Anatomical basis for ultrasound-guided infiltration of the saphenous nerve in the subsartorial canal

Abstract

The present work is aimed studying the visibility and position of the vasto-adductor membrane with ultrasonography and demonstrating that injection performed under this membrane allows to infiltrate the saphenous nerve. It was analyzed with ultrasonography in four cadaveric subjects and in 13 volunteers. This membrane was clearly visible and methylene blue was located underneath it after injection in all cadaveric subjects. This study demonstrates that it can be used as a reliable anatomic landmark when performing an injection for both regional anesthesia and in the treatment of saphenous nerve tunnel syndrome.

Introduction

The saphenous nerve (SN) is the largest sensory terminal branch of the femoral nerve [1, 2]. It runs 1–2 cm proximal to the inguinal ligament and then at the deep aspect of the anterior thigh follows the course of the femoral artery through the femoral triangle and subsartorial canal- also named “Hunter canal” or “adductor canal”—to the adductor hiatus. Here it divides into an infrapatellar branch [3, 4] which is distributed to the skin of the anterior inferior aspect of the knee, and a second branch, which is distributed to the medial aspect of the knee, leg and ankle [5, 6] (Fig. 1). It provides sensory innervation to the medial aspect of the lower leg and foot. Blockade of the SN may be helpful in both regional anesthesia (RA) and in the treatment of saphenous nerve tunnel syndrome [7–9]. Improved success of such a procedure in RA has been reported when the latter is performed both with the use of external landmarks, including perifemoral, subsartorial, transsartorial or medial femoral condyle approaches [10, 11] or under the guidance of ultrasound (US) [12, 13]. Similarly, injection of steroids into the subsartorial canal has been advocated in subsartorial tunnel syndrome that leads to disabling pain of the medial aspect of the thigh, knee and leg [14–16] as the most reliable therapeutic and diagnostic tool. While the procedure can be performed on the basis of simple clinical landmarks [17–19], the use of US to guide the needle within the subsartorial canal seems more reliable concordant with similar anesthetics studies. The vasto-adductor membrane (VAM) [20], a thin but strong membrane that covers the saphenous nerve, and separates it from the subsartorial plexus in the adductor canal, is known as a good sonographic anatomical landmark to guide infiltration of the saphenous nerve.

Fig. 1.

The anatomy of the subsartorial canal. The sartorius muscle (S) is retracted showing the saphenous nerve, and the Vasto-adductor membrane. VM vastus medialis muscle, AM adductor magnus muscle, AL adductor longus muscle, G gracilis muscle, S sartorius muscle, RF rectus femoris muscle

The purpose of our study was to determine the visibility and position of the VAM along the thigh with US in a population of cadaveric specimens and volunteers and demonstrate that injections performed deep to this membrane include the saphenous nerve.

Materials and methods

Anatomical study

Eight thighs of four cadaveric subjects including two males and two females with a mean body mass index (BMI) of 22.7 (20–25.4) were selected and placed in the supine position. Permission to dissect the cadaveric specimens was granted by our Anatomy department. The same operator with more than 10 years’ experience in musculoskeletal imaging performed an US examination of the thighs of each subject. The anterior aspect of each thigh was imaged with a 12–5 MHz linear array transducer (model HDI 5000; Philips Medical Systems, Eindhoven, Netherlands). The subsartorial canal, lying deep to the sartorius muscle, was examined in an axial plane to identify the linear membrane around the femoral pedicle, neighboring the saphenous nerve which is considered to represent the VAM. Once detected, dynamic scanning of the VAM was performed (more effective than static imaging), sweeping the probe down from the hip to the knee. The structure to be considered as VAM had to be attached to the vastus medialis muscle laterally and the adductor magnus/longus muscle medially. Membrane visibility was scored on a 3-level scale: 0-not visible; 1-identified with difficulty; 2-clearly identified. The exact position of the VAM along the thigh was measured using the same following landmarks as reported by Tubbs [20]. The distance between the proximal edge of the VAM and anterior superior iliac spine (ASIS) was measured cranially while the distance between the distal edge of the VAM and the adductor tubercle (AT) was measured distally. The position and length of these edges were also measured. Subsequently, a 22G spinal needle was placed using the in-plane approach under the echoic line considered to represent the VAM and 0.1 mL of methylene blue was injected. Using anatomical dissection, the sartorius muscle was opened transversally and reclined to access the VAM. The exact position of the methylene blue was noted and the position and length of the proximal and distal edges measured.

Ultrasound study in volunteers

Following institutional ethics committee approval and informed written consent 13 volunteers, recruited from our medical staff (six males and seven females) with a mean age of 31 years (27–38) were included in this study. This composed of 26 thighs to be examined. None of the subjects reported any current or past medical history of pain in the lower limbs, previous surgery or trauma to the thighs. The BMI range of all of volunteers was within the normal limits (BMI < 25 kg/m²). In each subject, the subsartorial canal was examined with a 12–5 MHz linear probe (model iU22; Philips Medical Systems, Eindhoven, Netherlands) following the same protocol as described for the cadavers. The subject was placed in the supine position, the lower limb in a resting position. The visibility, position and dimensions of the image considered to represent the VAM were noted.

