Lumbar neuraxial ultrasound


By Drs Adrian Chin & Ben Crooke (with Michael Fredrickson)


Many anesthesiologists justifiably question the need for ultrasound guidance when performing neuraxial blockade (ultrasound guided spinal or epidural block), particularly when the landmark/palpation technique is efficient, reliable and associated with a high success rate. Compared to peripheral blocks, neuraxial blocks have the unique advantage of having readily identifiable needle endpoints: CSF flow for spinal block; distinct loss-of-resistance for epidural block. However, ultrasound imaging can assist successful and safe neuraxial needle placement for several reasons.

First, studies have shown the conventional palpation technique frequently results in a more caudad estimation of a specific lumbar spinal interspace: e.g. the L2/3 interspace is often estimated to be the L3/4 interspace. 1, 2 That finding combined with the conus medullaris often extending more caudad than the L1 level may explain previous reports of neuraxial blockade related spinal cord trauma resulting in persistent neurological deficit. 2

Rising obesity rates have added further relevance for neuraxial sonography. A prominent iliac crest fat pad increases imprecision when estimating the position of the lumbar interspaces, while higher depths from the skin to the vertebral column make estimation of both the position of the midline and the required needle insertion angle more challenging.

Evidence suggests neuraxial ultrasound can more accurately locate the midline, the desired spinal interspace, and the required needle insertion angle and depth. This manifests as reduced needle attempts and improved neuraxial block success rates, particularly for trainees. 3-5 Pre-procedural sonography has been particularly advocated for spinal anesthesia in patients with obesity, kyphoscoliosis and those having had previous spinal surgery including Harrington rods. 6-9

Skill level

Pre-procedure midline marking: basic

Pre-procedure needle insertion point marking and estimation of the required needle angle: intermediate

Real-time ultrasound guided spinal or epidural insertion: advanced/not advised

Procedure time: 5 mins

Common indications

Pre-procedural ultrasound can be used whenever neuraxial block is considered, but is more commonly used in those patients in whom the technique is anticipated to be technically challenging.

Sedation: Not required


Vertebral gross anatomy

Relevant vertebral anatomy for neuraxial sonography (Fig. 1A+B):

Transverse scanning (probe transverse to the long axis of the spine) – when at the level of the spinous process, the lamina acan be visualised. With the probe between adjacent spinous processes, the articular processes/facet joints and transverse processes are seen (Fig 1B).

Paramedian sagittal scanning (probe parallel to the long axis of the spine) – moving laterally, the lamina, articular processes/facet joints and then transverse processes are seen (Fig 1A).

The posterior complex represents the ligamentum flavum, epidural space and posterior dura. The anterior complex represents the anterior dura, posterior longitudinal ligament and the posterior vertebral body.


Fig. 1A+B. Lumbar vertebrae

Vertebral sonography

As stated above, when placed transverse to the long axis of the spine, there are two distinct views depending on the neuraxial structures imaged (Fig. 2). In the “(transverse) spinous process” view, the spinous process casts a large midline shadow between the hyperechoic vertebral lamina. In the “(transverse) interlaminar or interspinous” view, the articular processes/facet joints and transverse processes are visible. With the latter view, an acoustic window exists between adjacent spinous processes, enabling visualization of the posterior and anterior complexes. While the “target” for spinal anesthesia is the posterior complex, visualization of the anterior complex denotes a clear sonographic window (and therefore needle route) through the interlaminar space. 4

When placed to the side of the long axis of the spine (paramedian sagittal view), there are four distinct views depending on the neuraxial structures imaged (Fig. 2). In the “paramedian sagittal transverse process” view, the trident sign can be seen: a finger like shadowing behind the transverse processes. In the “transverse sagittal articular process” view, camel humps can be seen: continuous articulations of adjacent articular processes. In the “paramedian sagittal laminar” view, horse heads or the sawtooth sign can be seen: non-continuous bone due to the interlaminar spaces. With this view the posterior and anterior complexes can be seen, particularly if the probe is angled medially towards the spine (“paramedian sagittal oblique” view). 


Fig. 2. Neuraxial ultrasound: Six sonographic views of the neuraxial spine.

Surface landmarks

As previously mentioned, conventional surface landmarks can be unreliable.

Needle: Not applicable



1. Curved array low frequency probe (2-5 MHz)

2. Ultrasound gel

3. Marking pen

4. Gauze (to remove gel)

5. Alco wipe® (to clean pen tip if coated with gel)

Procedure (Video 1)

1. After positioning the patient either sitting or lateral, place a low frequency (2-5MHz) curved array probe transverse to the long axis of the spine at the middle of the patient’s lower back.

2. Optimise the image depth, frequency and time-gain-compensation.

3. Mark the midline. With a mark on the middle of the probe, the probe is aligned transversely so there is symmetry between the left and right bony neuraxial structures (each side forms a mirror image of the other side). Depending on the cephalocaudad probe position, either the lamina (“spinous process or lamina view”) or the transverse processes and articular processes/facet joints (“interspinous process or interlaminar view”) will be visible. Sliding the transversely applied probe in a cephalad direction, a marking pen is used at intervals to mark the skin adjacent to the middle of the long edge of the probe. To avoid having to mark through gel, start caudad and mark above the probe while moving in a cephalad direction. The accuracy of this technique assumes there are no major spinal deformities (scoliosis, metalware).

