Skeletal Radiology

, Volume 42, Issue 4, pp 579–586

High-resolution magnetic resonance-guided posterior femoral cutaneous nerve blocks

Authors

    • Russell H. Morgan Department of Radiology and Radiological ScienceThe Johns Hopkins University School of Medicine
    • Musculoskeletal Radiology, Russell H. Morgan Department of Radiology and Radiological ScienceJohns Hopkins University School of Medicine
  • Cary Bizzell
    • Russell H. Morgan Department of Radiology and Radiological ScienceThe Johns Hopkins University School of Medicine
  • Sudhir Kathuria
    • Russell H. Morgan Department of Radiology and Radiological ScienceThe Johns Hopkins University School of Medicine
  • Aaron J. Flammang
    • Siemens Corporate ResearchCenter for Applied Medical Imaging
  • Eric H. Williams
    • Department of Plastic, Reconstructive, and Maxillofacial SurgeryThe Johns Hopkins University School of Medicine
    • Dellon Institute for Peripheral Nerve Surgery
  • Allan J. Belzberg
    • Department of NeurosurgeryThe Johns Hopkins University School of Medicine
  • John A. Carrino
    • Russell H. Morgan Department of Radiology and Radiological ScienceThe Johns Hopkins University School of Medicine
  • Avneesh Chhabra
    • Russell H. Morgan Department of Radiology and Radiological ScienceThe Johns Hopkins University School of Medicine
Technical Report

DOI: 10.1007/s00256-012-1553-8

Cite this article as:
Fritz, J., Bizzell, C., Kathuria, S. et al. Skeletal Radiol (2013) 42: 579. doi:10.1007/s00256-012-1553-8

Abstract

Objective

To assess the feasibility, technical success, and effectiveness of high-resolution magnetic resonance (MR)-guided posterior femoral cutaneous nerve (PFCN) blocks.

Materials and methods

A retrospective analysis of 12 posterior femoral cutaneous nerve blocks in 8 patients [6 (75 %) female, 2 (25 %) male; mean age, 47 years; range, 42–84 years] with chronic perineal pain suggesting PFCN neuropathy was performed. Procedures were performed with a clinical wide-bore 1.5-T MR imaging system. High-resolution MR imaging was utilized for visualization and targeting of the PFCN. Commercially available, MR-compatible 20-G needles were used for drug delivery. Variables assessed were technical success (defined as injectant surrounding the targeted PFCN on post-intervention MR images) effectiveness, (defined as post-interventional regional anesthesia of the target area innervation downstream from the posterior femoral cutaneous nerve block), rate of complications, and length of procedure time.

Results

MR-guided PFCN injections were technically successful in 12/12 cases (100 %) with uniform perineural distribution of the injectant. All blocks were effective and resulted in post-interventional regional anesthesia of the expected areas (12/12, 100 %). No complications occurred during the procedure or during follow-up. The average total procedure time was 45 min (30–70) min.

Conclusions

Our initial results demonstrate that this technique of selective MR-guided PFCN blocks is feasible and suggest high technical success and effectiveness. Larger studies are needed to confirm our initial results.

Introduction

The posterior femoral cutaneous nerve (PFCN) is a sensory nerve that is formed by the sacral plexus and carries only sensory fibers (Fig. 1) [1]. The perineal branch (PBPFCN) innervates the lateral perineum, proximal medial thigh, posterolateral aspect of the scrotum/labium majus, and a portion of the penis/clitoris; the inferior cluneal branches provide innervation to the inferior buttock area, and the descending cutaneous portion innervates the posterior aspect of the thigh and the popliteal area [14].
https://static-content.springer.com/image/art%3A10.1007%2Fs00256-012-1553-8/MediaObjects/256_2012_1553_Fig1_HTML.gif
Fig. 1

Illustration demonstrating the anatomy of the posterior femoral cutaneous nerve (with permission, [1]), which carries cutaneous sensory fibers and innervates the gluteal region, the perineum, and the back of the thigh and leg

PFCN neuropathy represents an important differential diagnosis of chronic perineal pain, clunealgia, and posterior thigh pain [212]. Etiologies include compression, entrapment, repetitive trauma from cycling, impact injury, and injection injury; however, in many cases, no etiology may be identified [2, 58, 10].

