Portal de Periódicos da CAPES

Paravertebral by proxy – time to redefine the paravertebral block

2018; Wiley; Volume: 73; Issue: 10 Linguagem: Inglês

10.1111/anae.14348

1365-2044

I. Costache, Amit Pawa, Faraj W. Abdallah,

Shoulder Injury and Treatment

There has been a recent focus in the regional anaesthesia literature on proximal approaches to block intercostal nerves. Several newly described techniques have led us to question what truly defines a paravertebral block (PVB) 1-4. We propose the concept that these novel blocks are not entirely new, and that they exert their effect by spread to the paravertebral space, thus should be collectively considered as 'paravertebral by proxy' blocks. We consider these proximal blocks near to the thoracic midline as distinct from the lateral chest wall blocks such as the serratus plane block, which have also increased in popularity of late. The lateral chest wall blocks target solely the lateral cutaneous branch of the ventral ramus of the spinal nerve, as opposed to the proximal approaches which block both the ventral and dorsal rami of the spinal nerve near the paravertebral space. The thoracic paravertebral space has been traditionally defined as a triangular-shaped area bounded medially by the vertebral bodies, intervertebral discs and intervertebral foraminae; anteriorly by the parietal pleura; and posteriorly by the superior costotransverse ligaments, ribs and transverse processes. This space is continuous with the epidural and intercostal spaces. The belief that the PVB requires the needle tip to be in the traditionally defined paravertebral space was unchallenged before the ultrasound era because it was not possible in routine clinical practice to verify the precise needle-tip position. Recently, several techniques with injections outside the thoracic paravertebral space have been described. Each of the retrolaminar 1, intercostal/paraspinal 2, erector spinae plane 3 and midpoint transverse process to pleura (MTP) 4 blocks can achieve blockade of the thoracic nerve roots in the paravertebral space without actually entering this space directly with the block needle (Fig. 1). These PVB variants have recently been collectively described as 'paraspinal blocks' 5. From an anatomical standpoint, if the definition of a PVB reflects blockade of the nerve roots in the paravertebral space, then we suggest that these indirect ways of achieving a PVB be thought of as achieving a paravertebral spread by proxy, without placing the needle tip in the paravertebral space, in close proximity to the pleura. This concept is not entirely new, for studies of abdominal wall blocks, such as the transversus abdominis plane and quadratus lumborum blocks, have also demonstrated spread to the paravertebral space without direct needling. Our novel MTP block 4 clearly exemplifies this concept, whereby we have repeatedly observed that injections posterior to the superior costotransverse ligament produce a successful PVB; and, importantly, we have frequently noted anterior pleural displacement on injection with the needle tip clearly evident posterior to the superior costotransverse ligament. To that end, it is also very likely that we have been performing these 'new' blocks and achieving a paravertebral spread by proxy unknowingly for decades, through the use of the traditional PVB landmark technique. Based on ultrasound images of the paravertebral space and simple geometry, it is clear that the traditional landmark approach of contacting the transverse process and then redirecting and advancing the needle tip 1 cm further (either in a cephalad or caudad direction) does not guarantee needle-tip placement anterior to the superior costotransverse ligament. This would suggest that previous studies using a landmark technique may have involved a portion of blocks being achieved without directly accessing the paravertebral space, that is, by proxy. This hypothesis takes into consideration the geometrical fact that the final needle-tip position with the landmark technique is highly dependent on the initial contact point with the transverse process (Fig. 2). Evidence from dermatomal spread patterns seems to support this theory. Multilevel injections seem to have been a prerequisite to successfully block all targeted dermatomes when using the landmark technique. In contrast, multilevel injections may not be needed when dynamic ultrasound-guided PVB is performed. The ultrasound-guided technique reliably achieves spread to the paravertebral space, even with small volumes (1 ml) injected in the intercostal space, according to a cadaver study 6. Radiographic, clinical and cadaveric studies indicate that a single-level dynamic ultrasound-guided PVB consistently produces multidermatomal block and extensive cranio-caudal spread 7-10. For example, studies by Renes et al. 11 and Marhofer et al. 12 have presented evidence of multidermatomal spread, six and four dermatomes, respectively, with dynamic ultrasound-guided PVB using a single injection with 20 ml of local anaesthetic. Furthermore, both single- and multilevel ultrasound-guided PVB have been shown to achieve similar dermatomal spread, according to a recent study 8. Ultrasound-guided injection in the paravertebral space can achieve a wide spread, even to other anatomically continuous areas, including the epidural space, the contralateral paravertebral space and multilevel paravertebral space 9. It is, therefore, becoming clear with the recent ultrasound-guided paravertebral block literature that multiple injections are in fact not required in order to achieve multilevel thoracic blockade. Correspondingly, evidence from high-fidelity simulation also corroborates this proposition. Observations from a simulation study suggest that failed thoracic PVB may be due to injection either too distant from the space, or a needle in the lung 13. However, as some superficial needle positions may eventually result in successful blocks via a 'paravertebral by proxy' 4, we are left with the concerning possibility that a considerable proportion of failed blocks involved needling through the pleura, as seen with the 12% incidence of pleural puncture in the high-fidelity simulation model 13. Nonetheless, it is challenging to prove or dismiss this suspicion, as the proportion of pleural punctures resulting in clinical pneumothorax is unknown, and studies of needle-tip position of failed PVB are lacking. Finally, although successful block of somatic intercostal nerves has been demonstrated with all of the 'paravertebral by proxy' blocks, it remains unclear if all of these PVB variants reliably block the sympathetic chain. Blockade of the sympathetic chain seems to play a role in enhancing acute pain control, and also in preventing the development of persistent postoperative pain. A number of studies have demonstrated that an isolated sympathetic block that spares somatic innervation is capable of improving acute postoperative pain following upper extremity 14, 15 and breast surgery 16. Similarly, the sympathetic block associated with thoracic PVB is believed to play a role in preventing persistent postoperative pain following mastectomy 17. The exact underlying mechanism is not fully understood, but a modulatory role of the sympathetic nervous system in pain sensation has been proposed 18. Regardless of the mechanism, truncal blocks that produce a combination of somatic and sympathetic blockade are thought to be superior to fascial plane block alternatives that primarily block somatic innervation. Clinically, these novel blocks offer alternatives to the traditional PVB and are helpful to clinicians new to PVB, uncomfortable with PVB or in case of technical difficulties such as poor image quality or obese patients 19. As our understanding of relevant anatomy evolves, we are more convinced that it is time to abandon the long-held belief that the paravertebral space needs to be directly accessed in order to achieve spread of local anaesthetic to the thoracic nerve roots. It does seem timely to review the definition of a PVB, and to consider all injections outside the paravertebral space as 'paravertebral by proxy' blocks. That said, and as obvious from the numerous 'unknowns' in Table 1, further studies are needed to identify the technique-specific optimal volumes for these paravertebral by proxy blocks, to compare their clinical efficacy and to investigate whether they produce a sympathetic block. In the meantime, in clinical practice, we would recommend direct paravertebral block as the optimal technique sited with ultrasound guidance. In circumstances where the operator is uncomfortable siting a block in such proximity to pleura, or where the image obtained is suboptimal, the operator should consider siting an MTP or ESP block to achieve paravertebral by proxy. AP has received honoraria from GE Healthcare for teaching, and consults for B Braun Medical Ltd. No external funding or other competing interests declared.