https://orcid.org/0000-0002-1066-7015
Figures
Abstract
This study aimed to histologically investigate whether the compatibility of donor and recipient nerves in distal nerve transfer for radial and ulnar nerve palsy is suitable for restoring nerve function. Partial median to radial nerve transfer for radial nerve palsy and partial median to ulnar nerve transfer for ulnar nerve palsy were performed in 10 cadaveric upper limbs fixed using the Thiel technique. Histological analysis of the nerve samples at the coaptation site focused on the number of myelinated axons. Each recipient and donor nerve was identified in all specimens without any anatomical variations. While median-radial nerve transfer techniques showed an adequate number of donor axons, median-ulnar nerve transfer techniques showed a shortage of donor axons. The insufficiency of donor axons compared to the recipient axons may explain the challenges in reinnervating the recipient muscles. Combining the two different nerve transfers may compensate for the shortage of donor axons and improve motor recovery. Type of study and Level of evidence: Therapeutic, Level III.
Citation: Kodama A, Kunisaki A, Tanaka T, Ishibashi S, Tsuji K, Munemori M, et al. (2025) Histological study of donor/recipient feasibility in distal nerve transfer for the upper limb nerve injury. PLoS One 20(5): e0322331. https://doi.org/10.1371/journal.pone.0322331
Editor: Richa Gupta, PGIMER: Post Graduate Institute of Medical Education and Research, INDIA
Received: August 14, 2024; Accepted: March 19, 2025; Published: May 5, 2025
Copyright: © 2025 Kodama et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are with in the paper.
Funding: This work was supported by funds from JSPS KAKENHI (Grant Number JP 22K09357) to Akira Kodama. https://www.jsps.go.jp/The funder did not play any role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. There was no additional external funding received for this study.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Proximal nerve injuries in the upper limbs present a significant challenge to achieving adequate functional recovery, even with nerve repair or transplantation interventions. Historically, tendon transfer procedures such as pronator teres to the extensor carpi radialis brevis (ECRB), flexor carpi radialis (FCR) to the extensor digitorum communis (EDC), and palmaris longus to the extensor pollicis longus (EPL) for radial nerve palsy, and ECRB transfer to the adductor pollicis for ulnar nerve palsy have been the primary approaches to address these injuries [1,2]. Although these techniques have demonstrated reliability and historical success, they often result in non-physiological kinematics and reduced power grip [3]. Notably, for ulnar nerve palsy, the acquired pinch strength remains significantly below normal even after tendon transfer procedures [2,4]. Nerve transfer is a potential solution for reconstructing radial and ulnar nerve palsies [5–7]. Nerve transfers offer distinct advantages by reducing the distance from the nerve coaptation site to the target muscle compared with a nerve graft or nerve repair at the injured site. This feature enhances the effectiveness of reinnervation, reduces the required time compared to autografts, and holds the promise of restoring more physiological functions compared to tendon transfers. In radial nerve palsy, nerve transfers from the flexor digitorum superficialis (FDS) branch of the median nerve to the ECRB and from the FCR branch of the median nerve to the posterior interosseous nerve (PIN) have shown promise for minimizing the loss of recipient motor units through synergistic motor donor selection [7]. Bertelli’s observations indicated superior outcomes in wrist motion and finger extension in patients with nerve transfers compared with tendon transfers [5]. Several techniques have been reported for ulnar nerve palsy, including the distal anterior interosseous nerve (AIN) transfer to the deep branch of the ulnar nerve (DBUN) [6]. However, functional improvement has not been consistently observed in any of these reports. Some reports suggest reinnervation of the adductor pollicis (ADP) and interosseous muscles [6,8,9]. However, others have concluded that AIN-to-DBUN transfer does not provide sufficient intrinsic muscle reinnervation to prevent clawing [10,11].
Bertelli et al. introduced the concept of transferring the opponens pollicis motor branch (OPB) to the terminal division of the deep branch of the ulnar nerve (TDDBUN), which innervates both ADP and first dorsal interosseous (FDI) muscles [12]. However, clinical results remain fragmented across different institutions, lacking a consistent view of the functional prognosis and donor/recipient selection.