The manifestations of PFCN neuropathy vary, which is thought to be influenced by the level and location of the abnormality along the nerve and overlap with other nerves [3, 4]. PFCN neuropathy selectively affecting the PBPFCN is a major differential diagnosis of pudendal neuralgia owing to substantial overlap of their perineal areas of innervation [3]. Differentiation of those two entities is of critical importance for appropriate management, as conservative and surgical treatment options differ, and misdiagnosis may lead to inappropriate management [2, 3, 13]. Neuropathy of the cluneal branches of the PFCN may cause pain and paresthesia referred to the inferior lateral buttock area (clunealgia) [4]. Finally, involvement of the more distal, descending cutaneous PFCN branches may present as pain and paresthesia of the posterior thigh [1, 14].

In cases of unsatisfactory results of conservative therapy or failed selective regional blocks of other nerves (e.g., pudendal nerve), selective diagnostic nerve blocks of the PFCN can be helpful to assess the nerve’s contribution to the patient’s symptoms and to identify potential surgical targets [4, 14]. However, in order to ensure the validity of the pain response of PCFN blocks, the general principles of accurate perineural drug delivery and objective assessment of the distribution of the injectant are fundamental prerequisites.

Without adequate image guidance, accurate needle positioning for PFCN blocks is challenging. Anatomic landmark-based, non-image-guided techniques have been described [1417]; however, because of the lack of visualization of the PFCN and injectant, no objective assessment of the validity would be possible with those techniques. A computed tomography (CT)-guided technical report of PFCN injection has been published; however, the low contrast-to-noise ratio of CT renders the reliable identification of the PFCN difficult [4]. High-resolution magnetic resonance imaging (MRI), in contrast, reliably displays the course of the PFCN [1821] (Fig. 2), which may then be used for accurate targeting with an MR-compatible injection needle for selective perineural drug delivery and visualization of the spread of the injectant. To the best of our knowledge, no prior reports have scientifically assessed the technical success and effectiveness of MR-guided PFCN blocks.
https://static-content.springer.com/image/art%3A10.1007%2Fs00256-012-1553-8/MediaObjects/256_2012_1553_Fig2_HTML.gif
Fig. 2

Three Tesla, T1-weighted turbo spin-echo magnetic resonance images (repetition time, 600; echo time, 20; slice thickness, 4 mm; echo train length, 6; field of view, 360 × 253 mm; base resolution, 384; phase resolution, 70 %) demonstrate the anatomy of the posterior femoral cutaneous nerve (PFCN). The PFCN travels with the sciatic and pudendal nerve (circle, a) and exits through the infrapiriformis foramen (circle, b). The PFCN and sciatic nerve descend laterally under the gluteus maximus muscle (circle, c), whereas the pudendal nerve descends medially into the pudendal canal (arrow, c). At the inferior edge of the gluteus maximus muscle, the PFCN gives rise to the inferior cluneal branches (not seen). Approximately 4 cm (range 3–5.5 cm) inferior to the ischial tuberosity, the perineal branches arise (not seen). Then, the PFCN separates from the sciatic nerve (arrow, d) and travels posteriorly down the leg under the fascia lata (circle, d). Note contrast material in the urinary bladder, which is inconsequential for the demonstration of the PFCN

Therefore, the purpose of this study was to assess the feasibility, technical success, and effectiveness of high-resolution MR-guided PFCN blocks.

Materials and methods

This study was approved by our institutional review board. The study was Health Insurance Portability and Accountability Act-compliant and the informed consent requirement was waived.

Patients

A consecutive cohort of patients was identified retrospectively by searching our patient information systems for MR-guided perineural injections of the PFCN. Search criteria included the ‘PFCN’, ‘Cluneal branch’, and ‘Perineal branch’ as the target, the specific descriptor for the 1.5-T MRI system, and the procedure code for percutaneous drug delivery.

The search resulted in 12 PFCN procedures, which were performed in 8 patients [6 (75 %) female, 2 (25 %) male; mean age, 47 years; range, 42–84 years]. The patients were referred for diagnostic blocks of the PFCN because of chronic pain and paresthesia in the perineal area and negative previous pudendal nerve blocks (n = 4), in the inferior lateral buttocks area (n = 2), and in the posterior thigh (n = 2), which was refractory to conservative therapy.

