Pulsed Radiofrequency Energy for Treatment of Chronic Pain Syndromes
Pulsed radiofrequency energy (PRFE) has multiple medical applications including pain modulation, wound healing, and bone repair.1,2 PRFE refers to a noninvasive, nonthermal method of delivering nonionizing electromagnetic energy to a targeted area with a frequency ranging from 1 to 1000 Hz.
The pulsatile delivery, with bursts lasting from 10 µsec to 1 msec, allows for dissipation of heat making this method non-ablative. The FDA has cleared such shortwave diathermy devices for adjunctive use in the palliative treatment of postoperative pain and edema in superficial soft tissue.3,4
Mechanisms for proposed analgesia include modulating calcium and calmodulin pathways,5 increasing endogenous opioid precursor mRNA,6 altering transcription of cytokine and matrix metalloprotease levels,7 increasing chondrocyte proliferation,8 and enhancing noradrenergic and serotonergic descending inhibitory pain pathways.9 However the exact mechanism is still not completely understood at this time.
There are several studies that have demonstrated some effectiveness with PRFE. In a recent meta-analysis of 16 controlled trials, 11 trials with PRFE had positive outcomes and 5 trials with neutral outcomes that were measured across different pain conditions.3 There were 2 double blind randomized controlled trial that showed PRFE was more effective than sham in shoulder pain and knee osteoarthritis.10,11 Additionally other trials and case series have shown PRFE to provide significant pain reduction in chronic back pain, non-specific wrist pain, plantar fasciitis, and postoperative pain.12-15
The role and effectiveness of PRFE in treating various chronic pain conditions is still being defined, and it is unclear if certain conditions fail to respond. Additionally there is very little literature detailing whether treating beyond 4 weeks is beneficial. Our goal for performing this review was to determine whether PRFE was beneficial for chronic pain and if additional benefit occurred with those who received 8 week treatment.
Study Method and Design
The objective of this observational series was to record the effects and duration of PRFE treatment across various pain syndromes. A retrospective analysis was performed via chart review on 40 patients in an outpatient Veteran’s Affairs (VA) Physical Medicine and Rehabilitation Clinic in Long Beach, California. Internal Review Board (IRB) approval was attained through the Southern California Institute for Research and Education before collecting any data. Inclusion criteria comprised of every patient that received PRFE treatment within the principal investigator’s patient population at the VA.
Eligible patients were required to demonstrate chronic pain unrelieved by a recent procedure or injection. Exclusion criteria comprised of any patient who did not take a PRFE device home after the initial office trial. It should be noted that the treating physician did not offer this treatment to anyone with active infection, cancer, or an implantable electronic stimulating device.
Patients prescribed PRFE were first offered an outpatient initial trial of transcutaneous PRFE. A treatment applicator pad was placed directly over the site of maximal pain and a nonionizing, nonthermal carrier frequency energy was emitted at 27.12 MHz for 30 minutes. The pulsed RF (pulsed width 42 µsec, pulse frequency 1 kHZ) has a predicted electrical field strength of 591 V/m at 5 cm above the applicator.
After receiving the trial, patients were then instructed to take the PRFE device home to use twice daily for 30 minutes. Patients were then instructed to return to the clinic within 4 weeks to evaluate their pain and activities of daily living (ADL). If the patient’s pain improved but their ADL had not, they had the option to extend their trial for an additional 4 weeks.
The reviewer collected each subject’s medical diagnosis as determined by radiographs, laboratory results, and clinical judgment. Numeric pain scores were also collected via chart review, which included pain before initial trial (initial), 30 minutes after trial (post-trial), 4 weeks, and 8 weeks. This was scored on a numeric rating scale of 0-10 with 0 reflecting no pain, 1-3 mild pain and mild interference on ADL, 4-6 moderate pain and moderate interference on ADL, and 7-10 severe pain and inability to perform ADL.
Forty subjects met inclusion criteria, and no patient was excluded due to trial failure. Therefore 40 patients completed the initial trial, which was reduced to 36 subjects at 4 weeks and 21 subjects at 8 weeks. Patient dropout occurred for various reasons: pain resolved with treatment, twice daily frequency was inconvenient, insufficient efficacy, and patient inability to access electricity to run device. It is important to note that no subject was reported to discontinue treatment secondary to adverse side effects.
Numeric pain scores were averaged for each stage of the study (7.9 initial, 5.7 post trial, 4.6 at 4 weeks, and 3.0 at 8 weeks), which showed consistent reduction of pain over time (Figure 1). Statistical analysis using the Wilcox Signed Rank Test for a nonparametric two-tailed test was performed across the 40 patients. Results showed that the post-trial, 4 weeks, and 8 weeks average pain scores were all statistically reduced compared to initial pain scores (P<0.001). Additionally, there was statistical improvement from post-trial to 4 weeks (P<0.006), which was maintained without further change from 4 weeks to 8 weeks.
The data was analyzed by patient diagnosis. Each of the patients treated had at least one diagnosis and 3 patients met criteria for two diagnoses that explained their pain symptoms (eg, spinal stenosis and lumbar radiculopathy). As a result, the 40 patients had a total of 43 diagnoses after being grouped into the following categories: rotator cuff tear, failed back syndrome, myofascial pain, radiculopathy, spinal stenosis, diabetic neuropathy, knee osteoarthritis, lumbar facet arthritis, ankle osteoarthritis, and postsurgical abdominal pain (Table 1).
Due to the low sample sizes, statistical significance of the diagnoses was not reported to prevent false comparisons. Nevertheless, all 10 diagnoses showed a generalized trend of improvement over time with few exceptions: slight worsening of failed back syndrome from post-trial to 4 weeks, no change in diabetic neuropathy during post-trial to 8 weeks, no change in knee osteoarthritis at 4 and 8 weeks, and some worsening of postsurgical abdominal pain at 8 weeks (Figure 2). However it should be noted that these exceptions occurred only in diagnoses with low sample sizes of 3 or less, but regardless all diagnoses showed improvement from their initial pain scores.