Cutaneous nerve fields of the anteromedial lower limb—Determination with selective ultrasound‐guided nerve blockade

Abstract Background This study aimed to determine the peripheral cutaneous nerve fields (CNF), their variability, and potential overlap by selectively blocking the intermediate (IFCN) and medial (MFCN) femoral cutaneous nerves and the infrapatellar branch of the saphenous nerve (IPBSN) in healthy volunteers. Methods In this prospective study, ultrasound‐guided nerve blockades of the IFCN, MFCN, and IPBSN in 14 healthy volunteers were administered. High‐frequency probes (15–22 MHz) and 1 ml of 1% lidocaine per nerve were used. The area of sensory loss was determined using a pinprick, and all fields were drawn on volunteers' skin. A three‐dimensional (3D) scan of all lower limbs was obtained and the three CNF and their potential overlap were measured. Results The mean size of innervation areas showed a high variability of peripheral CNF, with 258.58 ± 148.26 mm2 (95% CI, 169–348.18 mm2) for the IFCN, 193.26 ± 72.08 mm2 (95% CI, 124.45–262.08 mm2) for the MFCN, and 166.78 ± 121.30 mm2 (95% CI, 94.1–239.46 mm2) for the IPBSN. In 11 volunteers, we could evaluate an overlap between the IFCN and MFCN (range, 4.11–139.68 ± 42.70 mm2), and, in 10 volunteers, between the MFCN and IPBSN (range, 11.12–224.95 ± 79.61 mm2). In only three volunteers was an overlap area found between the IFCN and IPBSN (range, 7.46–224.95 ± 88.88 mm2). The 3D‐scans confirmed the high variability of the peripheral CNF. Conclusions Our study successfully determined CNF, their variability, and the overlap of the MFCN, IFCN, and IPBSN in healthy volunteers. Therefore, we encourage physicians to use selective nerve blockades to correctly determine peripheral CNF at the anteromedial lower limb.


K E Y W O R D S
anesthesia, lower extremity, neuralgia, pain, peripheral nerves, ultrasonography
The classical teaching about CNF covered by cutaneous nerves at the anteromedial limb in most cases is limited by several factors. A knowledge of the CNF at the anteromedial lower limb is based on anatomical textbooks (Ladak, Tubbs, & Spinner, 2014;Standring & Gray, 2008) or observational studies of hypoesthesia after several surgeries (Cohen et al., 2017;Tanavalee, Limtrakul, Veerasethsiri, Amarase, & Ngarmukos, 2016). These descriptions lack differentiation between distinct cutaneous nerve branches (e.g., medial or intermediate femoral cutaneous nerves), do not describe the possible overlap of CNF, and do not distinguish the possible variabilities. Hence, it is urgent that better ways be found to understand these differences, and thus, enhance the correct localization of CNF at the anteromedial lower limb. This may increase the correct identification of the level of nerve damage, and consequently, avoid insufficient treatment and the chance of chronic neuropathic pain. Further, accurate localization of the CNF may increase our basic knowledge about the CNF and the possible variabilities.
Therefore, in the present study, we hypothesized that selective nerve blockades of the intermediate (IFCN) and medial (MFCN) femoral cutaneous nerves and the infrapatellar branch of the saphenous nerve (IPBSN) can depict the CNF. We wished to determine the CNF (area of sensory loss), their variability, and potential overlap by selective injection (blockades) of local anesthetics around these branches under ultrasound guidance in healthy volunteers.