Critical considerations in nerve transfer surgery include factors such as the proximity of the donor nerve to the target muscle, type of axon, size match and suture tension between the donor and recipient, number of axons, and synergistic function of the original muscle compared to the recipient muscle [13]. Surgeons performing these procedures have reported inconsistent results, which may be partly due to differences in the number of axons between donor and recipient nerves [14,15]. Therefore, based on the theory that the differences in functional recovery outcomes between nerve transfer techniques may be due to the compatibility of donor and recipient nerve axon counts according to the technique used, our cadaveric study aimed to perform a histological investigation to determine how the compatibility of donor and recipient nerves varies according to the technique of distal nerve transfer for high radial and ulnar nerve palsy.
Methods
This study was approved by the ethical review board of Hiroshima University. The approval number is E2021-2526.
Cadaveric specimens
This study involved ten upper limbs from six cadavers donated by three females and three males aged 71–97 years (average age: 86.8 years). All cadavers were donated voluntarily by individuals during their lifetime, and consent was obtained from their families upon donation. Cadaver dissections were performed from 18 July 2021–9 July 2023. The cadavers were fixed an average of 24.7 Hours (20–28) after their death, and the cadavers were dissected an average of 166.5 days (31–353) after death. The donation and declaration forms comply with the Guidelines for Cadaver Dissection in Education and Research of Clinical Medicine in domestic societies [16] and the Act on Body Donation for Medical and Dental Education in our country. The Thiel embalming protocol, which preserves tissue flexibility and natural color, was used to preserve cadaveric specimens [17,18].
Surgical techniques
Partial median nerve to radial nerve transfer for radial nerve palsy.
A volar skin incision was made over the mid-forearm, from the antecubital fossa. The distal insertion of the pronator teres is released from the radius to facilitate exposure of the median nerve. After identification of the radial nerve, the PIN was tracked distally and released from the supinator. The superator branches were separated from the PIN.
The FCR branch of the median nerve is transferred to the PIN (FCR/PIN), and the FDS branch to the ECRB (FDS/ECRB). When multiple FDS branches were present, the branch with the largest diameter was selected as the donor nerve ().
Partial median nerve to ulnar nerve transfer for ulnar nerve palsy.
The following techniques were performed in anticipation of nerve transfer in ulnar nerve palsy. The terminal branch of the AIN was transferred to the DBUN (AIN/DBUN) as previously described [19].
The incision was made from the level of the Guyon Canal to the mid-forearm. The DBUN was identified within the Guyon Canal and dissected between the DBUN and the superficial branch of the ulnar nerve(SBUN). The AIN was cut where it started to branch in the middle portion of the pronator quadratus and mobilized proximally so that it could be transposed to the ulnar nerve (Fig 2). The DBUN was cut at the coaptation site and the distal cut end was brought into contact with the AIN for coaptation.
In addition, we performed a technique that transferred OPB to TDDBUN [12]. A skin incision was made over the first web space through the thenar region and down to the carpal tunnel. The recurrent branch of the median nerve was dissected from the carpal tunnel. The OPB was identified by its location deep in the abductor pollicis brevis (APB) muscle and was dissected intramuscularly. We identified TDDBUN by identifying the ADP muscle and separating the adductor heads deep into it (Fig 3).
In each technique, the donor and recipient nerves are moved to the coaptation site. After confirming that suturing was possible, the nerve was harvested at 10 mm from the coaptation site.
Histological analysis
Nerve samples were collected from each specimen at the level of the coaptation site and then fixed at 4 °C in 2.5% glutaraldehyde (Nacalai Tesque Inc., Kyoto, Japan) with 0.1 M Millonig’s buffer (pH 7.4) for 2–8 weeks. After post-fixation in 1% aqueous osmium tetroxide, samples were dehydrated in an ascending series of alcohols (50–100%) and propylene oxide (Katayama Chemical, Osaka, Japan). The samples were embedded in an epoxy resin (Nisshin EM Co. Ltd., Tokyo, Japan) to preserve their orientation, and cured at 60 °C for 48 h. Multiple 1 μm semithin transverse sections were cut with an ultramicrotome (Reichert-Jung, Leica, Tokyo, Japan) and stained with 1% toluidine blue (Sigma-Aldrich, St. Louis, MO, USA) for 1 min.
