On slaughtering and burying holly Cows - Intraneural InjectionBy Professor Miguel Angel Reina, MD, PhD For many years, decades and even centuries, certain dogmas have been introduced into the culture of Medicine and especially into Acute Pain Medicine and Regional Anesthesia. Entire industries have been built around these beliefs. These include multimillion dollar RA block needle designs, pressure monitoring devices and many more. My Group and I from Madrid and Barcelona have for many years now focused on slaughtering these holly cows. The research of our group always focused on the truth rather than on conventional wisdom, culture or dogmas. For example, for many years we believed that the fibers of the dura run parallel and up and down and therefore we should place spinal needles in a certain way to avoid cutting these fibers. Our scanning electron microscopy work [Reina, et al. An In Vitro Study of Dural Lesions Produced by 25-Gauge Quincke and Whitacre Needles Evaluated by Scanning Electron Microscopy. Reg Anesth Pain Med 2000; 25: 393-402] clearly showed the truth and the reader is encouraged to read this and the many other myths that we busted by searching Miguel Angel Reina on Google Scholar. But, this is not where it ended. Our group recently slaughtered a few more wholly cows. In a recent talk at the ESRA meeting in Ireland, I presented a continuous education lecture I called “What’s New in Intraneural Injection? Basic Research” (ESRA 2018: Reina MA, Boezaart AP, Sala-Blanch X, Monzó E, server A, Bigeleisen P). In this CME lecture he highlighted the recent work in the last year or two of our group as follows: “It has generally been accepted for many years that intraneural injection during attempted nerve block may lead to nerve injury (1). Researchers, however, do not agree on what constitutes “intraneural” injection. Some describe intraneural injection as an injection deep to the circumneurium (formerly the paraneurium (2) while most anesthesiologists accept that it is inside the epineurium but outside the perineurium of the fascicles. Intrafascicular injection has been proposed by a few to be truly intraneural, and numerous animal studies have shown that it is not only possible to inject into a fascicle, but also that spread of the injectate can be as far central as the spinal cord. Our research group conducted a series of studies attempting to unravel these controversies. We first compared the microstructures of nerves in commonly studied species to those of humans (3) and explored the validity of extrapolating these animal data to humans. High-resolution, light microscopic images were obtained from cross-sections of sciatic nerves at their bifurcation from fresh rat, rabbit, pig, sheep, dog, and human cadavers. Various microanatomical characteristics were measured and compared between the species. We were able to show that there were some interspecies similarities but the vast majority of the microanatomical measurements of all five species differed significantly from those of humans (3). We could convincingly conclude that although some of the metrics could reasonably be expected to differ due to the size of the species, e.g., nerve cross-sectional area, there was little microanatomical similarity between the sciatic nerves of humans and those of the non-primate mammalian species studied. These results may seriously call into question the validity of the conclusions reached over the years by direct translation from these species to humans. It may be more important than previously realized into which compartment a local anesthetic agent is injected for single-injection nerve blockade for determining onset and duration of action of nerve blocks. Furthermore, the compartment into which a catheter for a continuous peripheral nerve block is placed may also be critical for its efficacy. The common markers used (methylene blue and India ink) in anatomical studies of intraneural injections have multiple problems, including extravasation into non-injected locations due to their small and inconsistent molecular sizes. This confounds our understanding by making reproducible research into nerve compartments difficult and may be responsible for some confusion. With the intention of improving the results and being able to do more in-depth microanatomical research, we have used a different marker to study the intraneural spread after injection in in vitro studies. We collected blood from patients and prepared it as heparinized blood solution. In our studies, the injected marker was comprised of erythrocytes. It proved to be a marker that always maintains the same shape and size, is compatible with all histological stains, is easily recognizable, and whose spread is limited by the barriers formed by collagen fibers and/or cellular layer and does not diffuse through these. Our marker was then injected into fresh cadavers in a manner similar to routine clinical practice for ultrasound-guided peripheral nerve blocks to form a so-called “doughnut” by “hydrodissecting.” The nerves, and structures of the cadavers were then prepared and examined by light microscopy to evaluate advantages and disadvantages of the marker. Although deliberate intraneural injection was avoided in the first phase of this study, the marker was identified inside all of the nerve compartments except inside fascicles. Apart from leaking through the needle entry site in some instances, the heparinized blood solution did not show extravasation into neighboring nerve compartments. The tissues were not distorted and the cells did not form clots. Nerve membranes and compartments containing heparinized blood solution could be clearly identified with routine staining. We concluded that this technique enabled us to accurately study the nerve membranes and compartments. It appeared to be superior to other markers because it did not leave the compartments into which it had been injected, it did not distort the tissue, and it was easily visible under light microscopy. It is important to note that the use of ultrasound does not prevent intraneural injection. The ultrasound image did not allow us to identify which part of the injected solution had intraneural spread. In 2006, Bigeleisen (4) published work where he deliberately injected local anesthetic intraneurally. This resulted in excellent nerve blocks without evidence of nerve injury. Multiple authors (5) have subsequently reported that peripheral nerve intraneural injection is not only a frequent occurrence, and has been for many years, but is also probably safe and even desirable. This notion has been challenged recently (6). Our group attempted to further clarify these controversies but not to argue the virtues (or dangers) of peripheral nerve intraneural injection in clinical practice. The introduction of ultrasound and a better understanding of the microanatomy of nerves, fascia, and membranes forming the barriers that surround them has challenged the common perception that peripheral nerve intraneural injection is harmful. For decades, intraneural injection to obtain anesthetic blockade of combined and single peripheral nerves was commonly cautioned against and regarded as dangerous and the cause of permanent