Combined spinal-epidural anesthesia (CSEA) is popular for elective cesarean delivery because the intrathecal component provides rapid surgical anesthesia and epidural catheterization enables prolongation of analgesia, treatment of patchy block, and control of postoperative pain. When inserting needles during the performance of this technique, it is our experience that patients variously may experience discomfort (pain, pressure, pushing, or paresthesia) during dermal puncture, subcutaneous or deeper injection of local analgesic solution, inadvertent needle impingement on periosteum, insertion through the meninges of the spinal needle when using the “needle through needle” technique (after localization of the epidural space using the epidural needle), and during insertion of the epidural catheter. Accordingly, it is our practice, as is common elsewhere, to warn patients of all injections, needle insertions, and epidural catheter placement. Although local infiltration of lidocaine or other local anesthetic is routinely used to minimize discomfort during insertion of epidural and spinal needles, a search of the literature reveals that the incidence of spontaneous vocalization, involuntary movement, and patient acknowledgment of any physical sensation during dural puncture by the spinal needle has not been reported. Accordingly, we investigated the incidence of any spontaneous vocalization, involuntary movement, and patient acknowledgment of sensation accompanying thecal penetration by a pencil-point spinal needle and the efficacy of lidocaine injected into the epidural space through the epidural needle before insertion of the spinal needle in reducing any such responses and sensations.
With institutional approval and written patient consent, 43 consecutive ASA physical status I–II patients presenting for elective cesarean delivery under CSEA performed by the authors were block randomized (sequentially by toss of a coin in groups of four to ensure equal numbers per group) on their arrival in the preoperative anesthetic induction room to receive either 3 mL lidocaine 2% plus epinephrine 1:200,000 (group lidocaine) or 3 mL preservative free normal saline (group saline) injected into the epidural space via a Tuohy epidural needle before needle through needle insertion of a spinal needle.
With patients in the sitting position, after application of noninvasive monitoring devices (electrocardiogram, arterial blood pressure, Sao2), insertion of an IV cannula, and IV administration of ondansetron for prophylaxis of emesis, each patient was advised to anticipate skin cleansing, application of sterile drapes, and subcutaneous and interspinous injection of local anesthetic solution (3 to 5 mL 2% lidocaine with epinephrine 1:200,000) at the L2-3 to L4-5 spinous interspace. A 17-gauge Tuohy epidural needle was then sited using loss of resistance to air, after which either 3 mL normal saline or lidocaine was immediately injected. One minute thereafter, without advising the patient that sensation might occur, a 27-gauge Whitacre pencil-point spinal needle was inserted into the epidural needle and gently advanced over 0.5 to 1 s until it was felt to pierce the dura mater and pia mater.
During insertion of the spinal needle, each patient was closely observed by the attending anesthesiologist and nurse (who were blinded and unblinded, respectively, as to the nature of the injectate) for any involuntary movement or spontaneous vocal response in reaction to dural puncture. Immediately after thecal puncture, irrespective of whether involuntary movement or spontaneous vocalization had just occurred, each patient was asked, “Did you feel that?” by the anesthesiologist. When there was a free flow of cerebrospinal fluid from the spinal needle, 1.2 to 2.0 mL 0.75% bupivacaine plus fentanyl 25 μg was slowly injected intrathecally, with intermittent barbotage, after which the spinal needle was removed and an epidural catheter was inserted to a depth of 5 cm in a cephalad direction. An aspiration test was then performed on the epidural catheter, after which a test dose (3 mL) of bupivacaine 0.5% was injected into the epidural space, followed 3 min thereafter by injection of a definitive dose (7 to 12 mL) of the same solution. Any complaint of sustained and severe paresthesia or aggravation of paresthesia by intrathecal injection was regarded as an indication to remove the spinal needle and repeat the above procedure at an alternative lumbar interspace. An epidural infusion of bupivacaine and fentanyl was continued for 24 to 48 h after surgery under the care of the anesthesia department’s acute pain team, and each patient was visited after cessation of epidural infusion by one of the investigators, who inquired into residual sensory or motor impairment. Patient demographics (age, weight, parity, ASA physical status) were recorded on a customized data collection sheet, as were any of the described responses occurring during dural puncture and any neurological sequelae after cessation of the epidural infusion.
Statistical analysis was performed used the χ2 and Student’s t-tests, with P < 0.05 regarded as significant. A sample size of approximately 18 patients per group was calculated on the basis that a 50% reduction in incidence of any subjective or objective patient response (involuntary movement, spontaneous vocalization, and solicited acknowledgment of having felt any physical sensation) occurring during thecal penetration would indicate clinical efficacy of lidocaine, with confidence limits of 95% and power of 80%.
