Needle-through-needle labor analgesia: rapid intrathecal onset with the extendable flexibility of an epidural catheter
The combined spinal-epidural (CSE) technique marries the rapid, dense onset of a subarachnoid block with the titratable duration of an epidural catheter. First described by Coates and Mumtaz in 1982 and popularized for labor analgesia through the 1990s, it is now a mainstay "walking epidural" option in modern obstetric anesthesia.
The sitting position is preferred for CSE in laboring patients because it maximizes the vertical distance between spinous processes and helps identify the midline in patients with a large gravid abdomen. The patient is asked to round the lower back ("mad-cat" or "angry cat" posture) — flexing the lumbar spine widens the interlaminar and interspinous gaps by several millimeters, easing needle passage between adjacent laminae.
Lateral decubitus positioning is an alternative, useful when a support person cannot help maintain sitting balance, when the patient has significant aortocaval compression concerns, or after neuraxial opioid administration causes sedation. Lateral position slightly reduces the accuracy of midline identification but reduces vasovagal risk during a contraction.
Regardless of position, the anesthesiologist palpates the iliac crests bilaterally; a line connecting them (Tuffier's or the intercristal line) crosses the spine at approximately the L4 vertebral body or the L3-L4 interspace, providing a reliable external landmark before sterile prep and draping.
L3-L4 is the most commonly used interspace for CSE in labor, with L2-L3 as an acceptable alternative. The adult spinal cord (conus medullaris) typically terminates at L1-L2, so needle insertion at or below L3-L4 minimizes any theoretical risk of direct spinal cord trauma; insertion above L2 is avoided for elective techniques for this reason.
Ultrasound-assisted landmark identification is increasingly used, particularly in obese parturients, where manual palpation of the intercristal line has been shown to be inaccurate in up to 30% of cases (frequently 1-2 levels higher than the true anatomic space). A brief pre-procedural ultrasound scan improves first-pass success and reduces the number of needle passes.
Anatomic imprecision matters: manual landmark palpation misidentifies the intended lumbar interspace in roughly one-quarter to one-third of patients, and body mass index is the strongest predictor of error — reinforcing why careful technique and, where available, ultrasound guidance improve first-pass success.
CSE is chosen specifically when both rapid onset and flexible, extendable duration are clinically valuable — the classic scenario being active labor with an anticipated long course, where the mother wants fast relief now but the anesthesia team wants the option to extend the block for many hours or convert it to surgical anesthesia if a cesarean becomes necessary.
Compared with epidural-only technique, CSE analgesia begins in 2-5 minutes rather than 15-20 minutes, and typically uses a lower total local anesthetic dose because the intrathecal component is so much more potent milligram-for-milligram than the epidural route.
Compared with single-shot spinal anesthesia, CSE preserves the ability to redose indefinitely via the catheter, so a spinal block with a fixed 60-120 minute duration is not a limiting factor for labors that outlast it, and the same catheter can later provide a dense surgical block for cesarean delivery without repeating the dural puncture.
| Product | Indication | Trial Design | Key Result |
|---|---|---|---|
| Combined Spinal-Epidural (CSE) | |||
| Epidural (single technique) | |||
| Single-shot Spinal | |||
| Continuous Spinal Catheter |
Locating the epidural space relies on a tactile principle rather than direct visualization: the ligamentum flavum offers dense, gritty resistance to a syringe plunger, and that resistance vanishes abruptly the instant the needle tip crosses into the fat- and vessel-filled epidural space.
From posterior to anterior, a midline Tuohy needle traverses: skin, subcutaneous fat, the supraspinous ligament (connecting the tips of adjacent spinous processes), the interspinous ligament (between adjacent spinous processes), and finally the ligamentum flavum — a tough, elastic band that connects adjacent laminae and is the last resistant structure before the epidural space.
The needle is introduced a few millimeters, then the stylet is removed and a low-friction glass or plastic syringe containing 2-3 mL of saline (or air) is attached. The needle-syringe unit is advanced slowly in small increments — often 1-2 mm at a time — while constant or intermittent gentle pressure is applied to the plunger.
While traversing ligament, the dense collagenous tissue prevents the plunger from moving — injected fluid cannot be forced in, and the plunger springs back when released. This is felt as firm, "gritty" resistance. The moment the needle tip pierces the ligamentum flavum and enters the epidural space, the resistance disappears almost instantaneously: the plunger glides forward with a distinct "give," and often a small amount of saline can be injected freely.
