish]. Rev Esp Anestesiol Reanim. 2001;48:217–22
23. Grau T, Leipold RW, Conradi R, Martin E, Motsch J. Efficacy of ultrasound imaging in obstetric epidural anesthesia. J Clin Anesth. 2002;14:169–75
24. Vallejo MC, Phelps AL, Singh S, Orebaugh SL, Sah N. Ultrasound decreases the failed labor epidural rate in resident trainees. Int J Obstet Anesth. 2010;19:373–8
25. Bowens C, Dobie KH, Devin CJ, Corey JM. An approach to neuraxial anaesthesia for the severely scoliotic spine. Br J Anaesth. 2013;111:807–11
26. Yeo ST, French R. Combined spinal-epidural in the obstetric patient with Harrington rods assisted by ultrasonography. Br J Anaesth. 1999;83:670–2
27. McLeod A, Roche A, Fennelly M. Case series: ultrasonography may assist epidural insertion in scoliosis patients. Can J Anesth. 2005;52:717–20
© 2015 International Anesthesia Research Society
28. Costello JF, Balki M. Cesarean delivery under ultrasound-guided spinal anesthesia [corrected] in a parturient with poliomyelitis and Harrington instrumentation. Can J Anesth. 2008;55:606–11