| Volunteers
Healthy volunteers were recruited via notices at the Department of Biomedical Imaging and Image-guided Therapy of the University of Vienna and word-of mouth acquisition. Written, informed consent was F I G U R E 1 Illustration of the course of the intermediate (green) and medial (red) femoral cutaneous nerves and the infrapatellar branch of the saphenous nerve (blue) [Color figure can be viewed at wileyonlinelibrary.com] obtained from all volunteers. Inclusion criteria were age over 18, and exclusion criteria were any known polyneuropathy, myopathy, chronic disease that could cause peripheral neuropathy, current or lower limb pain, previous lower limb surgeries, numbness, hypoesthesia/ paresthesia around the knee, previous use of any local anesthetics of any serotype for any reason within 1 year prior to enrollment, and potential contraindications associated with local anesthetic administration. Volunteers' history, medications, current age, size, and weight were assessed before the examination. approximately 10-15 cm above the knee joint line. The sartorius muscle was then carefully screened distally until a tubular structure that pierced the fascia lata overlying the sartorius muscle, and the nerve lying in a fat pad, was assumed to be the IPBSN. To correctly assign the IPBSN to the origin from the saphenous nerve, the former was followed proximally until its origin. For the MFCN and IFCN, ultrasound examinations started with a transverse view at the middle/ distal third of the anterior thigh superficial to the sartorius vastus medialis and rectus femoris muscles. The subcutaneous fat overlying these muscles was then carefully screened distally and proximally for tubular, in contrast to fat, hyperechoic structures that were presumed to be the MFCN and the IFCN. The IFCN was explored at the level overlying the vastus medialis, the rectus femoris, and, sometimes, the sartorius muscle, whereas the MFCN was explored further medially in the region of the vastus medialis and sartorius muscle. If detected, the branches were followed proximally to assess the origin from the femoral nerve so as not to confound the branches with the lateral femoral cutaneous nerve. The branches were then screened for the best locations for infiltration before splitting into further branches. To avoid a concomitant blockade of surrounding nerves, tissue layers (nerves running in fascial layers or fat pads) approximately 1-3 cm away from the origin of the nerve branches were used as blockade locations.

| Ultrasound technique and interventions
Every nerve branch was infiltrated on a different day (in total, three consecutive days), beginning with the IFCN on Day 1, followed by the IPBSN on Day 2, and ending with the MFCN on Day 3.
The technique used to blockade the nerve branches was as follows. For all injections, the skin surrounding the needle entry point was cleaned with betadine solution. Local anesthetic administration was performed with a lateral-to-medial or medial-to-lateral approach (depending on the nerve location) using a 23-gauge, 1.5-3.0-in. needle (depending on the muscle size and depth). Of note, 1 cc of lidocaine (Xylanaest purum 2%, Gebro Pharma GmbH, Austria) was injected adjacent to the nerve. All interventions were accomplished using a free-hand technique. The radiologist oriented the needlesyringe parallel to the transducer's imaging plane and angled at approximately 45 relative to the transducer's footprint. The needle was slowly advanced under permanent ultrasound guidance. Once the needle tip was seen adjacent to the nerve, lidocaine was delivered under real-time monitoring. The depth of infiltration (needle tip) was measured on the ultrasound image and the side of infiltration was marked on the volunteers' skin. All injection sites were screened for hematoma or other structural alterations 10 min after injection.
Finally, all subjects were asked to report all adverse events during the study by completing a checklist that included: pain at the injection site lasting more than 3 days; large bruise or blood clot at the injection site; uncontrolled bleeding at the injection site; increase of pain, skin rash, or itchiness; and fever or flu-like symptoms. Figure 2 provides examples of ultrasound-guided nerve blockades of all three nerves.

| Sensory blockade evaluation and drawing of innervation areas
The area of sensory loss was assessed by an independent investigator (C.S.) every 5 min for the first 30 min. The blocked area was tested using a pinprick and compared with the contralateral side. The response was scored with the following scale: 0 = no perception; 1 = reduced sensation; and 2 = normal sensation compared with the other side. Finally, a sensory map was drawn with a surgical marker on patients' skin relative to the marked point, which represented Grades 0 and 1 on the scale. For that reason, three different colors were used: green = IFCN; blue = IPBSN; and red = MFCN.

| Post-processing and 3D innervation mapping
A hand-held three-dimensional (3D) scanner system (3D Systems 350470 Sense2 3D-Scanner) was used to obtain photo-textured 3D surface models of the examined lower limbs. The scanner is not certified for medical use and has not been validated or proven for such purposes. Scans of all volunteers' lower limbs were performed along a circular path around the longitudinal axis of standing subjects with the help of a tripod-based guidance system. All scans were performed on Day 3 after all CNF were drawn on volunteers' skin. If necessary, scans were carried out in multiple rows and the resulting 3D models were merged in postprocessing. Postprocessing of the surface models was performed with Cinema 4D R18. To confirm the accuracy of the surface models and to determine the reliability of the 3D scanner, distances between multiple control points were measured on the lower limbs and verified in the 3D models. For that reason, 20-40 cm lines were drawn on a healthy volunteer who was not taking part in the study. In vivo measurements were matched with those obtained by the 3D scan of the volunteer. Measurements deviated by less than 1 mm. Finally, all scanned CNF and their overlap were measured.   There were no significant differences between CNF in male and female volunteers for the MFCN (p = .106) and IPBSN (p = .147),
A correlation analysis showed no correlations or significant differ-   (Anloague & Huijbregts, 2009;Gray & Clemente, 1985;Thiel, 1999). This knowledge is further supported by the actual work of Pivec et al. (2018) where this pattern was described using ultrasound to detect the nerves. Nevertheless, the course, distribution, and branching pattern seem to be highly variable and should be further addressed in future anatomical works.