Photomicrographs of these sections were obtained using a fluorescence microscope (KEYENCE, Osaka, Japan) connected to a digital camera and computer. The nerve diameter, fascicle number, and cross-sectional area of the individual fascicles were measured at 100X magnification, followed by the number of myelinated axons at 200X magnification. The cross-sectional areas were measured using ImageJ software (National Institutes of Health, Bethesda, MD, USA). The total fascicle area was calculated as the sum of the cross-sectional surfaces of all the fascicles. Myelinated axons were counted using ImageJ software. The cutoff value for the inclusion of axons was 4 μm. Axon density was calculated as the ratio of the axon number to the fascicle area. The donor-to-target ratios of the axon numbers were also calculated.
Results
Each recipient and donor nerve was identified in all specimens, without anatomical variations. Furthermore, in each specimen, all donor nerves could contact the recipient nerves without tension.
For the partial median nerve to radial nerve transfer, a comparison of the donor and recipient nerves revealed that the FCR had a significantly smaller fascicle area and a significantly lower number of axons than the PIN. In contrast, FDS had a similar fascicle area and number of axons as ECRB, showing no significant differences (Table 1, Fig 4).
Calibration bars represent 100 µm.
In the partial median nerve to ulnar nerve transfer, the AIN had a significantly smaller fascicle area and fewer fascicles and axons than the DBUN. OPB was also inferior to TDDBUN in terms of fascicle number, fascicle area, and axon number (Table 2, Fig 5). The individual donor-to-recipient axon number ratios for each technique were as follows: FCR/PIN, 1:3.8; FDS/ECRB, 1:0.9; AIN/DBUN, 1:5.2; OPB/TDDBUN, 1:4.5. Axon density was not significantly different between the donor and recipient for any nerve transfer technique.
Calibration bars represent 100 µm.
Discussion
Of the four types of nerve transfers examined in this study, all methods other than FDS/ECRB resulted in a significantly lower number of donor nerve axons than the recipient. However, these techniques have been reported to restore useful functions in clinical studies [5–8,12,19,20].
Thus, nerve transfer using a donor nerve with fewer axons than that of the recipient nerve may be successful. An animal study using rat tibial nerves showed that one axon in the proximal stump could maintain up to three or four collateral sprouts [21]. Animal experiments on nerve transfer with different donor-to-recipient axon ratios using rabbit upper extremity models have shown that 50% of the normal muscle power is achieved after one-third ulnar nerve transfer to the median nerve. However, the number of axons has not been examined [22].
Given these findings, we assumed that a lack of donor axons could be tolerated up to a donor-to-recipient axon number ratio of 1:3, and compared the donor/recipient suitability for each technique. The donor/recipient axon count ratios for each procedure in the previous cadaver study are shown in Table 3.
In the FCR/PIN and FDS/ECRB transfers, both our results and those of a previous study [23] showed that a donor-recipient ratio of 1:3 was almost met, suggesting that a sufficient number of donor axons were obtained in the median to radial nerve transfer. On the other hand, another Cadavor study chose PT/ECRL for wrist extension reconstruction, with a donor-recipient ratio of 1:0.9, comparable to our FDS/ECRB results.PT/ECRL is also a promising method to restore function [24]. However, PT-ECRL is often used for tendon transfer as an aid for wrist extension until nerve recovery is achieved when performing nerve transfer with FDS/ECRB [3,7]. Martin’s study states that the FCR/PIN lacks the donor axons needed to restore function. The difference from our results may be due to the fact that FCR have individual differences in the number of branches, which may lead to variations in the number of axons in each Branch. Cheah et al. harvested the distal ends of the primary nerve branches of 23 upper limb muscles from 10 fresh-frozen cadaveric upper limbs for quantitative histomorphometry [25]. Their results were similar to ours in terms of Median to Radial nerve transfer. However, this study sampled the distal end of the primary nerve branch of the muscle, which is a different harvest site from the nerve coaptation site that simulates nerve transfer (Table 3).
In contrast, in median-ulnar nerve transfers, donor nerves were significantly deficient in AIN/DBUN, with a donor/recipient ratio of 1:4.8 in this study and in a study by Schenck et al. [15]. Üstün et al. reported a higher rate of donor axons compared with our results [26]. Still, this difference may be because the samples were taken from the level of the pisiform, which is not the original coaptation site, as Schenck et al. mentioned [15]. Although the SBUN and DBUN were identified at the level of the Guyon’s canal, and these nerves were divided proximally, crossing fascicles were observed between the two branches, and differences in technique in deciding whether to include these in the DBUN or SBUN may have resulted in different axon counts in the recipients.