Saline is generally preferred over air for LOR in obstetric patients: injected air that inadvertently enters a vein can (rarely) cause venous air embolism, and residual air in the epidural or subdural space has been associated with patchy blocks and, if it tracks intrathecally, an increased risk of post-dural-puncture headache. Saline LOR also avoids interference with subsequent CSF identification, since saline (unlike air bubbles) is not easily confused with cerebrospinal fluid.
The entire epidural space is only 3-5 mm deep in the lumbar midline — roughly the thickness of a stack of two or three coins — which is why LOR technique demands slow, controlled, millimeter-scale advancement rather than continuous forward pressure.
Skin-to-epidural-space depth averages 4-6 cm in a normal-weight adult but correlates strongly with body mass index — in patients with obesity, depths of 7-9 cm or more are common, and a standard Tuohy needle (8-10 cm working length) may occasionally be insufficient, requiring a longer needle.
Other factors that increase technical difficulty include prior lumbar spine surgery (scar tissue obliterating the ligamentum flavum plane), scoliosis (asymmetric trajectory), calcified interspinous ligaments in older or multiparous patients, and inability of the patient to flex adequately during a painful contraction. Multiple needle passes are associated with a higher rate of accidental dural puncture (roughly 1-2% overall, higher with difficulty) and postpartum backache.
The defining maneuver of CSE: once the Tuohy needle tip is confirmed in the epidural space, a long, small-gauge pencil-point spinal needle is passed down through its lumen and advanced a further short distance to reach the subarachnoid space — using the epidural needle purely as an introducer/guide.
With the Tuohy needle bevel/opening oriented cephalad and firmly seated in the epidural space, its stylet-free lumen accepts a long (119-127 mm), fine-gauge pencil-point spinal needle. Because the epidural needle is typically 8-10 cm and the spinal needle is deliberately longer, the spinal needle protrudes 10-15 mm beyond the Tuohy tip once fully inserted — just enough to traverse the dura and arachnoid mater and enter the subarachnoid space, without traveling far enough to risk contacting the cauda equina or anterior epidural structures.
A subtle "pop" or change in tactile feedback is often felt as the spinal needle pierces the dura-arachnoid complex, analogous to (but finer than) the loss-of-resistance sensation of the epidural needle itself. Pencil-point (non-cutting) tips such as Whitacre or Sprotte designs are used specifically because they part dural fibers rather than cutting them, substantially lowering the risk of post-dural-puncture headache compared with older cutting-tip (Quincke) designs.
Once the spinal needle is felt to have crossed the dura, its stylet is withdrawn. Free, spontaneous flow of clear cerebrospinal fluid at the needle hub — usually appearing within a few seconds — confirms subarachnoid placement. If CSF does not appear promptly, gentle rotation of the needle by 90 degrees or a very slight further advance (1-2 mm) may be attempted, since the needle opening can lie tangential to the dural sac.
If no CSF is obtained despite reasonable attempts, this is termed a "failed dural puncture" or negative wet tap and occurs in roughly 5-10% of CSE attempts — commonly because the epidural space was unusually deep relative to spinal needle length, or the epidural needle deviated off the true midline. In this situation the anesthesiologist may proceed with epidural-only technique, use a longer spinal needle, or reposition and reattempt.
A confirmed CSF flashback is the single most important safety checkpoint in CSE: intrathecal drug is only injected once free-flowing, clear CSF is unambiguously seen — never based on "feel" alone.
The subarachnoid space at L3-L4 or L2-L3 contains the cauda equina — a mobile bundle of freely floating lumbosacral nerve roots — rather than the spinal cord itself (which ends at L1-L2 in most adults). This mobility is protective: nerve roots are typically displaced rather than pierced by a fine-gauge needle, which is one reason paresthesia during needle advancement, while it should prompt withdrawal and redirection, only rarely reflects permanent injury.
The distance from ligamentum flavum/epidural space to the dural sac is small and fairly consistent (a few millimeters), which is precisely why a standard 10-15 mm needle protrusion reliably reaches CSF across most adult body habitus once the epidural space itself has been correctly identified — the epidural needle placement, not the spinal needle length, is the primary source of technical variability.
