Comparison of radiofrequency ablation of genicular nerve with genicular nerve block in patients with osteoarthritis of the knee

Abstract

Background and Aim:

Genicular nerve block has recently become a promising treatment option along with radiofrequency ablation in the management of osteoarthritis of the knee.

Material and Methods:

Sixty patients of age above 50 years with a pain pattern consistent with osteoarthritis of the knee, with their X-ray knee joint findings corresponding to their clinical symptoms, and with a failure to respond to conservative treatment were included in this study. The patients were divided into two groups of 30 each as follows: group I: patients were administered fluoroscope-guided radiofrequency ablation of genicular nerve using RF electrode set to a temperature of 60℃ for 120 sec and group II: patients were administered fluoroscope-guided genicular nerve block using 9 ml drug solution comprising 8 ml of 0.25% bupivacaine plus 1 ml (40 mg) of methylprednisolone.

Results:

Both techniques of knee injection were effective and provided good pain relief to the patients with symptomatic osteoarthritis of the knee. Pain scores were clinically less in group I compared to group II at all time intervals of the study period. However, they were statistically significant between the two groups at 2, 3, 6, and 12 months (P < 0.05) with lesser Numerical Rating Scale in group I compared to group II. Western Ontario and McMaster Universities osteoarthritis index was clinically better in group I at all time intervals of the study period, but was statistically significant at 2 weeks, 1 month, and 12 months after the procedure (P < 0.05).

Conclusions:

Fluoroscope-guided radiofrequency ablation of genicular nerve is better than fluoroscope-guided genicular nerve block in terms of improvement in pain score, physical status, and patient satisfaction.

Introduction

Osteoarthritis (OA) is common form of arthritis which leads to long-term pain and disability. It exerts a significant burden on the individual and the community through reduction in quality of life, diminished employment capacity, and an increase in healthcare costs. The incidence of OA is rising because of the increased longevity of life and obesity.[1,2] Interventional modalities available are intra-articular knee injection of local anesthetic, steroid, hyaluronic acid, saline, medical ozone, growth factor or platelet-rich plasma, saline lavage, saphenous nerve block, genicular nerve block, arthroscopic partial meniscectomy, radiofrequency ablation, etc.[3,4,5]

Genicular nerve block (GNB) is a promising treatment option for the management of OA knee pain. This procedure provides pain relief by inhibiting the nerve fibers innervating the knee joint. The superomedial, inferomedial, and superolateral genicular nerves are the main innervating articular branches for the knee joint. Genicular nerve injections have been performed under fluoroscopic guidance, in which needle placement is successfully done with reference to bony landmarks.[6,7]

Radiofrequency ablation (RFA) acts by stopping nociceptive pain input from the periphery to the central nervous system through A-delta and C-fibers without destroying the motor or sensory fibers (A-beta). The postulated mechanism of action for RFA is heat generation resulting in thermocoagulation and localized neuronal tissue destruction. RFA is used in treating conditions ranging from sympathetic derived nerve pain to chronic knee pain. This area is stimulated for identification of nerve position and to make sure that no motor nerves are activated as evidenced by absence of fasciculations. The RF electrode tip heats up the targeted local tissue within few millimeters to a temperature typically greater than 47°C (ranging from 70 to 90℃) for 120–150 sec, generated through an electromagnetic field with a frequency of 250 Hz. RFA can be offered as part of a multimodal approach to pain control in patients with OA of the knee.[8,9,10,11]

We conducted this prospective study with the primary objective of comparing the efficacy of RFA of genicular nerve with genicular nerve block in OA of the knee in terms of producing pain relief and improvement in disability. The secondary objectives were ease of administration of the block and side effects, if any.

Material and Methods

The present prospective, randomized study was conducted in the pain management center of a post-graduate institute after obtaining approval from the institutional ethical committee and was registered in the Clinical Trials Registry-India vide registration number CTRI/2020/05/025118. Sixty patients of either sex and age above 50 years attending the pain clinic and fulfilling the following three criteria were included in the study: history, physical examination, and pain pattern consistent with OA of the knee; X-ray knee joint findings corresponding to the patient’s clinical symptoms; and failure to respond to 6 weeks of conservative treatment and not willing for surgery or waiting for surgery. Patients with known contraindications to genicular nerve block or intra-articular knee injection; history of adverse reactions to local anesthetics or steroids; and previous history of genicular nerve block or intra-articular injection of the knee were excluded from the study.