The study population (n = 43) comprised two groups of patients (lidocaine, n = 22; saline, n = 21) which were similar for number, age, weight, parity, and ASA physical status.
Subjective and/or objective responses to dural puncture occurred in 2 (9%) patients given lidocaine and 17 (81%) patients given saline (P < 0.0005). In those given lidocaine, only 2 patients answered the question “Did you feel that?” in the affirmative; neither pateint simultaneously involuntarily moved or spontaneously vocalized during insertion of the spinal needle. With saline, 16 (76%) patients answered the question “Did you feel that?” in the affirmative; of these patients, 7 (33%) involuntary moved and 13(64%) spontaneously vocalized at that time. One patient given saline moved in response to dural puncture but did not report having perceived any physical sensation during thecal puncture.
In no instance was the spinal needle withdrawn, redirected or resited because of paresthesia occurring either during insertion of the spinal needle or during subsequent intrathecal injection of local anesthetic. Further, after cessation of epidural therapy, no patient manifested or complained of any residual neurological deficit.
Early research of the nerve supply to the spinal column suggested that spinal sinovertebral nerves innervated only bone and vascular structures and did not penetrate the dura mater (1). Subsequent reports supported these findings (2,3) but also demonstrated that the anterior aspect of the dura mater also received innervation from the sinovertebral nerves (4,5). This apparent absence of a definitive nerve supply to the posterior aspect of the dura mater probably served as the rationale during the last millennium for the teaching that dural puncture is painless (6). However, cytoimmunochemistry studies in rats (7,8) demonstrate that sensory nerves from upper lumbar dorsal root ganglia are prominent in the lumbar peridural membrane and innervate the lower lumbar dura mater directly. Coincidental with these findings has been the introduction and popularization of CSEA for cesarean delivery, labor analgesia, and major subcostal surgery.
The CSEA needle through needle technique is popular because it provides the technical advantage of accurate insertion of a small gauge spinal needle via the epidural needle and the clinical advantages of a rapid onset of surgical anesthesia (produced by intrathecal injection), the facility to supplement inadequate spinal block, and a route to provide postoperative analgesia (by epidural injection or infusion). An ancillary benefit arising from introduction of the CSEA needle through needle technique is that it provides a ready means of directly observing the sensory responses of conscious human subjects to direct stimulation of the posterior spinal meninges.
A limitation of our study is that we did not inquire into the exact nature of the physical sensation or sensations that caused our patients to spontaneously move, vocalize or, on direct questioning, acknowledge having felt at the moment of thecal puncture. Notwithstanding this, our findings of responses to or an acknowledgment of having experienced a physical sensation at the moment of dural puncture by approximately 80% of parturients presenting for elective cesarean delivery suggest that there are pain or stretch receptors in the posterior aspect of the dura mater, that the spinal needle has impinged on a spinal nerve in the cauda equina or filum terminale, or that the spinal needle has stimulated pain or stretch receptors in the dural sheath surrounding one or other lumbar spinal nerves.
The sensation caused by dural puncture appears to be transient and mild. In our study, the absence of sustained paresthesia during intrathecal injection of local anesthetic and absence of residual nerve damage after discontinuation of postoperative epidural infusion suggests that the cause of our patient responses to thecal penetration is unlikely to be irritation of, or damage to, a spinal nerve by the spinal needle, or intraneural injection of lidocaine. Furthermore, although we consider it good practice to reduce the incidence of these responses to dural puncture by administering lidocaine in this manner, doing so has two potential problems. It is possible that injection of lidocaine into the epidural space in the manner described may obtund paresthesia (which is regarded as a symptom of potential nerve damage) caused by subsequent impingement of the spinal needle on a lumbar spinal nerve. Also, should unrecognized dural puncture by the epidural needle occur, this may result in inadvertent intrathecal injection of 3 mL lidocaine 2% containing 1:200,000 epinephrine, which would result in rapid onset of a significant block.
However, we believe that it is unlikely that a 3-mL volume of lidocaine 2% injected into the epidural space 1 minute before insertion of the spinal needle will obtund spinal nerve paresthesia, as the onset of anesthesia after epidural injection of conventional doses of lidocaine takes 5 to 15 minutes to develop when instituting epidural anesthesia per se (9). Our choice of a delay of 1 minute after epidural injection of lidocaine to insertion of the spinal needle was rationalized by data suggesting that the onset of analgesia after subcutaneous injection or mucus membrane application of lidocaine occurs within 1 minute (10,11), and the failure to find published data describing the effect on the dura mater of lidocaine injected in the manner described.
