Thoracic paravertebral nerve block, nerve stimulator guidance and the endothoracic fascia

We wish to congratulate and thank Dr Naja and colleagues on their continuing work toward refining the art and science of thoracic paravertebral nerve blockade [1]. Their enthusiasm for their work is obvious and the concepts their work generates are innovative and useful. The earliest reference regarding the use of a nerve stimulator to facilitate thoracic paravertebral blockade was, to our knowledge, made by Bonica and Buckley [2]. More recently, the concept has been rejuvenated [1, 3, 4]. The reasons for this renewed interest may vary, but most investigators seem to be interested in improving the predictability, reliability and safety of thoracic paravertebral blocks. Naja et al. [1] have demonstrated that there is ‘a clear association … between the point of injection within the thoracic paravertebral space (TPVS) and the distribution pattern of injected dye.’ Furthermore, they attribute a ‘paravertebral’ (TPVS) pattern of spread to injection in the immediate proximity of the paravertebral nerve, anterior (ventral) to the endothoracic fascia. Finally, they claim that the injection must be ventral to the endothoracic fascia and that this is a matter of ‘definition’. It is not clear what is meant by ‘definition’. Because a needle is advanced further anteriorly in order to elicit a more refined motor endpoint with a nerve stimulator does not prove the needle is anterior to the endothoracic fascia, which is an ill-defined structure. It is not clear exactly what it is, what its boundaries are, or where the spinal nerves are in relationship to it (most depictions of the endothoracic fascia place the spinal nerve dorsal to the structure in the paravertebral space [5], not ventral, as Naja et al. claim). Its structure and its properties (for example as a potential barrier to mass movement or diffusion) have not been defined. It has not been identified except at surgery or autopsy. Endoscopic views of the paravertebral space have failed to identify it [6]. Therefore, although the conclusion that deposition of dye close to a thoracic spinal nerve affects the pattern of distribution is supported by the work of Naja et al., it is premature to conclude that the pattern of spread is explained by differential placement relative to the endothoracic fascia. Naja et al. did not examine the clinical correlates of the pattern of dye distribution (for example the extent and quality of blockade and the efficacy of analgesia). An examination of their data shows that although the pattern of distribution of dye does correlate with the minimal stimulating current obtained, 43% of the patients (6/14) in the 0.5 mA group had a segmental spread more consistent with that seen in the 2.5 mA group (2.3 segment spread). This suggests that the extent of analgesia may not have differed, except in those patients who had a pure TPVS distribution pattern (4.5 segment spread). Subsequently, no predictions can be made relative to the extent and quality of analgesia that might be expected. The degree of spread within a potential space may be influenced by a number of factors other than the presence of a suspected barrier or the weight of the patient. Distribution may be influenced by the compliance of the targeted space, the pressure head during injection (the rate of injection), the characteristics of the injectate (for example the viscosity), compartmental barriers to uniform spread, as well as the injection volume and the size of the subject [7]. Naja and colleagues have suggested elsewhere [8] that the use of a nerve stimulator for performing thoracic paravertebral blocks ‘is associated with a higher success rate and fewer complications than standard thoracic paravertebral techniques.’ We would like to believe that is true but we remain unconvinced. An insufficient number of patients have been studied. The use of the technique has been largely restricted to a few enthusiastic experts [1, 3, 4]. Fine-tuning a motor response could conceivably lead to more complications (for example pneumothorax, arterial injury or injection, nerve injury, epidural or intrathecal injections, spinal cord injury). Until more convincing evidence is produced to support the concept of enhanced efficacy, spread and safety with discriminatory motor endpoints, it may be prudent to use the nerve stimulator in a supra-maximal mode to identify proximity to the correct fascial plane rather than attempt further needle refinements to minimise the stimulating current [3]. Ultrasound technology has helped refine our knowledge of peripheral regional anaesthesia and may do the same for paravertebral anaesthesia [9]. A hybrid technique that utilises as many of our senses as possible may ultimately prove the most useful. We thank Drs Lang and Saito for an in-depth discussion of our manuscript. We agree that it is unclear if the difference in the spreading pattern can be attributed to the endothoracic fascia structure. As they point out, this fascia is an imprecise anatomical structure, and it can, of course, be questioned if the injection points in our study were in fact made on either side of it. It is clear, however, that the spreading pattern is dependent on whether the injection point is close to the nerve structures or not. Injections close to the nerve will result in a more favourable distribution of the injectate and are, in our opinion, associated with a better and more predictable clinical result [1–3]. Drs Lang and Saito also question whether it is safer to use a nerve stimulator-guided technique with the aim of advancing the needle into a position close to the nerve. It is argued that the manipulation of the needle to reach this position must result in an increased risk of complications, mainly unintentional pleural puncture or pneumothorax, and that the number of patients studied is insufficient to claim that the technique is as safe as more traditional approaches. However, on comparing two studies on the failure rate and complications of PVB, the incidence of unintentional pleural puncture or pneumothorax was almost identical between the study using a traditional ‘loss-of-resistance’ technique (n = 367) [4] and a later study [2] where the large majority of the blocks were performed by placing the needle tip close to the nerves with the aid of nerve stimulation (n = 662) (1.1% and 0.5% vs. 0.8% and 0.5%, respectively). Instead, the factor that appeared to increase the risk of these complications was the performance of bilateral blocks [5], which is most probably caused by a distorted anatomy of the side pointing downward if the patient is in the lateral position when the block is performed. This is illustrated by the fact that the approach to the dependent side has to be slightly modified, as described in one of our recent papers [2]. Lastly, we agree with them that the recent use of ultrasound to guide nerve blocks is an interesting development that appears to improve the success rates and quality of many peripheral nerve blocks [6]. However, in our experience it has proved difficult to identify the nerve structures within the paravertebral space, even with sophisticated ultrasound machines and probes. These problems are related both to the complex ultrasonographic anatomy of the paravertebral space and because it is located deep to the skin plane, thus necessitating the use of lower ultrasound frequencies that reduce the resolution of the ultrasonographic picture. This will limit its use in positioning the needle in the optimal position close to the nerve, but can be used to determine entry into the paravertebral space. Future work will need to clarify whether ultrasound will be of clinical benefit when performing PVBs. M. Z. Naja and Makassed General Hospital Beirut, Lebanon E-mail: [email protected]P. A. LönnqvistAstrid Lindgrens Children's Hospital Stockholm, Sweden E-mail: [email protected]

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.