Detailed clinical history was taken, and all patients were subjected to examination in the pain clinic. After obtaining informed and written consent, the procedure was explained in detail to patients. Numerical Rating Scale (NRS; 0–10) was explained to each patient for assessment of pain before performing the procedure. Patients were placed in supine position with pillow under the knee to flex the leg at the knee joint. Strict aseptic precautions were followed. Lignocaine (1%) was infiltrated subcutaneously.

Patients were randomly divided into two groups of 30 each using computer-generated randomized number table. In group I (n = 30), a 100-mm, 22G with 10-mm bent-tip RFA needle was positioned at the superolateral, superomedial, and inferomedial periosteal areas connecting the femur and tibia shafts to their respective epicondyle under fluoroscopic guidance after stimulating threshold of 0.15 V at 50 Hz before nerve ablation to optimize needle positioning. With sensory capture confirmed at 0.15 V and absence of motor stimulation, the electrode target temperature was set to 60℃ for 120 sec. In group II (n = 30), a 23-G, 10-cm spinal needle was advanced under fluoroscopic guidance and the superomedial genicular nerve (SMGN) was blocked at the junction of the femoral medial epicondyle and shaft of the femur, the superolateral genicular nerve (SLGN) was blocked at the junction of the femoral lateral epicondyle and shaft of the femur, and the inferomedial genicular nerve (IMGN) was blocked at the junction of the tibial medial epicondyle and shaft of the tibia. Three milliliters of drug solution was injected at each of the three sites (total 9 ml drug solution comprising 8 ml of 0.25% bupivacaine plus 1 ml [40 mg] of methylprednisolone). After the procedure, a sterile bandage was applied. No more than three injections/RFA was performed during the 12-month study period. The time interval between the two procedures was not less than 1 month [Figure 1].

Figure 1.

Fluoroscopic-guided genicular nerve block: (a) superomedial branch, (b) superolateral branch, (c) inferomedial branch, and (d) lateral fluoroscopic view and radiofrequency ablation

A sample size of 30 per group was calculated to achieve a power of 85% to show a difference of 9.6 units in the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) and 2 units change in NRS with 5% type I error. A difference of 5 units in the WOMAC index and 2 units change in NRS was found to be clinically relevant in previous studies, and was also used for sample size calculation.[12]

Pain relief, patient satisfaction, and improvement in disability were the primary outcomes of the study. The secondary outcomes were ease of administration of block, side effects, and complications related to the procedure.

Pain was assessed using NRS (0–10). Patients were asked to sit on a chair, stand, and walk before rating their pain. NRS was measured and recorded half an hour before the procedure and at half an hour, 2 weeks, 1 month, 2 months, 3 months, 6 months, and 12 months after the procedure.

Patient satisfaction was assessed 2 weeks, 1 month, 2 months, 3 months, 6 months, and 12 months after the procedure on a 4-point scale as follows: Excellent: when the pain is completely resolved or diminishes by 75% or more, Good: when diminution of pain is 50%–74%, Fair: when diminution of pain is 25%–49%, and Poor: when diminution of pain is less than 25% or there is an increase in pain.

The WOMAC index was used to assess patients with OA of the knee using 24 parameters. The WOMAC index was calculated half an hour before the procedure and at 2 weeks, 1 month, 2 months, 3 months, 6 months, and 12 months after the procedure.

Pain during injection was recorded to determine the ease of administration of the technique. All injections were fluoroscope guided, and the needle was advanced under fluoroscopic guidance. Pain was assessed on a 4-point scale during administration of the injectate: 1 for no pain, 2 for mild pain, 3 for moderate pain, and 4 for severe pain. Side effects and complications like pain during administration of drug solution, pain at the injection site, swelling, and fever, if any, were recorded.

The patients were followed for 12 months after the initial procedure to determine if further injections were required. Repeat injections were carried out using the same approach as the initial procedure, if pain relief was not adequate (NRS >4). The time interval between the two consecutive injections was recorded.

Statistical Package for the Social Sciences (SPSS) version 20 (International Business Machine SPSS Statistics Inc., Chicago, IL, USA) was used for statistical analysis. Unpaired t-test (for quantitative data to compare two independent groups) and Friedman test (for quantitative data within two groups) were used for quantitative data comparison of all clinical indicators. Chi-squared test was used to compare the qualitative data. Results were considered to be statistically significant if P value was ≤ 0.05.