4. Identify the lumbosacral junction. The probe is orientated to obtain a paramedian sagittal laminar view (Fig. 2). After identifying a short hyperechoic line at approx. the L4/5 level (L5 lamina), the probe is  moved caudally until a continuous hyperechoic line is identified (the sacrum). A gap should be evident between these two lines with the more deeply located anterior complex visible deep to this gap.

5. Mark the L1-5 vertebral levels. Maintaining a paramedian sagittal orientation, the probe can then be moved cephalad, with a marking pen again at the midpoint of the long edge of the probe to mark each lamina.

6. Estimate the required needle direction. The probe is rotated transversely at the desired level (e.g. L3-4) and a transverse interlaminar view is obtained. The estimated required needle direction is obtained with slight cephalad-caudal tilting and sliding until the best image of the posterior and anterior complexes is obtained. The resulting probe tilt is a good estimate of the required needle direction.

7. Estimate depth to the dura. Using the ultrasound calipers.

8. Mark optimal needle insertion point by marking the 4 midpoints of the long and short edges of the probe. The probe is put down and a horizontal and vertical line is constructed. Where they intersect is the optimal needle insertion point. The vertical line should correspond with the previously marked midline.

9. Check the optimal insertion point by re-applying probe and ensuring a good view of the anterior complex.

Additional sonographic views of the spine can be obtained using a paramedian sagittal orientation and sliding laterally through the paramedian laminar, articular process and transverse process views. The paramedian oblique view is obtained by tilting the probe medially, aiming to highlight the posterior and anterior complexes through the interlaminar space. This view can be used for real-time ultrasound guided spinal anesthesia, although this is an advanced skill.

Local Anesthetic regimen: Not applicable

Specific complications

The only complications specific to ultrasound pre-scanning for neuraxial block are the theoretical risks related to transferring gel into the epidural and subarachnoid spaces. The clinical relevance of this potential problem is at present unknown. All ultrasound gel should be removed from the back before commencing the needle procedure.

Pitfalls include: 

1. Imprecise skin markings.

2. Moving the patient after marking the back but before needle placement.

3. Misidentification of the lumbosacral junction; failing to recognise anomalies of the junction (present in 12% of the population), 11

4. Confusing the anterior for the posterior complex. This may result in inadvertent dural puncture during epidural blockade due to overestimation of the depth to the epidural space.


Clinical PEARLS

1. It is difficult to use a marking pen through ultrasound gel. Start low and move up, marking above the probe. An Alco wipe can be used to clean the gel off the pen.

2. Appropriate positioning will “open up” the intervertebral space.

3. The anterior complex is normally brighter than the posterior complex.

4. The measured depth to the neuraxial structures may be equal to or shorter than the actual depth due to compression of subcutaneous tissue by the probe.

5. Be precise with the pen, and reapply the probe over the needle insertion point to confirm the correct position.


Video 1. Pre-procedure scanning for lumbar neuraxial blockade. 


1. Broadbent, C.R., et al., Ability of anaesthetists to identify a marked lumbar interspace. Anaesthesia, 2000. 55(11): p. 1122-6.

2. Reynolds, F., Damage to the conus medullaris following spinal anaesthesia. Anaesthesia, 2001. 56(3): p. 238-47.

3. Perlas, A., Evidence for the use of ultrasound in neuraxial blocks. Reg Anesth Pain Med, 2010. 35(2 Suppl): p. S43-6.

4. Chin, K.J., M.K. Karmakar, and P. Peng, Ultrasonography of the adult thoracic and lumbar spine for central neuraxial blockade. Anesthesiology, 2011. 114(6): p. 1459-85.

5. Chin, K.J. and A. Perlas, Ultrasonography of the lumbar spine for neuraxial and lumbar plexus blocks. Curr Opin Anaesthesiol, 2011. 24(5): p. 567-72.

6. Costello, J.F. and M. Balki, Cesarean delivery under ultrasound-guided spinal anesthesia [corrected] in a parturient with poliomyelitis and Harrington instrumentation. Can J Anaesth, 2008. 55(9): p. 606-11.

7. Prasad, G.A., P.S. Tumber, and C.M. Lupu, Ultrasound guided spinal anesthesia. Can J Anaesth, 2008. 55(10): p. 716-7.

8. Chin, K.J., et al., The use of ultrasound to facilitate spinal anesthesia in a patient with previous lumbar laminectomy and fusion: a case report. J Clin Ultrasound, 2009. 37(8): p. 482-5.

9. O'Donnell, D., A. Prasad, and A. Perlas, Ultrasound-assisted spinal anesthesia in obese patients. Can J Anaesth, 2009. 56(12): p. 982-3.

10. Carvalho, J.C., Ultrasound-facilitated epidurals and spinals in obstetrics. Anesthesiol Clin, 2008. 26(1): p. 145-58, vii-viii.

11. Bron, J.L., B.J. van Royen, and P.I. Wuisman, The clinical significance of lumbosacral transitional anomalies. Acta Orthop Belg, 2007. 73(6): p. 687-95.