nerve damage. However, what is becoming increasingly popularized is that during unintentional or intentional intraneural injection, it may be essential to distinguish between intrafascicular and extrafascicular injection. Extrafascicular intraneural (sub-epineurial but extra-perineurial) injection has been associated with short onset times and prolonged effects without major complication, whereas intrafascicular injection is thought to be associated with axonal injury and neurological complications, although this has yet to be documented in humans. As a second phase of our project, we used 22-G Stimuplex D needles and simulated deliberate ultrasound-guided peripheral nerve intraneural injection in the sciatic and median nerves of fresh cadavers. After the nerves and major neighboring blood vessels were identified with ultrasound, heparinized blood solution was injected in the nerves without any difficulty. These injections closely mimicked anesthetic nerve blocks and the swelling of the nerves was clear. The successive transversal microtome cross-section slices of the nerves in paraffin wax of every succeeding slice was performed. The epimysium, circumneurium, epineurium, and perineurium formed mechanical barriers that limited heparinized blood solution spread outside the particular compartments into which it had been injected. In some samples, the intraneural injection spread outside of the nerve most likely because the needle hole went into the epineurium. Furthermore, small movements of the needle tip when performing the technique could possibly deposit part of the solution outside the nerve. The heparinized blood solution cells were identified within the structure of the nerve, mainly in the extrafascicular compartment intermingled with the interfascicular adipocytes inside the epineurium. The injected heparinized blood solution displaced adipocytes without rupturing them and created pathways between them. In addition, extrafascicular spread within the epineuriums of the nerves did not disrupt the epineurium in any of the images examined. A large number of erythrocytes from the heparinized blood solution were observed between layers that formed the perineuriums surrounding the fascicles. Erythrocytes were not found inside any of the fascicles examined (i.e., inside the endoneurium) except in a single sample of one sciatic nerve and one median nerve examined under low magnification. These erythrocytes displaced the axons without deforming them. The morphology of axons was not affected. The erythrocytes did not surround the axons, and the axons were all bundled together without any disruption or deforming of the endoneurium. Following this observation, a closer and more intense analysis of the minute details of the microanatomy of the nerve was performed using high magnification and high resolution of the images in the two samples. In each sample, we observed septae of differing sizes within the fascicle. Incomplete septae were formed by thin perineurial layers that extended to the center of the fascicles where islands of intrafascicular erythrocytes from heparinized blood solution appeared. The erythrocytes in the heparinized blood solution still did not leave the perineurial layer structure and were found between the collagen fibers located within the space between these perineurial layers. On low magnification, however, these perineurial cells were not visible, thus creating the false appearance of an island of heparinized blood solution cells within the fascicle. Under ultrasound guidance, it was clear that every compartment into which we could inject heparinized blood solution, we would be able to create a so-called “doughnut sign.” This sign is therefore not limited to a specific sought or unsought compartment. “Doughnuts” can form if the injection is sub-epimyseal, sub-circumneural (outside the nerve), and/or subepineural (intraneural but extrafascicular). The ultrasound images allowed us to visualize the gross anatomy, which differed from the histological studies at high magnification and high resolution that allowed us to identify the true final compartment where the erythrocytes were injected. The microanatomy allowed us to obtain data that we could not visualize with ultrasound. Finally, no “perineural spaces,” as Moore described, surrounded the axons, and the endoneurium appeared solid without the possibility of fluids spreading longitudinally in it. The characteristics of the endoneurial tissue formed by the sum of collagen fibers that assumed the form of the outside barriers of individual “tunnels” for each axon is a key issue in the new findings related to intraneural injections and possible intrafascicular spread. For decades, this tissue was considered soft enough for an intraneural injection to enter the fascicles. That possibility was never proved or disproved by previous studies. We have, however, been able to show that the intrafascicular tissue is not soft. On the contrary, it is a very compact tissue that prevents the entry of solutions and provides resistance to intrafascicular spread. This has conclusively been proven by our study. The consistency of intrafascicular tissue, however, is probably not similar to all of the fascicles of all the nerves, and each nerve must be studied separately. Similarly, the consistency of the endoneurium inside the fascicles is most likely different is each animal species. Much research remains to be done, and our team has been working on this line of research for years. It is, however, extremely important to realize that these studies and extrapolations have only been performed on peripheral nerves and no conclusions can be reached for nerve roots because these have not yet been studied. Devastating complications have been reported following intra-root injections. The conclusions of our work are only applicable to the sciatic and median human nerve, and no extrapolation of our conclusions to other nerves should or could be reached before these nerves have been studied. Our studies are ongoing regarding deliberate intraneural injections at the plexuses and nerve root levels and we must still complete the analysis of these samples.
From our and other studies, we conclude that:
- Intraneural injections into peripheral nerves have been occurring for many years and are a very frequent phenomenon - with or without ultrasound guidance.
- Intraneural injection into peripheral nerves may even be desirable, but should never be confused with central nerves - spinal roots, trunks divisions, and cords.
- It is most probably not feasible to extrapolate animal data with human data. We have not studied any primate nerve microanatomy but all the work done in the past on dogs, rabbits, and other animals is most likely not valid. Although we tried, it was very difficult to inject into the endoneurium of fascicles of humans.
- The spread of solutions through the cells of the perineurial layers and perineurial septae requires further research.
- Catheter placement for low-volume, low-concentration local anesthetic continuous nerve blocks outside the circumneurium will most likely lead to secondary block failure – that is, once the primary high-volume, high-concentration block has worn off”.