Statistical analysis

The data gathered for each group were compared with the results of the benchmark study from Tubbs [20] using Student's t-test. Results were expressed in mean and extreme values. P values < 0.05 were considered significant.

Results

Anatomical study

The VAM was clearly visible with US, with an average visibility score of 2. The average length of the VAM was 5.2 cm (range 3.5–8) with US and 7.6 cm (range 5–10.5) during dissection. The average distance from the proximal edge to the ASIS was 26.7 cm (range 25–30) with US and 27.1 cm (range 26–29.5) during dissection. The average distance from the distal edge to the AT was 12.8 cm (range 11–14) with US and 10.5 cm (range 8.512) during dissection. The distal half of the VAM was composed of thin and obliquely oriented fibers while the proximal-half fibers were much thicker and more transversely oriented. The methylene blue injection site was observed an average of 14.3 cm (range 12–16) proximal to the AT. Detailed data is provided in Table 1.

Table 1.

Anatomical data for four cadavers

	Ultrasound results	Anatomical dissection
VAM visibility score (0–2)	2	Not available
Puncture site (to AT (cm))	14.3 (12–16)	14.3 (12–16)
Methylene blue under VAM (%)	100%	100%
Length of VAM (cm)	5.2 (3.5–8)	7.6 (5–10.5)
Distance to ASIS (cm)	26.7 (25–30)	27.1 (26–29.5)
Distance to AT (cm)	12.8 (11–14)	10.5 (8.5–12)

VAM vastoadductor membrane; ASIS anterior superior iliac spine; AT adductor tubercle; VAM visibility score: 0-not visible; 1-identified with difficulty; 2-clearly identified

In all cases, the methylene blue was located deep to the VAM, thus confirming the echoic line visible on US to be this anatomical structure (Figs. 2 and 3). In two cases in two different cadavers, the VAM was barely separated proximally from the distal expansions of the fascia cruralis during dissection. In all cases, the SN was identified deep to the VAM along the anterolateral aspect of the femoral artery and was surrounded by methylene blue in all cases. Distally, the subsartorial branch of the nerve ran caudally, parallel to the adductor longus tendon. In two cases, collateral branches of the SN crossed the VAM to the skin of the medial aspect of the thigh, while in one case true anastomosis of one of these branches with the obturator nerve was observed.

Fig. 2.

Sonogram of an ultrasound-guided infiltration of the saphenous nerve in the subsartorial canal in a cadaveric subject. Before (A) and after (B) injection. A superficial femoral artery, VM vastus medialis muscle, AL adductor longus muscle, AM adductor magnus muscle, S sartorius muscle, arrows: vastoadductor membrane, asterisks: Methylene blue under the vastoadductor membrane

Fig. 3.

Methylene blue located under the VAM prior (A) and subsequent (B) to a longitudinal incision. The saphenous nerve appears to be stained with methylene blue, arrow: vastoadductor membrane, arrowhead: saphenous nerve

Ultrasound study in volunteers

The VAM was visible with US in all patients (100%), with an average visibility score of 1.6, including a score of 2 for 16 thighs (61.5%) and a score of 1 for the other ten (38.5%). The average VAM length was estimated at 5 cm (range 2.7–13). The average distance from the proximal edge to the ASIS was 25.5 cm (range 18–32.5) and average distance from the distal edge to the AT 13.7 cm (range 10.5–21) (Table 3).

Table 3.

Results of the ultrasound study in volunteers compared with the study by Tubbs

	Ultrasound results	Tubbs’ results	Comparison
Length of VAM (cm)	5 (2.7–13)	7.6 (5.5–15)	p < 0.001
Distance to ASIS (cm)	25.5 (18–32.5)	28 (20–32)	NS
Distance to AT (cm)	13.7 (10.5–21)	10 (7–15)	p < 0.001

VAM vastoadductor membrane; AT adductor tubercle; ASIS anterior superior iliac spine; NS no statistically significant differences

Discussion

Injection of the subsartorial canal is proposed when anesthesia of the medial knee and leg is required or for the treatment of saphenous nerve tunnel syndrome. Several studies have recommended the use of US to guide the procedure, and reported this technique to be more reliable than prior techniques that were based on skin surface clinical landmarks [13–19]. These studies concur that the saphenous nerve in the thigh is hardly targeted directly; due to its deep position and thin caliber as well as the absence of a relevant anatomical landmark for needle placement. In an anatomical study, Andersen has suggested that simple placement of the needle tip under the sartorius muscle may not be sufficient to reach the saphenous nerve as when injected, substances spread anterior to the superficial aspect of the VAM [21]. This may explain why in a study by Tsai assessing this clinical practice, non-guided injection of anesthetics in the subsartorial canal is effective in only 77% of cases [22]. Cowlishaw [23] considered it is important to achieve subsartorial spread of local anesthetic both superficial and deep to the VAM, to provide excellent analgesia for knee arthroplasty surgery. The rationale is to reach both the saphenous nerve and the other branches of the subsartorial plexus (branches from the saphenous nerve, from the medial cutaneous nerve of the thigh, the anterior section of the obturator nerve) which are situated superficial to the VAM.