The intrathecal injection is the therapeutic heart of CSE: a tiny volume of local anesthetic combined with a lipophilic opioid, deposited directly into cerebrospinal fluid bathing the nerve roots, produces dense analgesia within minutes at a fraction of the dose an epidural would require.
A typical labor CSE intrathecal injectate combines a very low dose of bupivacaine (1.25-2.5 mg) with a lipophilic opioid — fentanyl 10-25 mcg or sufentanil 2.5-5 mcg — in a total volume of about 1-2 mL. This "minimal local anesthetic dose" strategy is deliberate: it aims to block small, unmyelinated C-fibers carrying visceral labor pain while sparing larger motor fibers, preserving enough lower-extremity strength for many patients to stand or take assisted steps (the classic "walking epidural").
The opioid component acts on mu-opioid receptors in the substantia gelatinosa of the dorsal horn, producing analgesia synergistic with, but mechanistically distinct from, the local anesthetic's sodium-channel blockade — this combination allows a much lower local anesthetic dose than would be needed alone, which is precisely why motor block is often minimal in early labor.
Drug deposited directly into CSF acts on nerve roots that are essentially bathed in it, with no dura, epidural fat, or connective tissue barrier to cross. Diffusion through CSF alone can establish a working block within 2-5 minutes. By contrast, an epidural injection must diffuse across the dura and arachnoid (a much slower, rate-limiting process) before reaching CSF and nerve roots in adequate concentration, which is why epidural-only analgesia characteristically takes 15-20 minutes to establish.
This speed difference is the central clinical justification for choosing CSE when a laboring patient is in significant distress and needs rapid relief — for example, in advanced or rapidly progressing labor where a slower-onset epidural alone might not catch up with the pace of cervical dilation.
Because so little drug is needed intrathecally (roughly one-tenth the local anesthetic mass of a typical epidural bolus), the CSE spinal component delivers faster, denser analgesia while often using less total local anesthetic than an epidural alone would require for the same effect.
A successful low-dose intrathecal injection typically produces a sensory block to approximately the T10 dermatome (umbilicus level) within 3-5 minutes, sufficient to blunt the visceral pain of uterine contractions while sparing higher thoracic segments needed for adequate ventilatory mechanics.
Because the intrathecal component causes rapid sympathetic blockade, maternal blood pressure and fetal heart rate are monitored closely for the first 20-30 minutes — maternal hypotension occurs in roughly 10-15% of patients and is treated promptly with intravenous fluid, left uterine displacement, and phenylephrine or ephedrine as needed, since maternal hypotension can reduce uteroplacental perfusion and produce transient fetal heart rate decelerations.
After the spinal needle is withdrawn, the Tuohy needle — still correctly seated in the epidural space — is used to thread a flexible catheter that will provide analgesia for the remainder of labor. Because the spinal component already provides analgesia, the catheter cannot be functionally verified by patient-reported relief alone, making an explicit test dose essential.
The spinal needle is withdrawn first, leaving the Tuohy needle in place as the conduit. A soft, flexible, multi-orifice epidural catheter is then passed through the Tuohy needle and advanced 4-5 cm beyond the needle tip into the epidural space — a depth chosen to balance adequate purchase (reducing the risk of the catheter slipping out with maternal movement) against excessive depth (which increases the risk of the catheter exiting through an intervertebral foramen or coiling asymmetrically, producing a unilateral block).
The Tuohy needle is then withdrawn over the catheter while the catheter is held steady, and the catheter is secured to the back with an adhesive dressing, typically looped over the shoulder, before the patient repositions.
A standard test dose of 3 mL of lidocaine 1.5% with epinephrine 1:200,000 (equivalent to 45 mg lidocaine and 15 mcg epinephrine) is injected through the catheter with continuous maternal heart rate monitoring.
• Intravascular placement: epinephrine produces a transient maternal heart rate increase of more than 20 beats per minute within 20-40 seconds — a positive test indicates the catheter tip lies within an epidural vein and must be withdrawn and repositioned. • Intrathecal placement: lidocaine 45 mg given intrathecally would rapidly produce a dense motor block (inability to lift the legs) within 3-5 minutes — evidence the catheter has migrated into the subarachnoid space rather than the epidural space. • Negative test: no significant heart rate change and no rapid dense motor block supports (but does not with certainty prove) correct epidural placement.