MR-guided injection

In patients with symptoms in the perineal area, the PFCN was targeted 2 cm below the ischial tuberosity, proximal to the PBPFCN (Fig. 3) [3]. In patients with symptoms affecting the inferior lateral buttock area, the cluneal nerves were targeted under the gluteus maximus muscle (Fig. 4) [4]. In patients with symptoms in the posterior thighs, the distal portion of the PFCN located between the gluteus maximus and the long head of the biceps femoris was targeted (Fig. 5).
https://static-content.springer.com/image/art%3A10.1007%2Fs00256-012-1553-8/MediaObjects/256_2012_1553_Fig3_HTML.gif
Fig. 3

Magnetic resonance (MR)-guided diagnostic block of the right posterior femoral cutaneous nerve (PFCN) proximal to the perineal branch in a patient with chronic perineal pain. a T1-weighted axial MR image 2 cm below the ischial tuberosity demonstrating the PFCN (arrow) posterior to the sciatic nerve. b Intermediate-weighted axial MR image following needle placement demonstrates the tip of the injection needle (arrow) subjacent to the PCFN. c Axial MR fluoroscopic gradient echo FLASH 2D sequence demonstrating monitoring of the hyperintense injectant (arrows). d Axial T1-weighted MR image with spectral fat saturation demonstrates the gadolinium-enhanced injectant (white arrow) surrounding the PFCN (black arrow) consistent with a technically successful block. The block resulted in anesthesia of the perineal area indicative of a clinically effective block. e Coronal T1-weighted MR image with spectral fat saturation demonstrates the gadolinium-enhanced injectant (white arrow), which is centered 2 cm below the ischial tuberosity (black arrow) with approximately 2 cm cranial and caudal extension, and resultant block of the perineal branch of the PFCN. Localized delivery at this level avoids confounding spread to the ischial spine and pudendal nerve in Alcock’s canal (gray arrow)

https://static-content.springer.com/image/art%3A10.1007%2Fs00256-012-1553-8/MediaObjects/256_2012_1553_Fig4_HTML.gif
Fig. 4

Magnetic resonance (MR)-guided diagnostic block of the left inferior cluneal nerves in a patient with chronic inferior buttock pain and paresthesia. a T1-weighted axial MR image above the ischial tuberosity demonstrates the PFCN (arrow) proximal to the cluneal nerves’ origins. b Intermediate-weighted axial MR image following needle placement demonstrates the tip of the injection needle (black arrow) adjacent to the PCFN (white arrow). c Coronal T2-weighted MR image demonstrates the injectant (white arrow) surrounding the PFCN branches consistent with a technically successful block. The block resulted in anesthesia the inferior buttock region indicative of a clinically effective block

https://static-content.springer.com/image/art%3A10.1007%2Fs00256-012-1553-8/MediaObjects/256_2012_1553_Fig5_HTML.gif
Fig. 5

Magnetic resonance (MR)-guided diagnostic block of the left distal portion of the posterior femoral cutaneous nerve (PFCN) in a patient with chronic posterior thigh pain and paresthesia. a T1-weighted axial MR image demonstrates the PFCN (arrow) posteriorly under the fascia lata. b Intermediate-weighted axial MR image following needle placement demonstrates the tip of the injection needle (black arrow) subjacent to the PCFN (white arrow). c Axial T2-weighted MR image with fat saturation demonstrates the injectant (white arrow) surrounding the PFCN, consistent with a technically successful block. The block resulted in anesthesia of the posterior thigh indicative of a clinically effective block

In one patient with symptoms in the perineal area (n = 1) and in one patient with symptoms in the inferior lateral buttock area (n = 1), two diagnostic blocks of the same PFCN region were performed a week apart in order to assess reproducibility (confirmatory diagnostic blocks). In four patients, one diagnostic block was performed, and in two patients, bilateral diagnostic blocks were performed in a single session.