Further, should inadvertent dural puncture have occurred during insertion of the epidural needle, it is also unlikely that the brisk flow of cerebrospinal fluid that usually emits from a 17-gauge Tuohy epidural needle so placed would not be recognized. However, should dural puncture by the epidural needle perchance not be recognized, the risk of total spinal associated with inadvertent intrathecal injection of 60 mg of lidocaine (3 mL lidocaine 2% containing 1:200,000 epinephrine) is negligible, as doses of 65 to 75 mg lidocaine are recommended for spinal blockade in routine obstetric anesthesia (12). The onset of spinal anesthesia after intrathecal injection of lidocaine during establishment of conventional spinal anesthesia usually occurs within 1–3 minutes (13). Recognition by the anesthesiologist of a similar rapid onset of anesthesia after injection of lidocaine to obtund responses to dural puncture should be regarded as an indication that unanticipated intrathecal injection has occurred and alert the anesthesiologist to the hazard of administering any further intrathecal injection.
Finally, we acknowledge that the dura mater covering the brain is especially sensitive to traction and stretch (14). Our finding that 80% of patients respond to or acknowledge having experienced dural puncture may be related to our use of a pencil-point spinal needle, which, by indenting and stretching the meninges before perforation may stimulate stretch receptors in the spinal meninges. This suggests that Quincke type (cutting edge) spinal needles may cause less dural stretch and induce fewer responses during dural puncture, which is the subject of a continuing study that we are currently performing.
Minimizing needle stick discomfort by previous application of topical anesthetic creams or injection of local anesthetics is recommended as a modality of good patient care (10,15) and may enhance patient satisfaction with their anesthetic management. The results of this study suggest that consideration be given to injecting lidocaine into the epidural space before insertion of the spinal needle when performing CSEA for elective cesarean delivery or other major elective surgery.
The authors wish to thank Moira Priestnal-van den Berg and Anne Starr for secretarial assistance.
1. Luschke H. Die nerven des menschlichen wirbelkanal. Tubingen: Laupp and Siebeck, 1850;49.
2. Bridge CJ. Innervation of spinal meninges and epidural structures. Anat Rec 1959;166:533–63.
3. Pedersen HE, Blunck CFJ, Gardner E. The anatomy of lumbosacral posterior rami and meningeal branches of spinal nerves (sino-vertebral nerves). J Bone Joint Surg 1956;38A:377–91.
4. Edgar MA, Nundy S. Innervation of the spinal dura mater. J Neurol Neurosurg Psychiatry 1966;29:530–4.
5. Kimmel DL. Innervation of spinal dura mater and dura mater of the posterior cranial fossa. Neurology (Minneapolis) 1961;11:800–9.
6. Atkinson RS, Rushman GB, Alfred Lee J. A synopsis of anaesthesia, 10th ed. Bristol: Wright, 1987;662–721.
7. Kumar R, Berger RJ, Dunsker SB, Keller JT. Innervation of the spinal dura: Myth or reality? Spine 1996;21:18–26.
8. Konnai Y, Honda T, Sekiguchi Y, et al. Sensory innervation of the lumbar dura mater passing through the sympathetic trunk in rats. Spine 2000;25:776–82.
9. Berde CB, Strichartz GR. Local anesthetics. In: Miller RD, ed. Anesthesia, 5th ed. Volume 1. New York: Churchill Livingstone, 2000;492–522.
10. van den Berg AA, Abeysekera RMMS. Rationalizing venous cannulation; patient factors and lidocaine efficacy. Anaesthesia 1993;48:84.
11. van den Berg AA. ENT and eye anaesthesia: To spray or not to spray—a rationalization. Anaesthesia 1993;48:74.
12. Pedersen H, Santos AC, Finster M. Obstetric anesthesia. In: Barash PG, Cullen BF, Stoelting RK eds. Clinical anesthesia, 3rd ed. Philadelphia: Lippincott, Williams and Wilkins, 1997:645–68.
13. Local anesthetics. In: Collins VJ. Principles of anesthesiology, 2nd ed. Philadelphia: Lea and Febiger, 1979:865–87.
14. Raskin NH, Peroutka SJ. Headache, including migraine and cluster headache. In: Braunwald E, Faua AS, Kasper DL, et al., eds. Harrison’s principals of internal medicine, 15th ed. New York: McGraw Hill, 2001:70–9.
15. Bell DJ, Schutz JR, White WD, Muir HA. Evaluating the pain of epidural placement in patients [abstract]. Anesthesiology 2003;98 (Supplement 1):58 SOAP A117.