Epinephrine-containing test doses are interpreted cautiously in laboring patients because uterine contractions themselves can transiently raise maternal heart rate, and beta-blocked or otherwise medicated patients may blunt the expected tachycardic response — clinical judgment and incremental dosing remain essential.
Because the intrathecal spinal component has already produced analgesia, a laboring patient cannot reliably tell the anesthesia team whether the catheter itself is working until its first independent top-up — the defining safety trade-off of CSE compared with a standalone epidural, whose catheter is testable and confirmed before the spinal block would ever mask it.
With a standalone epidural, an inadequate or malpositioned catheter is usually apparent quickly because the patient still has pain, prompting early replacement. With CSE, the spinal component provides 60-120 minutes of effective analgesia regardless of whether the epidural catheter itself is correctly sited — so a catheter that has failed (kinked, migrated out of the epidural space, single-sided, or subcutaneous) may not declare itself until the spinal block wears off, sometimes well into labor or, worst case, at the point an emergency cesarean is needed.
This is the central, well-recognized risk-benefit trade-off of CSE: reported rates of epidural catheter replacement after CSE range from roughly 3-7% in most series, similar to or only slightly higher than epidural-only technique, but the consequence of a "silent" failure is more consequential because it may only be discovered under time pressure. Some services mitigate this by giving a small epidural test/priming dose at initial placement and by re-checking the block early, once the spinal component has partially receded (around 45-60 minutes).
The epidural catheter is what turns a fixed-duration spinal block into an anesthetic that can be sustained for a marathon labor or rapidly deepened into a full surgical block — the single greatest advantage CSE holds over single-shot spinal anesthesia.
As the intrathecal component wears off (typically 60-120 minutes after injection), the epidural catheter takes over as the sole analgesic route. Most units transition to a continuous epidural infusion (for example, dilute bupivacaine 0.0625-0.125% or ropivacaine 0.1-0.2% combined with fentanyl 1.5-2 mcg/mL) delivered by programmed intermittent bolus or patient-controlled epidural analgesia, supplemented by clinician-administered top-up boluses as needed.
This is the core practical benefit of the "E" in CSE: labor duration is unpredictable and can extend many hours beyond a single spinal block's working life, and the indwelling catheter allows analgesia to be extended indefinitely without another needle pass, another dural puncture, or another window of vulnerability while the block is re-established.
If an urgent or emergency cesarean becomes necessary, the same epidural catheter already in place can usually be "topped up" to a dense surgical block far faster than starting fresh neuraxial anesthesia — often the single greatest time-saving advantage of CSE in an intrapartum emergency.
A typical conversion protocol uses 2% lidocaine with epinephrine and fentanyl, or 0.5% bupivacaine, given as incremental 5 mL boluses (total 15-20 mL) over 5-10 minutes, aiming for a bilateral sensory block to cold or pinprick at the T4 dermatome (nipple line) — the level required to blunt visceral surgical stimulation from peritoneal traction and uterine exteriorization. Sodium bicarbonate is sometimes added to 2% lidocaine to speed onset by increasing the non-ionized fraction available to cross nerve membranes.
Because epidural top-up relies on an already-functioning catheter, a catheter that turns out to be inadequate (see Stage 5) may force a rescue single-shot spinal or, rarely, general anesthesia under true time pressure — reinforcing why early recognition of a marginal catheter, well before an emergency arises, is so important.
A well-functioning CSE catheter can typically achieve a T4 surgical block in 10-20 minutes — compared with 20-30+ minutes to place and dose a fresh epidural, or the delay and hemodynamic considerations of induction for general anesthesia — which is why CSE is often favored for patients at elevated risk of intrapartum cesarean.
Rapidly extending a block from a T10 labor level to a T4 surgical level produces a correspondingly larger sympathetic blockade, and maternal hypotension is common (roughly 10-15% of top-up dosing episodes) — left uterine displacement, co-loading with intravenous crystalloid, and a phenylephrine infusion or boluses are standard mitigations, since maintaining maternal blood pressure protects uteroplacental perfusion and fetal wellbeing throughout the conversion.
Motor block also intensifies substantially with cesarean-strength top-up dosing (Bromage scale progressing toward complete lower-extremity paralysis), which is expected and appropriate for surgical immobility but is a further reminder that the "walking epidural" character of the low-dose labor block is deliberately abandoned once denser surgical anesthesia is required.