| 3D innervation mapping
Although cutaneous nerve damage at the anteromedial lower limb is known to cause neuropathic pain (Ginanneschi et al., 2013;Laffosse et al., 2011;Pivec et al., 2018;Sundaram et al., 2007), it seems to be a minor concern in our clinical routine. This may be primarily explainable by the fact that, until recently, these nerves were not accurately identifiable by imaging and only little is known about their anatomical distribution. In 2006, Lundblad, Kapral, Marhofer, and Lonnqvist (2006) first identified and blocked the main branch of the IPBSN by ultrasound. One decade later, the study group of Pivec et al. (2018) observed the different branches of the anterior femoral cutaneous nerve (IFCN and MFCN) with ultrasound. Based on these findings, our study aimed to determine the anatomical pattern of the CNF. To the best of our knowledge, until now, neither anatomical study nor anatomical textbook has described these patterns (Ladak et al., 2014;Standring & Gray, 2008 damage. In addition to the described variations, the most constant pattern was found for the IPBSN that always supported the infrapatellar region and the IFCN, which always supported the thigh in a midline position. Further, the posterior popliteal fossa was innervated by the IPBSN (No. 2) and MFCN (No. 14), which is normally supported by the posterior femoral cutaneous nerve (Ladak et al., 2014). Some of these findings may be explained by the variable course (Kerver et al., 2013;Pivec et al., 2018) or the communication between several nerves, of which some form plexuses around the knee (Gray & Clemente, 1985;Ladak et al., 2014;Thiel, 1999 (Greenberg, 2003;Ladak et al., 2014). Nevertheless, there remains much ambiguity about the CNF at the anteromedial lower limb. This article is not intended as a comprehensive work in this regard, but rather, is intended to encourage future research. A primary limitation previously, which was that cutaneous nerves were not detectable via any imaging modality until recently, has now been addressed by the excellent resolution of ultrasound (Meng et al., 2015;Tagliafico, Bignotti, Cadoni, Perez, & Martinoli, 2015). With a slight adaption of our presented assessment protocol (e.g., perhaps a slight elevation of the lidocaine dose, or an enhancement of the 3D scan), in vivo determination of CNF of a whole extremity seems to be possible in the near future. Our study has several limitations. As mentioned above, we cannot exclude concomitant blockade of different cutaneous nerves. By precisely evaluating the spread of the local anesthetic under ultrasound-guidance and avoiding blockade of a cutaneous nerve in direct proximity to another, we tried to prevent such a mistake. In two cases, we could not find a single MFCN or IFCN. Sometimes, only one, rather than two, branches existed (Anloague & Huijbregts, 2009)  No. 3, we were unable to detect the IFCN, and therefore, no IFCN of the CNF was evaluable. Another limitation is the fact that we included only three cutaneous nerve branches and did not include the obturator or the genitofemoral nerve to evaluate the CNF of the whole anteromedial thigh. This was mainly limited by ethical considerations to avoid five nerve blockades, with all the possible side effects, on healthy volunteers. Moreover, it has to be noted that, due to the assessment protocol, to avoid concomitant blockades, and therefore, blocking the nerve rather than the origin more distally, we could not observe the most proximal part of any of the CNF in our study.
Other limitations to be noted are the small sample size and the inherent subjectivity in pinprick testing, which may lead to biases in correctly evaluating CNF.
In conclusion, our study successfully determined CNF, their variability, and overlap of the anterior femoral cutaneous nerves and the infrapatellar branch of the saphenous nerve in healthy volunteers.