Anatomical dissections are not obsolete

This Invited Commentary accompanies the following original article: Albokrinov AA, Fesenko UA. Spread of dye after single thoracolumbar paravertebral injection in infants. A cadaveric study. Eur J Anaesthesiol 2014; 31:305–309. Despite recent advances in imaging techniques such as ultrasound, computed tomography and magnetic resonance imaging, the cadaver study from Albokrinov and Fesenko, published in this issue of the European Journal of Anaesthesiology, clearly demonstrates that human dissection studies remain a cornerstone of our understanding of regional anaesthesia. This is especially true when it comes to nerve block techniques in neonates and infants, as extrapolation from adult anatomy can mislead. One example of this is the anatomical relationship between the anterior superior iliac spine and the ilioinguinal and iliohypogastric nerves in neonates.1,2 Moriggl and Eichenberger reported a number of important cadaver studies in adults that gave new and important insights into paravertebral blocks (PVB),3–5 permitting effective blockade with ultrasound guidance (USG). The authors of the present study should be congratulated on their new information regarding PVB in children, obtained from a combination of USG block and subsequent anatomical dissection.6 The cadaver study in itself is no small achievement, as navigating the difficult waters of consent and ethical issues associated with cadaver dissections in small children is far from easy, and is virtually impossible in a number of developed countries nowadays. This study departs from the classic landmark-based approach and the nerve stimulator-guided technique previously used in children, and offers a new approach to thoracolumbar PVB in small children. The technique used is similar to the lateral USG approach suggested by Marhofer7 and Boretsky et al.,8 who both advocate needle-tip placement just lateral to the transverse process, within the most medial part of the intercostal muscle apparatus. Albokrinov and Fesenko describe a technique in which the final position of the needle tip comes to lie between the ventral lateral part of the transverse process and the most dorsal part of the psoas muscle. In adults, this position would be dangerous as it would not be possible to visualise the needle tip, but in neonates and infants, this is still possible due to lack of ossification of the transverse process. The margin for error is, however, very small, and visualising the needle tip throughout its course remains essential. It has previously been shown that low thoracic paravertebral blockade cannot spread caudally beyond T12 because of the origin of the psoas muscle.9 This necessitates separate injections to block the lumbar nerves as they emerge from their intervertebral foramina. The approach described by Albokrinov and Fesenko must, therefore, involve a certain degree of hydrodissection by the injectate, separating the origin of the psoas from the transverse process. This hydrodissection must not only allow spread to the most caudad part of the thoracic paravertebral space, but must also produce a communication to the L1 paravertebral pocket, as the L1 nerve root was found to be stained in all individuals. This can be explained by the prone position of the cadavers in which the blocks were performed, in contrast to the lateral decubitus position that is normally used in clinical practice. We are also intrigued by the observation that nerve roots more caudal to L1 were also stained when the injectate volume was more than 0.3 ml kg−1. As to whether these high volume injections gained access to the L2-5 roots via the epidural route remains unanswered, as the investigators did not have consent to dissect the spinal cord. Dalens et al.10 provide evidence in support of epidural spread when they found that a medial approach to the lumbar plexus had a high incidence of blockade of both limbs. Previous studies have found two separate patterns of spread associated with PVB in both adults and children.11,12 These are a more localised ‘cloud-like’ appearance, or a ‘longitudinal’ pattern in which the injectate is seen to spread mainly within the paravertebral gutter in between multiple adjacent levels. Albokrinov and Fesenko have not classified their patterns of spread, but according to their description, it appears as if the spread in the vast majority of the individuals belong to the ‘cloud-like’ pattern. This is in line with previous findings of a preferential ‘cloud-like’ spread if the injection is performed in the most dorsal part of the paravertebral space.13 As technology advances, we may be able to identify the path of local anaesthetic spread using three-dimensional ultrasound.14,15 Until such time, cadaver studies such as these will continue to have a place in clinical practice. They provide information that is otherwise impossible to attain clinically in this non-verbal age group. In conclusion, Albokrinov and Fesenko describe a new and interesting approach to an ultrasound-guided PVB at the thoracolumbar junction in neonates and small infants. They have also provided some new insights into the anatomical issues related to PVB in this age group. Furthermore, they have been able to clarify the PVB volume needed for analgesia in various types of inguinal surgery, and they have also provided additional support for previous findings that were based mainly on radiograph studies. Despite the many merits of combining ultrasound-guided block followed by careful anatomical dissection in cadavers, it will not be possible to fully judge the adequacy of the findings of the present study until it has been investigated in one or more prospective randomised clinical trials. We certainly look forward to such data becoming available. Acknowledgements related to this article Assistance with the Invited Commentary: none. Financial support and sponsorship: none. Conflicts of interest: none. Comment from the editor: this Invited Commentary was checked by the editors but was not sent for peer review.