For OPB/TDDBUN transfer, although the donor/recipient ratio values differed between our results and those reported by Bertelli et al.[12], the number of donor axons is considered to support functional recovery. The cause of this discrepancy in the results remains uncertain; however, the difference between our measurements and those in the literature can be partially explained by the inclusion and exclusion criteria in axon identification. However, because there is only one previous report, further verification is required.
These results suggest that the number of donor axons in median-to-radial nerve transfer is adequate. However, the number of donor axons in median-to-ulnar nerve transfer is inadequate, especially for AIN/DBUN. A previous anatomical study suggested that insufficient FDI and ADP reinnervation after nerve transfer from the AIN to the DBUN may be due to insufficient axons in the AIN compared with the DBUN [15]. Some studies have stated that this inadequate reinnervation may be due to the low donor-to-recipient axon ratio and insufficient axons in AIN [10,11].
Combining AIN-DBUN and OPB-TDDBUN can compensate for the shortage of donor axons in the AIN/DBUN. In addition, the TDDBUN component of the DBUN can be subtracted from the number of recipient axons donated by the AIN, thus compensating for the lack of donor axons in AIN-DBUN transplantation. Indeed, the donor-recipient AIN:(DBUN-TDDBUN) ratio improved to 1:3.5 when AIN-DBUN and OPB-TDDBUN were performed simultaneously. (Table 3) Based on the results of this study, we recommend the combined nerve transfer of AIN-DBUN and OPP-TDDBUN for ulnar nerve palsy. As the two surgical sites are separated into the forearm and palm, simultaneous surgeries do not affect each other.
Several studies have histologically tested the compatibility of donor recipients with upper-limb nerve transfers using cadavers [15,23,24,26]. However, each of these studies examined only one technique. The strength of our study is that, in addition to verifying the results of existing studies, it clarifies the differences in the donor-recipient ratio for each nerve transfer technique by evaluating several techniques in the same cadaver. Evaluating multiple techniques can help predict functional outcomes between procedures and provide simulation guidance, such as the simultaneous use of the AIN-DBUN and OPB-TDDBUN for ulnar nerve palsy, as demonstrated in this study.
However, a donor/recipient ratio of at least 1:3 was preferable in our study. This idea was based on the findings of Jiang et al., who reported that up to 3–4 collaterals can be developed by one axon. In contrast, functional compensation is often observed in partially denervated muscles [27]. Gordon et al. described that single motor units enlarged to approximately five times their original size, resulting in the ability to compensate for up to 80% of motoneuron loss [28]. Based on motor unit enlargement, nerve coaptations with an axon ratio between 1:3 and 1:5 can be expected to have the same motor recovery ability as that of coaptations with an optimal ratio greater than 1:3.
This study has some limitations. First, the specimens used in this study were fixed cadavers rather than living organisms. Samples from fixed cadavers may differ in thickness from the nerve bundles and myelin sheaths in the live body. However, there is no difference in the number of axons between living and fixed tissues.
Second, the average age of the cadavers used in this study was 86.8 years old, which is much higher than the age range of patients who actually require nerve transfer surgery. A reduction in the diameter or density of myelinated fibers has been reported with aging in the peripheral nerves of humans and several animals [29,30]. The relationship between age and number of axons is unclear, although it is possible that they decrease. However, it is possible to compare the donor and recipient nerve sizes within the same individual. In addition, the surgical techniques used in the cadaver study influenced the results of the study, such as a larger surgical field than in actual surgery, making the surgery easier to perform and overlooking vascular preservation and hemostasis. Finally, soft tissue flexibility was reduced in fixed cadavers compared with living or freshly frozen cadavers, which may have affected the distance and route of nerve transfer. However, because Thiel embalming can retain flexibility close to that of actual living tissue, it has been used for several types of surgical training [31,32], 3D anatomical studies of the brachial plexus [33], and anatomical studies to avoid proximal radius PIN injuries during posterior and lateral approaches [34]. The simulation of nerve transfer in this study allowed for the same surgical technique as in living bodies, and the impact of thiel embalming on the surgical technique was considered limited.
