Can ultrasound-guided radiofrequency ablation of genicular nerves of the knee, be performed without locating corresponding arterial pulsations—a cadaveric study

Abstract

Introduction Given the rising prevalence of knee osteoarthritis, radiofrequency ablation of genicular nerves (RFA) has emerged as a promising treatment option for knee pain. The knee has an extremely complex and variable innervation with nearly 13 genicular nerves described. The frequently ablated genicular nerves are the superomedial (SMGN), the superolateral (SLGN), and the inferomedial (IMGN) genicular nerves. Conventionally, under ultrasound guidance, these nerves are ablated near the corresponding arterial pulsations, but due to the rich vascular anastomosis around the knee joint, identifying the arteries corresponding to these constant genicular nerves can be tedious unless guided by some bony landmarks. In this study, we have evaluated whether it is possible to accurately target these three genicular nerves by just locating bony landmarks under ultrasound in human cadaveric knee specimens. Methods Fifteen formalin-fixed cadaveric knee specimens were studied. SMGN was targeted 1 cm anterior to the adductor tubercle in the axial view. For SLGN, in the coronal view, the junction of the lateral femoral condyle and shaft was identified, and at the same level in the axial view, the crest between the lateral and posterior femoral cortex was targeted. For IMGN in the coronal view, the midpoint between the most prominent part of the medial tibial condyle and the insertion of the deep fibers of the medial collateral ligament was marked. The medial end of the medial tibial cortex was then targeted at the same level in the axial view. The needle was inserted from anterior to posterior, with an in-plane approach for all nerves. Eosin, 2% W/V, in 0.1 ml was injected. Microdissection was done while keeping the needle in situ. Staining of the nerve was considered a positive outcome, and the percentage was calculated. The nerve-to-needle distance was measured, and the mean with an interquartile range was calculated. Result The accuracies of ultrasound-guided bony landmarks of SMGN, SLGN, and IMGN were 100% in terms of staining, with average nerve-to-needle distances of 1.67, 3.2, and 1.8 mm respectively. Conclusion It is with 100% accuracy, that we can perform RFA of SMGN, SLGN, and IMGN under ultrasound guidance, by locating the aforementioned bony landmarks.