Our study confirms that the VAM- delineated as a thin membrane demarcating a sub-compartment within the subsartorial canal- is easily visible with the US in patients with a normal BMI (< 25 kg/m²). VAM was identified in all specimens. It is derived from the medial intermuscular septum and connect the medial edge of the vastus medialis muscle to the lateral edge of the adductor magnus muscle,

Despite its good visibility with US, variations in the size and position of the VAM along the thigh need to be considered:

We determined the average distance between the VAM from the proximal edge to the ASIS as a solid and reproductible clinical landmark. We found no statistically significant difference between the US measurements in both volunteers and cadavers to those measured during dissection in the present study. In addition, these measurements were concordant with the study by Tubbs [20] when assessing the average distance (Tables 2 and 3). Other authors like Bendtsen [24] prefer to use other anatomical landmarks, such as the floor of the femoral triangle, which is a groove between the inguinal ligament and the sartorius and adductor longus muscles; which is theoretically visible on ultrasound. Bendtsen [24] considers that there is no anatomical gap between the apex of the femoral triangle and the VAM. We have not assessed this in our study.

Table 2.

Results of the anatomical study in cadavers compared with the study by Tubbs

	Anatomical dissection	Tubbs’ results	Comparison
Length of VAM (cm)	7.6 (5–10.5)	7.6 (5.5–15)	NS
Distance to ASIS (cm)	27.1 (26–29.5)	28 (20–32)	NS
Distance to AT (cm)	10.5 (8.5–12)	10 (7–15)	NS

	Ultrasound results	Tubbs’ results	Comparison
Length of VAM (cm)	5.2 (3.5–8)	7.6 (5.5–15)	p < 0.01
Distance to ASIS (cm)	26, 7 (25–30)	28 (20–32)	NS
Distance to AT (cm)	12.8 (11–14)	10 (7–15)	p < 0.001

VAM vastoadductor membrane; AT adductor tubercle; ASIS anterior superior iliac spine; NS no statistically significant differences

Secondly, the total length of the VAM was significantly shorter and the distance from its distal edge to the AT was significantly longer on US than observed during dissection, a finding observed in both the present study and that by Tubbs [20] (Tables 2 and 3). These discrepancies have been explained in the study by Tubbs as being attributed to the morphology of the VAM. The distal region of the VAM is composed of thin and obliquely oriented fibers while its proximal fibers are much thicker and more transversely oriented. This particular appearance of the VAM was also clearly visible on dissection in this study and may account for the poorer visibility of the distal edge with US, leading to both an artifactual increase in its distance to the AT as well as shortening of the VAM length.

Low visibility of the distal VAM was considered of doubtful clinical significance as it did not impair overall VAM visibility and needle placement when a saphenous nerve procedure is performed. Following needle placement, the spread of the injectate was facilitated by the use of a high-volume approach; defined as the injection of 10–30 mls in anesthesiology [13–25] and up to 10 mls (e.g. including 1 ml of steroid and 9 mls of lidocaine) in subsartorial canal tunnel syndrome [18].

This study has some limitations. First, VAM visualization and measurement reproducibility was not assessed as only one operator performed the ultrasound examinations in both groups. However, findings were concordant with the study by Tubbs. Although localizing the proximal and distal edges of the VAM is a subjective task with potential inter-observer variations, good VAM visibility with US was demonstrated by the exact position of the methylene blue during dissections in all the anatomical cases. Another limitation of the study is a validation of good VAM visibility with US only in subjects with a BMI < 25 kg/m², limiting validation of reproducibility in larger patient populations. This was due to recruiting volunteers in our study who were generally young with athletic physiques. Nevertheless, the VAM remains relatively a superficial structure and it is, therefore, possible to achieve similar results in larger patients, but further studies may be required to assess this. A third limitation is anatomical in nature; the exact position of branches directed to the vast medialis relative to the VAM, may explain unintentional motor block during RA, has not been assessed in the present work [26]. Last but not least, further studies may be necessary to clarify the minimum effective volume of local anesthetics and/or steroids administered under the VAM membrane to achieve technical success.

Conclusion

Our study demonstrates that the VAM can be clearly identified with the US in patients with a normal BMI (< 25 kg/m²), and used as a reliable anatomical landmark when performing injections of the subsartorial canal. This ultrasound-guided technique may be helpful for both regional anesthesia and the treatment of saphenous nerve tunnel syndrome.

Author contribution

All of authors contributed to the design of the research, to the analysis of the results and to the writing of the manuscript.

Funding

No external funding for this manuscript.

Availability of data and material

All data generated or analysed during this study are included in this published article.

Code availability

Not applicable.

Declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Ethics approval

Permission to dissect the cadaveric specimens was granted by our Anatomy department. 13 volunteers were recruted following institutional ethics committee approval and informed written consent.

Consent to participate

Consent to participate was obtained for every individual person’s data included in the study.

Consent for publication

Consent for publication was obtained for every individual person’s data included in the study.