All procedures were performed using a clinical, wide-bore 1.5-T MR imaging system (MAGNETOM Espree; Siemens Healthcare, Erlangen, Germany) with patients in prone position and the table landmark centered at the inferior buttocks area. High-resolution T1-weighted MR images [turbo spin echo; repetition time (TR), 691; echo time (TE), 20; slice thickness (SL), 4 mm; echo train length (ETL), 5; field of view (FOV), 360 × 253 mm; base resolution (BR), 384; phase resolution (PR), 70 %; receiver bandwidth (BW), 102 Hz, acquisition time (TA), 3.6 min] were obtained before the interventional phase of the procedure using a multi-channel body coil (Siemens Healthcare). The course of the PFCN was mapped from proximal to where it becomes subcutaneous and the needle path was planned on the work station. The coil was then exchanged with a flexible loop-shaped radio frequency coil with a diameter of 19 cm (Siemens Healthcare). From this point on, all actions were performed inside the bore without additional table movements during the procedure. The skin entry point was determined by using a water-filled syringe as a pointing device at the level of the selected injection site, which was displayed by a continuously acquired single-slice T1-weighted spoiled gradient-echo MRI sequence (MR fluoroscopy; TR, 9.3; TE, 3.5; SL, 5 mm; FOV, 256 × 224 mm; BR, 256; PR, 56 %; BW, 180 Hz; TA, 0.8 s/frame) [22]. A MR-compatible fiducial marker was placed to mark the determined needle entry site. The interventional field was then prepped and draped in standard fashion. Conscious sedation was initiated and vital signs were monitored during the procedure according to our institutional policy. Superficial local anesthesia was employed using 1 % lidocaine. A 20-G needle of 10- or 15-cm length (MReye, G11583; Cook Medical, Bloomington, IN, USA) was then navigated interactively into the immediate vicinity of the PFCN. Axial turbo spin-echo MR images (TR, 1200; TE, 12; SL, 4 mm; ETL, 17; FOV, 256 × 224 mm; BR, 320; PR, 100 %; BW, 252 Hz; TA, 12 s) were acquired for visual assessment of the adequacy of the proximity of the needle tip to the nerve. In all patients and locations, a total amount of 4 ml was then slowly injected around the PFCN consisting of 1 ml of 1 % preservative-free lidocaine, 1 ml of 0.5 % bupivacaine, and 1 ml of non- particulate dexamethasone (10 mg/ml). In 5/12 cases (41.7 %), a small amount of gadolinium-diethylenetriaminepentacetate (DTPA) contrast medium (Magnevist, Bayer HealthCare Pharmaceuticals, Montville, NJ, USA) with a dilution factor of 1:300 was added, whereas in 7/12 cases (58.3 %) no gadolinium-based contrast medium was added. The injection was either monitored continuously with MR fluoroscopy (Fig. 3) or intermittently with a fast STIR sequence (TR, 1200; TE, 29; SL, 5; ETL, 5; FOV, 256 × 224 mm; BR, 192; PR, 45 %; BW, 542 Hz; TA, 32 s). The final distribution of the delivered injectant was visualized using an axial fat-saturated T1-weighted turbo spin-echo MR sequence (TR 500, TE 20, SL 4 mm, ETL, 6; FOV 200 × 200 mm, BR, 384, PR, 50 %; BW, 102 Hz; TA, 5.1 min) or an isotropic three-dimensional T2-weighted MRI data set with or without fat saturation [sampling perfection with application optimized contrasts using different flip angle evolutions (SPACE) sequence; TR, 1500; TE, 138; ETL, 8; SL, 1 mm; FOV, 300 × 207 mm; BR, 384; PR, 74 %; BW, 751 Hz; TA, 7 min] (Figs. 3, 4, and 5).

Assessment of outcome

Technical success was defined as MRI visualization of the injectant around the targeted portion of the PFCN. Assessments were made, in consensus, on the post-interventional fat-saturated T1- or T2-weighted MR images by two board-certified radiologists. Images were reviewed using a picture archiving and communication system diagnostic workstation (Leonardo; Siemens Healthcare).

The effectiveness of the PFCN blocks was assessed by clinical evaluation obtained 30–60 min after the injection. An effective block was defined as post-interventional anesthesia of the target area innervation of the PCFN depending on the level and side of the block using the tip of a blunt paper clip tip with comparison to the contralateral side [3]. In blocks above the level of the PBPFCN, an effective block was defined as anesthesia in the lateral perineal area and the posterior thigh. In cases of blocks involving the cluneal nerves, an effective block was defined as anesthesia in the inferior lateral buttocks area [4, 14]. In blocks of the distal portion of the PCFN, an effective block was defined as anesthesia in the posterior thigh area [14].