Results

The two groups were comparable in age, weight, and gender distribution [Table 1].

Table 1.

Distribution of age, gender, and weight in the two groups

Parameter	Group I FGRFAGN (n=30)	Group II FGGNB (n=30)	P
Age (in years) Mean±SD	58.86±6.29	57.63±4.29	0.40
Weight (in kg) Mean±SD	66.45±11.47	67.40±6.53	0.7
Male-to-female ratio	22:8 (73%:27%)	21:9 (70%:30%)	0.83

FGGNB=fluoroscopic-guided genicular nerve block, FGRFAGN=fluoroscopic-guided radiofrequency ablation of genicular nerve, SD=standard deviation

The variation in pain score at different time intervals when compared to pain score before injection was clinically and statistically significant (P < 0.05). In group I, the mean pain score (NRS score) half an hour before injection was 7.95 ± 0.95, which decreased to 1.73 ± 1.03 half an hour after injection. The pain score was 1.64 ± 0.95, 1.77 ± 1.11, 2.05 ± 1.09, 2.32 ± 1.24, 2.34 ± 1.26, and 2.40 ± 1.28 at 2 weeks, 1 month, 2 months, 3 months, 6 months, and 12 months after injection, respectively. In group II, the mean pain score (NRS score) half an hour before injection was 8.13 ± 0.73, which decreased to 2.3 ± 0.91 half an hour after injection. The pain score was 2 ± 1.14, 2.17 ± 1.55, 2.97 ± 1.34, 3.24 ± 1.42, 3.4 ± 1.52, and 3.6 ± 1.62 at 2 weeks, 1 month, 2 months, 3 months, 6 months, and 12 months after injection, respectively. Pain scores were clinically less in group I compared to group II at all time intervals of the study period. However, they were statistically significant between the two groups at 2, 3, 6, and 12 months (P < 0.05), with lesser NRS in group I compared to group II [Figure 2].

Figure 2.

Pain score (NRS) at different time intervals in the two groups. NRS = Numerical Rating Scale

The results showed clinically better patient satisfaction in group I compared to group II and the difference was statistically significant at half an hour, 2 months, 3 months, 6 months, and 12 months of the study period in group I (P < 0.05).

WOMAC score was clinically better in group I than group II at all time intervals throughout the study period. In group I, the mean WOMAC index half an hour before injection was 57.38 ± 14.16, which decreased to 14.89 ± 9.14 at 2 weeks after injection. WOMAC index was 14.89 ± 9.14, 17.20 ± 10.29, 19.38 ± 12.67, 19.42 ± 12.70, and 19.68 ± 12.82 at 1 month, 2 months, 3 months, 6 months, and 12 months after injection, respectively. In group II, the mean WOMAC index half an hour before injection was 61.55 ± 14.3, which decreased to 19.23 ± 8.81 at 2 weeks after injection. WOMAC index was 21.01 ± 10.02, 19.52 ± 9.17, 20.48 ± 11.64, 24.48 ± 14.64, and 26.48 ± 15.64 at 1 month, 2 months, 3 months, 6 months, and 12 months after injection, respectively. When WOMAC index was compared between the two groups, it was clinically better in group I at all time intervals of the study period, but was statistically significant at 2 weeks, 1 month, and 12 months after the procedure (P < 0.05) [Figure 3].

Figure 3.

WOMAC index at different time intervals in the two groups. WOMAC = Western Ontario and McMaster Universities Osteoarthritis Index

No patient in group I and three patients in group II required second injection during the 12-month study period. The interval between two injections was 180 ± 25 days. No procedural complication was observed in any of the patient in the two groups. Pain on administration of injectate was statistically significant between the two groups (P < 0.05), with lesser pain on injection in group II. The pain was temporary and was relieved within few minutes of the procedure; it required no intervention. Three patients in group I and four patients in group II reported increased pain after injection. Two patients each in both the groups reported soreness and swelling after the injection. The increased pain, soreness, and swelling were mild and resolved in 2–3 days after cold fomentation and concurrent medications for OA of the knee. No patient in any group reported fever, rash, itching, or any other side effect after the procedure.