Acknowledgments
We appreciate Mr. Nakatani and Mr. Shimizu of the Department of Anatomy and Developmental Biology, Graduate School of Biomedical and Health Sciences at Hiroshima University for the technical assistance with the cadaver preparation. The work was conducted with the facilities in the Natural Science Center for Basic Research and Development (N-BARD) at Hiroshima University NBARD-00165. And, We appreciate Ms. Morimoto, Ms. Tanaka, Ms. Ishikawa of N-BARD at Hiroshima University for their contribution to the preparation of semi-thin sections and tissue staining in this work.
References
- 1. Smith RJ. Extensor carpi radialis brevis tendon transfer for thumb adduction--a study of power pinch. J Hand Surg Am. 1983;8(1):4–15. pmid:6827050
- 2. Tsuge K. Tendon transfers for radial nerve palsy. Aust N Z J Surg. 1980;50(3):267–72. pmid:6931587
- 3. Lowe JB 3rd, Sen SK, Mackinnon SE. Current approach to radial nerve paralysis. Plast Reconstr Surg. 2002;110(4):1099–113. pmid:12198425.
- 4. Uraloğlu M, Orbay H, Açar HI, Sensöz O. Flexor pollicis brevis adductorplasty: an alternative method in ulnar nerve paralysis. Ann Plast Surg. 2006;57(1):110–4. pmid:16799320
- 5. Bertelli JA. Nerve Versus Tendon Transfer for Radial Nerve Paralysis Reconstruction. J Hand Surg Am. 2020;45(5):418–26. pmid:32093993
- 6. Novak CB, Mackinnon SE. Distal anterior interosseous nerve transfer to the deep motor branch of the ulnar nerve for reconstruction of high ulnar nerve injuries. J Reconstr Microsurg. 2002;18(6):459–64. pmid:12177812
- 7. Ray WZ, Mackinnon SE. Clinical outcomes following median to radial nerve transfers. J Hand Surg Am. 2011 ;36(2):201–8. pmid:21168979
- 8. Battiston B, Lanzetta M. Reconstruction of high ulnar nerve lesions by distal double median to ulnar nerve transfer. J Hand Surg Am. 1999;24(6):1185–91. pmid:10584939
- 9. George SC, Burahee AS, Sanders AD, Power DM. Outcomes of anterior interosseous nerve transfer to restore intrinsic muscle function after high ulnar nerve injury. J Plast Reconstr Aesthet Surg. 2022;75(2):703–10. pmid:34789435
- 10. Arami A, Bertelli JA. Effectiveness of Distal Nerve Transfers for Claw Correction With Proximal Ulnar Nerve Lesions. J Hand Surg Am. 2021;46(6):478–84. pmid:33341296
- 11. Gross JN, Dawson SE, Wu GJ, Loewenstein S, Borschel GH, Adkinson JM. Outcomes after Anterior Interosseous Nerve to Ulnar Motor Nerve Transfer. J Brachial Plex Peripher Nerve Inj. 2023;18(1):e1–e5. pmid:36644673
- 12. Bertelli JA, Soldado F, Rodrígues-Baeza A, Ghizoni MF. Transferring the Motor Branch of the Opponens Pollicis to the Terminal Division of the Deep Branch of the Ulnar Nerve for Pinch Reconstruction. J Hand Surg Am. 2019 ;44(1):9–17. pmid:30366737
- 13. Mackinnon SE, Novak CB. Nerve transfers. New options for reconstruction following nerve injury. Hand Clin. 1999;15(4):643–66, ix. pmid:10563268
- 14. Millesi H, Schmidhammer R. Nerve fiber transfer by end-to-side coaptation. Hand Clin. 2008;24(4):461–83, vii. pmid:18928894
- 15. Schenck TL, Stewart J, Lin S, Aichler M, Machens HG, Giunta RE. Anatomical and histomorphometric observations on the transfer of the anterior interosseous nerve to the deep branch of the ulnar nerve. J Hand Surg Eur Vol. 2015;40(6):591–6. pmid:25261412
- 16. Shichinohe T, Kondo T, Date H, Hiramatsu M, Hirano S, Ide C, et al. Guidelines for cadaver dissection in education and research of clinical medicine (The Japan Surgical Society and The Japanese Association of Anatomists). Anat Sci Int. 2022;97(3):235–40. pmid:35606673