Major complications were defined in accordance with the Society of Interventional Radiology guidelines [23], as complications that result in admission to the hospital for therapy, an unplanned increase in the level of care (worsening pain, hemorrhage, or infection), prolonged hospitalization, permanent adverse sequelae, or death. Assessment was performed on a procedure by procedure basis. Data were obtained by chart review.

The total length of time for a procedure (total procedure time) was defined as the time point from acquisition of the first MR image to the last MR image, including preprocedural MRI, nerve block, and post-procedural MRI.

Results

The course of the PFCN was mapped successfully in all patients on the high-resolution axial T1 images. MR-guided injections allowed accurate perineural drug delivery in all cases (12/12, 100 %). Post-interventional MR images demonstrated the injectant completely surrounding the PFCN at the targeted level in all cases.

Clinical evaluation demonstrated concordant anesthesia of the expected areas following blocks in all cases (12/12, 100 %). Specifically, patients described anesthesia of the perineal area following blocks above the PBPFCN (n = 7), anesthesia including the inferior lateral buttocks area in cases where the cluneal nerves were targeted (n = 2), and isolated anesthesia of the posterior thigh in cases where distal PCFN blocks were performed (n = 3). The results were also reproducible in the two patients who underwent confirmatory blocks.

No complications occurred during the procedure or during follow-up. All patients were discharged the same day, after 60 min of observation. No soft tissue abnormalities (i.e., internal bleeding or cutaneous bruising) occurred.

The median total procedure time for a single nerve block was 45 min (range, 30–70 min).

Discussion

Our initial results demonstrate that this technique of high-resolution MR-guided PFCN blocks at 1.5 T is feasible and affords high technical success and effectiveness. The absence of complications suggests a favorable safety profile.

High-resolution MR imaging guidance is advantageous because of the direct visualization of the course of the PFCN. Once the PCFN course is mapped on initial MR images, an appropriate needle path can be planned for accurate targeting and needle placement. Further, needle placement can be performed under real-time low-resolution MR fluoroscopy imaging for interactive needle control. Visualization of the needle allows accurate placement of the needle tip into the immediate vicinity of this small-caliber nerve and its tiny branches, thereby maximizing the validity of perineural injections, minimizing the risk of nerve injury, and limiting the volume needed for diagnostic blocks. Intermittently-acquired static MR images with higher signal and contrast to noise ratios can be used to confirm the final needle tip position prior to drug delivery. MRI at 1.5 T has been shown to display the needle tip within an error margin of 1 mm [24], which was found to be sufficiently accurate for all PFCN injection procedures in this study. In our study, all injections resulted in the expected perineural distribution of the injectant, and no signs or symptoms of PFCN or other neural injury occurred. MR images obtained following the injection directly visualized the injectant surrounding the PFCN and its smaller branches. This provided the operator an additional means of quality control in order to judge the technical adequacy of the block and its validity.

Another aspect of achieving an accurate block is to perform a selective injection and to minimize the spread of the injectant to the adjacent nerves, which may otherwise confound the selectiveness of the block [25]. This is of importance in patients with perineal pain, because symptoms may be related to pudendal neuropathy or the PBPFCN [3, 13]. In order to achieve an effective PFCN block, the injectant should be delivered sufficiently inferior to the ischial spine and the pudendal canal in order to avoid spread and anesthesia of this nerve, but proximal to the origin of the PCPFCN (Figs. 1 and 3). Tubbs et al. found that the PBPFCN courses 3 to 5.5 cm inferior to the termination of the sacrotuberous ligament with an average distance of 4 cm [3]. MRI guidance may be the most valuable approach for this block, as shown in our study, because of the direct visualization of the pudendal nerve, the course of the PFCN, and potentially occurring inadvertent spread of the injectant to the pudendal nerve, which was not observed in any of our cases (Figs. 2 and 3).