Discussion

In our study, both the groups were comparable regarding baseline patient profile. Majority of the patients were aged 50–70 years in both the groups. Mean weight was about 65 kg, and 70% of them were females. The female preponderance in our study could be attributed to the social milieu of our region. Females are regularly engaged in domestic work, agricultural work, animal husbandry, labor activities, etc. All these activities involve squatting, sitting on floor, and climbing stairs. Similar patient profiles have been observed in other studies.[6,10,13,14,15,16]

Pain scores were less clinically in group I than group II at all time intervals of the study period. However, they were statistically significant between the two groups at 2, 3, 6, and 12 months, with lesser NRS in group I compared to group II. This could be due to thermal lesioning of the peripheral sensory nerve endings resulting in the alleviation of pain. RFA of the genicular nerves blocks the nociceptive pain (A-delta and C-fibers) from transmitting to the central nervous system without destroying the motor or sensory nerves. During RFA, the high temperature of the tip of electrode disrupts the nerve tissue by protein denaturation and coagulation necrosis, which reduces the rest pain significantly but has a limited effect on the activity pain.[17]

We observed clinically and significantly better patient satisfaction in group I compared to group II at half an hour, 2 months, 3 months, 6 months, and 12 months after injection.

Both the techniques of injection were effective as shown by the WOMAC index and resulted in improved health status of patients. There was a statistically and clinically significant improvement in WOMAC index after injection in both the groups at all time intervals during the study period. When we compared WOMAC index between the two groups, it was clinically better in group I at all time intervals of the study period, but was statistically significant at 2 weeks, 1 month, and 12 months after the procedure. Similar to our study, improvement in pain scores and WOMAC index resulting in good patient satisfaction after RFA of the genicular nerve has been observed in other studies.[10,18]

In the present study, we had planned to repeat the injection/RFA, if required, using the same technique as the initial procedure, if pain relief was not adequate (NRS > 4). No patient in group I and three patients in group II required second injection during the 12-month study period. The second injection was performed after 180 ± 25 days.

Fluoroscopic guidance has several advantages for GNB. First, SLGN, SMGN, and IMGN traverse along the periosteal areas connecting the shaft to the epicondyle. Therefore, fluoroscope of the knee joint easily identify target areas for RFGN. Second, ultrasound does not provide clear visualization of small-gauge needles at deep tissue levels, whereas fluoroscopic imaging provides good needle visualization. Third, the use of real-time contrast fluoroscopy and digital subtraction angiography can prevent unintentional intravascular injection.

In our study, the direction of needle around the knee joint was guided by fluoroscope. Proper needle position around the knee joint was confirmed by non-ionic contrast medium (Omnipaque 300) with proper administration of the injectate around the nerves was confirmed under fluoroscopy. Fluoroscope guidance helped in redirecting the needle and drug solution. The use of fluoroscope while performing procedures could have been a major reason for significant improvement in pain scores and WOMAC index in the present study.

In our study, we observed that pain on administration of the injectate was the most common side effect. Most of the patients in group I had moderate pain and in group II had mild pain on injection. However, this pain was temporary, and it was relieved within few minutes of the procedure. It required no intervention. The other common side effect observed in our study was increased pain after knee injection (three patients in group I and four patients in group II). Two patients each in both the groups reported soreness and swelling after the injection. These side effects were mild and resolved in 2–3 days after cold fomentation. No patient in any group reported fever, rash, itching, or any other side effect after the injection. Similar side effect profile after knee injections has been reported by other authors.[11,14,19]

The present study has several limitations. First, the study results may have reflected the experience of one practitioner, which may have limited generalizability of the study findings. Secondly, the long-term effects should be evaluated in the future, based on the results of the short-term effects. In our study, we followed patients for 12 months, but trials could focus on the long-term outcomes up to 2 years after the interventions. Third, this study was not a double-blinded, controlled study as it was difficult to conduct a double-blinded, controlled study with fluoroscope like non-traditional modalities. Fourth, diagnostic nerve block was not done before RFA. We recommend the utilization of such a technique on many OA patients over a longer follow-up period. Fifth, we denervated only three branches innervating the knee joint. The results could have been different with denervating other branches. Sixth, pain measurement tools are bound by subjectivity and patient’s ability to understand evaluation scales.

Conclusions

Fluoroscope-guided RFA of the genicular nerve and genicular nerve block with local anesthetic and steroid are safe and effective techniques for management of symptomatic OA of the knee. Both these techniques provide good pain relief and improvement in physical disability to the patients. Fluoroscope-guided RFA of the genicular nerve is better than fluoroscope-guided genicular nerve block in terms of improvement in pain score, physical status, and patient satisfaction.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.