We dissected 24 Thiel embalmed feet in order to better understand the anatomical topography of the ankle’s medial side. The origin and distribution of TN, MPN, LPN, MCN and BN were registered according to specific landmarks and possible compression points.

Chronic heel pain syndrome is a commonly encountered condition among surgeons dedicated to foot and ankle [1, 2, 17, 18] but an accurate diagnosis is not always easy. A complex regional anatomy opens the door to many etiologies like space occupying lesions, bony prominences compression, posttraumatic or perineural fibrosis, muscle anomalies, among others that, altogether, make the neurological compression one of the most cited causes. However, in a considerable percentage of cases, the true cause can remain unknown [2, 19, 20].

The TT is a fibro-osseous tunnel beneath the FR, behind and inferior to the MM [7, 21]. Rosson et al. and Tekin et al. stated that, under FR, there were increased pressures leading to the clinical findings on TTS, only relieved with the retinaculum release [21, 22]. Also, some foot and ankle positions like increased eversion, dorsiflexion, or combined dorsiflexion-eversion may significantly increase TN tension on the TT [23,24,25]. Therefore, the compression of the TN and/or its main branches on the medial side of the ankle under the FR has been already established as a cause of the symptoms on TTS.

In 1984, Dellon and Mackinnon showed that in 90% of their cases, the bifurcation of the TN was within 1.0 cm of the DML and that in 95% of the cases it was under the FR [16]. More recently, Mattos et al. [26] and Gamie et al. [27] stated, on their work, that the MPN has its origin proximal to the DML in 95 and 100% of the cases, respectively. Moreover, Moroni et al. [8] presented an average distance of 16,4 mm from the center of DML to the TN bifurcation, a result close to the 18 mm that we obtained. However, despite the average distances being close to each other, the values found within each study are very variable ranging from 40 mm to -10 mm on Moroni’s [8] and from 0 mm to 40 mm on ours.

The FR starts, approximately, 20 mm proximal to DML and ends 20 mm distal to it [16]. We found the division of the TN within less than 20 mm from the center of the DML in 14 (58%) of our specimens and Moroni et al. on approximately 73% [8]. Bilge et al. proposed a classification for the TN division according to a reference line (1 cm width) from the tip of the MM to the medial process of the calcaneal tuberosity [28]. Type I represented a bifurcation proximal to the line, Type II overlying it and Type III distal to it. Their work concluded that 84% of the cases were type I, 12% type II and 4% type III [28]. This data is in close relation with ours, where, according to the mentioned classification, 88% of the cases were type I and 12% type II.

Being a site of compression and a well-defined target in TTS’s surgical treatment, the TT demands a good understanding of the anatomy behind the syndrome. Our results corroborated previous works that suggested a division of the TN at a variable distance from the superior limit of the FR [8, 16, 26, 27]. It is more common to find this division proximal to the FR which, in turn, suggests that it will be more common to find a compression of the main branches under the FR than the TN itself. Further in vivo studies are needed to understand if there is any relation between the occurrence of a TTS and the division of the TN outside the FR.

We tried to identify the relationship of the neural structures with anatomical landmarks that a surgeon can use in a daily basis and found that, on average, we could encounter the MPN, the LPN, the BN and the MCN, at 33 mm; 39 mm; 43 mm and 53 mm, respectively, from the center of the MM, on top of the DML. Moroni et al. did similar measurements and found the MPN, the LPN and the BN, on average, at 37 mm, 44 mm, 48 mm, respectively [8]. The values obtained by us and Moroni et al. are similar but we still can find a great variability among the reported distribution of TN and its branches on the current literature [7, 16, 29].

We found that there was a strong correlation between the DML’s length and the point where we encountered the MPN, the LPN the BN and the MCN. This means that the distance from the MM and the point where we find the MPN, the LPN and the BN under the FR is strongly correlated with the total distance from the MM and the medial process of the calcaneal tuberosity and, therefore, the size of the ankle. On that account, we hypothesize that the pathway for these nerves could be more predictable than we previously thought if we consider the size of the ankle. Heimkes et al. described that, at the inferior retinaculum limit, both MPN and LPN enter two separate fibro-osseous tubes [7]. We put on consideration that these tubes may condition a more regular distribution of the branches proximally. In this domain, Mattos et al. also found that MPN, along with MCN, was the nerve with the most constant distribution in relation to the DML [26]. There is still limited information on the literature about predictability of the BN trajectory however, our work can give some new insights on this topic, suggesting that distance from the MM where we find BN in the ankle is also strongly correlated with the size of the ankle itself.

This notion of a strong correlation can be extremely important as most works, ours included, report average distances where we can find the nerves on the medial side of the ankle. However, we should be aware that this distance is strongly correlated with the total distance from the MM to the medial process of the calcaneal tuberosity and, therefore, when we analyze these values we should always take into consideration the relation with total length of the DML and not the absolute value. We also believe that this information can shed some light into the way we analyze the trajectory of these nerves.