Although we did not compare MR-guided PFCN blocks with other techniques, it may be speculated that MRI guidance allows for improved selectivity compared with previously reported injection techniques, which relied primarily on anatomic landmark-based palpation techniques and CT guidance [4, 1417]. With such techniques, PFCN anesthesia might be achieved inconsistently and the validity of blocks would be uncertain because the spread of the injectant to adjacent nerves, such as the pudendal nerve, may go undetected. Further studies are desirable to compare MRI guidance with the described techniques.

We utilized T1-weighted or T2-weighted MRI for visualization of the location and spread of the injectant. Detection of the gadolinium-enhanced injectants with use of T1-weighted sequence with spectral fat saturation was based on the T1 shortening effect of the gadolinium-based contrast agent [22]. Injectants were also visualized with T2-weighted sequences, based on the intrinsically long T2 of the water-based injectant. Both techniques were equally sufficient for the localization and detection of the injectant. Gadolinium-enhanced injectants are useful for real-time T1-weighted MR fluoroscopic monitoring of the injection and detection of perineural spread (Fig. 3). The safety of perineural injection of gadolinium-based contrast material and the combination of anesthetic and steroids has been shown previously [26, 27]. T2-weighted visualization may be preferred for static visualization because it obviates the addition of gadolinium-based contrast. Either approach can be advantageous in the treatment of patients with hypersensitivity to iodine, for whom iodine-based contrast agents such as those used in CT and X-ray fluoroscopy are contraindicated [26, 27].

In contrast to CT guidance, MRI guidance does not result in exposure to ionizing radiation of patients and operators, and avoids radiation-associated health risks. Although the radiation dose of CT can be reduced with use of intermittent technique, new iterative algorithms, and low-energy settings, owing to the vicinity of the reproductive organs to the X-ray beam during PFCN injections, interventional MRI guidance complies favorably with the as low as reasonably achievable (ALARA) practice mandate and should be considered as the preferred approach.

The safety profile of this small group of patients was favorable, with no complications occurring during the procedure or during follow-up. However, larger studies are required to fully establish the safety profile of this technique.

No exclusive material was used to perform MR-guided PFCN blocks. Procedures were performed on a commercially available, clinical wide-bore MR imaging system, routine coils, and commercially available MR-compatible needles. The short bore magnet design allowed for convenient access to the isocenter, which was helpful in determining the skin entry site, needle placement, and injection inside the bore without the requirement of additional table movements. The procedure may be carried out in a similar fashion in magnets with a longer bore, depending on the distance to the isocenter. The increasing availability of wide-bore high field MRI systems [28] that allow improved patient access may increase the number of sites, where this procedure can be performed.

The mean procedure time of 45 min appears acceptable for this initial study. The procedure time seems longer when compared with CT-guided techniques, owing to the inherently longer time required for MRI acquisition. However, streamlining of the workflow and optimization of the MR protocol may reduce the length of time of the procedure. Also, we believe that the discussed advantages of MR guidance outweigh the potential differences in the procedure time. Newer augmented reality systems show promise in further decreasing the length of time for such procedures [29, 30].

Despite the apparent advantages of MR-guided PFCN blocks, the cost related to high-field MRI guidance and MR-compatible needles needs to be considered. However, this may be offset by more effective blocks and potentially improved patient management.

Our study had some limitations. The results of this study should be considered preliminary owing to the small sample size. The retrospective characteristics of our study may have introduced a selection bias. The length of the procedure time should be considered preliminary, because we did not differentiate between cases in which a single block was performed and cases in which blocks were performed bilaterally, which possibly resulted in some differences. The primary goal of this study was to assess the technical success and the effectiveness of the blocks, and we did not assess the contribution of the blocks to the entirety of patients’ work-up and management. This is because the determination and isolation of pain-generating structures is a complex process. It usually consists of testing of multiple putative anatomic structures, which may involve other pelvic abnormalities, and even the lower spine. Additional larger prospective studies would, inevitably, be required to fully assess the clinical value of MR-guided PFCN blocks. A bias may have been introduced through assessment of the technical success in consensus mode and by observers who performed some of the procedures. The post-interventional clinical assessment was also performed by the same team that performed the block.

Conclusions

Our initial results demonstrate that selective high-resolution MR-guided PFCN blocks are feasible with a suggested high technical success and effectiveness. The absence of complications suggests a favorable safety profile.

Copyright information

© ISS 2012