Paravertebral block

David Scott

Hugo Sellheim of Leipzig first described thoracic paravertebral block in 1905 for analgesia after abdominal surgery. It was a blind technique and fell out of use in subsequent years. Eason and Wyatt re-introduced paravertebral block in 1979, with some success but this was at a time when regional anaesthesia was not frequently practiced, so its use remained low. During the past two decades, with the resurgence of regional anaesthesia and the use of ultrasound, Karmakar, Lonnqvist, and Richardson have all contributed to the increasing popularity of paravertebral block.


The anatomy of the paravertebral ‘space’ is dependant on the level of the block. In the thoracic region it is defined by the transverse process of the vertebrae, and the costotransverse ligament posterior (superficially), and the endothoracic fascia anterior (deep), the dura mater medially, and lateral end of the transverse process laterally. It is a potential space through which the spinal and peripheral nerves pass through. At the thoracic level a small volume of local anaesthetic will cause blockade of the dermatome at that level, larger volumes will spread caudad and cephalad to block more levels. It is this characteristic which allows catheters to be placed. If too much catheter is placed it may migrate along a single nerve and result in a limited spread, so it is best to avoid attempting to feed a catheter too far.
At the lumbar level the paravertebral anatomy is different. The lung is no longer present, and is largely replaced by the psoas muscle. The lumbar plexus is formed by the anterior divisions of L2-4 and after leaving the intervertebral foramen it is enveloped by the psoas muscle, runs through the muscle eventually emerging on the anterior surface of the muscle at the level of the groin as the femoral nerve, and after the obturator nerve and lateral cutaneous nerve of the thigh have parted company.


There are several approaches described for paravertebral block. At the thoracic level there is an anatomy-based approach, a modification of this with ultrasound guidance, and three real time ultrasound guided approaches both in-plane and out of plane. None of these approached is perfect, and each has its problems. The nerves are not visible on ultrasound so all the approaches make use of surrogate anatomy for performance of the block. The landmark approach uses a loss-of-resistance technique where there is a change after passing through the costotransverse ligament.

At the lumbar level ultrasound is useful in identifying the transverse process, and the psoas muscle. It also allows the anaesthetist to identify the kidney and the anterior border of the psoas muscle – which is useful in preventing collateral damage. The lumbar plexus is generally not visible to ultrasound as it is surrounded by muscle so the nerve needs to be identified with the aid of a nerve stimulator – looking for a twitch in the femoral nerve distribution.
The value of this technique, and the effectiveness of the block is well reported. Thoracic paravertebral block has been investigated for effectiveness:
• 7 Randomised controlled trials reported on the rate of effective block for PVB, defined as complete or partial failure, ranged from 87% to 100%
• Any form of regional block will have more failures than GA, the PVB failure rate was not > 13%.
• Paravertebral block provides early improved pain relief (24 hours) compared with GA. After this time pain scores are similar (single shot)
• 4 Randomised controlled trials reported on length of hospital stay, with 3 studies reporting significantly shorter stays after Paravertebral block compared with GA (P= 0.05)
• Studies reported significantly higher levels of patient satisfaction with PVB block, compared with field block (P= 0.05), and GA (P = 0.008)


Every procedure we perform carries risk and complications. The recognised complications of Paravertebral block include: risk associated with all blocks (infection, hematoma, neural damage failure); specific complications for the procedure: pneumothorax, epidural (or subarachnoid) spread, hypotension, and mortality (local anaesthetic toxicity).
It is difficult to determine the rates of complications; I suggest that neurological damage is very rare, epidural spread hypotension and pneumothorax uncommon, and death very rare (only case I am aware of is local anaesthetic toxicity in a lumbar block that was not managed successfully).
Contraindications are general and specific. General contraindications include allergy, refusal, infection (at site of injection or severe systemic), and specifically coagulopathy. The recent guidelines for neuraxial blockade and coagulopathy provide a good guide for choosing or avoiding lumbar plexus block. Psoas compartment hematoma or abscess can result from lumbar plexus block.



These blocks are useful for unilateral trunk surgery and for hip surgery. They provide excellent analgesia and anaesthesia and are relatively simple to perform.
In conclusion paravertebral block is reliable and safe, it is easily performed and suitable as a continuous technique. It provides superior analgesia following mastectomy and axillary dissection and this may lead to reduced long term pain and reduced recurrence of cancer. The real question is why wouldn’t you do it?



Klein SM, Bergh A, Steele SM, Georgiade GS, Greengrass RA.  Thoracic Paravertebral block for breast surgery  Anesth Analg 2000;90:1402-5

Karmarkar MK.  Ultrasound-guided thoracic paravertebral block.   Techniques in Regional Anaesthesisa and Pain Management 2009;13:142-149
Renes SH, Bruhn J, Gielen MJ, Scheffer GJ, van Geffen GJ.  In-Plane Ultrasound-Guided Thoracic Paravertebral Block   Reg Anaesth  Pain Med. 2010;35:212-216

Fagenholz PJ, Bowler GMR, Carnochan FM, Walker WS Systemic local anaesthetic toxicity from continuous thoracic paravertebral block BJA 2012;109:260-262    - a cautionary tale of high dosing in small patient 


Joanne Martin

Mob: 021614655