Singh and Kumar stated that the medial septum, a dorsal extension of the medial border of the central plantar aponeurosis, also referred to as the deep fascia of the AH in some studies, could be the most important compressive structure, besides the FR, behind TTS [5]. Ghosh et al. also stated that both the LPN and MPN could be entrapped under the AH and this possibility should be excluded if no compression at the FR was found [30]. Therefore, we tried to address its importance by checking the distribution of the MPN, the LPN and the BN along the distal edge of the Heimkes Triangle, simulating the AH and the septum. We found them at 44 mm, 31 mm and 24 mm, respectively, from the medial process of the calcaneal tuberosity. Singh and Kumar also reported mean distances of points of penetration of TN branches through the foot septae from the posterior surface of the calcaneus of 56 mm for the MPN, 41 mm for the LPN and 29 mm for the BN [5]. The results obtained by us and Singh and Kumar are quite different. However, few works on the literature addressed these measurements and the data available is still scarce to determine if these structures have, in fact, a variable and unpredictable trajectory or if, on the other hand, could have a foreseeable passage. Moreover, slightly different anatomic landmarks were used between our (same as Dellon and Mackinnon [16] used) and Singh and Kumar’s [5] works which may also contribute to the differences found.

Surgical treatment of TTS is reserved for patients who do not respond to conservative measures. Unpredictable results may characterize surgical options, with some authors reporting relief of symptoms in only 50% of cases, others with poor results at long-term follow-up in more than 50% [2, 6]. Many studies focus on the necessity of including the nerve’s crossing through AH and medial septum on the surgical strategy for treating the TTS, but the best surgical strategy is highly controversial. Mullick and Dellon reported excellent results in 82% of the patients, after a 3 years follow-up, with decompression of four ankle tunnels, the FR and the tunnels for the MPN, the LPN and the MCN [6]. Thus, the distal tarsal tunnel seems to be undeniably connected to TTS despite remaining a challenging anatomical region. Our work tries to make it easier to understand the relationship between some landmarks and the nerves that need to be released under the AH and medial septum.

More recently, an ultrasound-based approach to the TTS was introduced with similar outcomes compared to open releases, minimizing soft tissue dissection, potential wound complications like infections and scars, with reducing recovery time and avoiding offloading [19, 31, 32]. The procedure is based on accurate identification of neural structures and a minimally-invasive cut of the FR and AH fascia [19, 31]. The techniques described are quite demanding and the surgeon must have a clear understanding of how the nerves are displayed, especially when releasing the distal TT [31].

Endoscopic procedures are also coming to light and precise anatomical references are needed so that either the correct placement of the portals and the safety of the technique in relation to the neurologic structures are assured [33]. The two portals endoscopic procedure was first described by Day and Naples [34]. On their study, the proximal portal was located over the palpable FR, about 1 cm proximal to the MM [34]. El Shazly et al. stated that the main indication for the endoscopic technique would be idiopathic proximal TTS with main trunk compression and proposed a revised proximal portal, 4 cm from the MM, that would facilitate the retinaculum release [33]. Our results could be in line with this new perspective about the proximal portal, suggesting that surgeons must take into consideration that the division of the TN may be variable and that can be found as proximal as 40 mm from the DML. This knowledge would be fundamental for safe portal placement and for a correct visualization of this division.

Awareness of possible points of division and trajectories is of tremendous importance for procedures like nerve blocks of the tibial nerve or its branches. Tibial nerve block was reported as a safe and effective method for controlling pain after outpatient surgery of hallux valgus [35]. The average distances from specific anatomical landmarks for the TN and its branches that are shown in our work gives important information for safe and effective blocks.

Lee at al. stated that an important percentage of nerve injuries after ankle replacement were due to excessive stretching during retraction, inadequate nerve release or improper protection during the incision [36]. Once again, an adequate knowledge of the position of the tibial nerve and its main branches according to specific landmarks like MM or the medial process of the calcaneal tuberosity will provide a greater security approaching this area.

Our study had some limitations. As it was a work based on cadaveric specimens the possible sample size was small (24 feet) and therefore our conclusions are limited. Moreover, all the subjects included were Caucasians and with more than 60 years old which can condition, in some degree, the results obtained. Additionally, we are aware that our study has also other limitations that are generally observed in studies that are performed in cadavers. However, we had done everything possible to minimize, at least in part, these limitations. One limitation is related to the changes induced by death and fixation procedures in the volume and trophicity of tissues. We tried to minimize this problem by using bony landmarks. Furthermore, we have measured some reference parameters and the results were very similar to the ones reported earlier in the scientific literature, a point that unequivocally provides robustness to our results. On the other hand, we highlight that our study was optimized by using full-body cadavers. In fact, in this way, all the studied structures, particularly the nerves, were intact. Additionally, we re-emphasize that all dissection procedures were carefully performed to maintain, as much as possible, the normal relationships between structures. Finally, another limitation of the present study is the absence of clinical information related to the studied specimens which prevented any correlation between anatomical and clinical factors.