- 17. Thiel W. The preservation of the whole corpse with natural color. Ann Anat. 1992;174(3):185–95. pmid:1503236
- 18. Thiel W. Supplement to the conservation of an entire cadaver according to W. Thiel. Ann Anat. 2002;184(3):267–9. pmid:12061344
- 19. Brown JM, Yee A, Mackinnon SE. Distal median to ulnar nerve transfers to restore ulnar motor and sensory function within the hand: technical nuances. Neurosurgery. 2009;65(5):966–77; discussion 977-8. pmid:19834412
- 20. Patterson JMM, Russo SA, El-Haj M, Novak CB, Mackinnon SE. Radial Nerve Palsy: Nerve Transfer Versus Tendon Transfer to Restore Function. Hand (N Y). 2022;17(6):1082–9. pmid:33530787
- 21. Jiang BG, Yin XF, Zhang DY, Fu ZG, Zhang HB. Maximum number of collaterals developed by one axon during peripheral nerve regeneration and the influence of that number on reinnervation effects. Eur Neurol. 2007;58(1):12–20. pmid:17483580
- 22. Lutz BS, Chuang DC, Chuang SS, Hsu JC, Ma SF, Wei FC. Nerve transfer to the median nerve using parts of the ulnar and radial nerves in the rabbit--effects on motor recovery of the median nerve and donor nerve morbidity. J Hand Surg Br. 2000;25(4):329–35. pmid:11057998
- 23. Sukegawa K, Suzuki T, Ogawa Y, Kobayashi T, Matsuura Y, Kuniyoshi K. A Cadaver Study of Median-to-Radial Nerve Transfer for Radial Nerve Injuries. J Hand Surg Am. 2016;41(1):20–6. pmid:26710730
- 24. Martins A, Novais Gameiro L, Clavert P. Transfers of the flexor carpi radialis and pronator teres nerves to radial nerve branches: a cadaveric study. J Hand Surg Eur Vol. 2022;47(9):968–70. pmid:35765852
- 25. Cheah A, Lee EY, Lim AYT. Upper Extremity Axon Counts and Clinical Implications for Motor Nerve Transfer. Plast Reconstr Surg. 2019;144(6):1044e–50e. pmid:31764654
- 26. Ustün ME, Ogün TC, Büyükmumcu M. Neurotization as an alternative for restoring finger and wrist extension. J Neurosurg. 2001;94(5):795–8. pmid:11354412
- 27. McComas AJ, Sica RE, Campbell MJ, Upton AR. Functional compensation in partially denervated muscles. J Neurol Neurosurg Psychiatry. 1971;34(4):453–60. pmid:4938133
- 28. Gordon T, Yang JF, Ayer K, Stein RB, Tyreman N. Recovery potential of muscle after partial denervation: a comparison between rats and humans. Brain Res Bull. 1993;30(3–4):477–82. pmid:8457897
- 29. Sakita M, Murakami S, Fujino H. Age-related morphological regression of myelinated fibers and capillary architecture of distal peripheral nerves in rats. BMC Neurosci. 2016;17(1):39. pmid:27342571
- 30. Verdú E, Ceballos D, Vilches JJ, Navarro X. Influence of aging on peripheral nerve function and regeneration. J Peripher Nerv Syst. 2000;5(4):191–208. pmid:11151980
- 31. Suzuki T, Suzuki-Narita M, Kubota K, Mori C. Updates on cadaver surgical training in Japan: a systematic facility at Chiba University. Anat Sci Int. 2022;97(3):251–63. pmid:35522373
- 32. Yiasemidou M, Roberts D, Glassman D, Tomlinson J, Biyani S, Miskovic D. A Multispecialty Evaluation of Thiel Cadavers for Surgical Training. World J Surg. 2017;41(5):1201–7. pmid:28144746
- 33. Van de Velde J, Bogaert S, Vandemaele P, Huysse W, Achten E, Leijnse J, et al. Brachial plexus 3D reconstruction from MRI with dissection validation: a baseline study for clinical applications. Surg Radiol Anat. 2016;38(2):229–36. pmid:26298831
- 34. Hohenberger GM, Schwarz AM, Maier MJ, Grechenig P, Dauwe J, Grechenig C, et al. Safe zone for the posterior interosseous nerve with regard to the lateral and posterior approaches to the proximal radius. Surg Radiol Anat. 2018;40(9):1025–30. pmid:29619502