This study was approved by the Institutional Review Board of the Chang Gung Medical Foundation, and this trial was registered at the ClinicalTrials.gov website (NCT02468297). We registered this trial retrospectively because of the following reasons. We began to enroll patients in April 2004 but stopped in November 2008. We needed to ensure that the amplitude of the electric current was around an appropriate range. Thus, the trial protocol was revised for performing a test to determine the resistance of the sole of every participant. After the revision of the protocol, we re-enrolled patients in April 2009. The manufacturer of EAS did not want to disclose the trade secret before the trial had been completed.

Informed consent was obtained from each participant. A total of 90 patients of both sexes were prospectively screened for study participation. All patients were outpatients of the Department of Rehabilitation, Orthopedics, and Chinese Medicine. We selected patients with the following inclusion criteria: diagnosis of CLBP, that is, the location of pain was below the 12th rib and above the horizontal gluteal crease and lasted for more than 12 weeks; age range of 20 to 60 years; and provision of signed informed consent. Patients were excluded if he or she met the following medical conditions, including cancer, rheumatoid arthritis, renal stone, diabetes mellitus, pacemaker implantation, under steroid treatment, fracture or surgical history of the back, spinal cord compression syndrome (eg, herniated intervertebral disc or spinal cord disorder), visceral organ infectious disease (eg, pancreatitis and pyelonephritis), or visual coordination disorder. Women who experienced menopause before 50 years of age, underwent ovary excision, or were pregnant were excluded. Further, women considering to be pregnant were asked to not enroll, and enrolled women were asked to agree to contraception or abstinence. Moreover, before joining the trial, their pregnancy test must be negative. Patients were also excluded if they were not free from previous participation in other trials within 30 days before joining this trial, showed contraindications to ibuprofen, or if they had poor heart, liver, gastrointestinal tract, and renal function. Further, if the physician suspected that EAS treatment might have adverse effects on the patient based on the physical examination and laboratory data at the first visit, the patient was not enrolled.

All patients were randomly assigned to either EAS group (EAS plus placebo NSAIDs) or NSAID group (sham EAS plus NSAIDs) based on a concealed allocation approach. A computerized random number table was used to determine group allocation. Numbered opaque sealed envelopes contained ibuprofen or placebo analgesics and numbered EAS, and no restrictions were placed on randomization. The envelopes for the analgesics and EAS were kept by a pharmacist who was not an assessor of the study. All assessors and patients were blinded to group allocation.

In the first visit (first week), patients were assessed by physical examination, vital sign evaluation, blood test (complete blood count, clinical chemistry, and pregnancy test), radiography of the lumbar spine in the anterior-posterior and lateral views, deep tendon reflex test, and sensation, range of motion, and pain scales. Subsequently, patients received EAS treatment (EAS group) or placebo EAS (NSAID group) for 1 hour 3 times a week for 6 weeks. Patients were assessed at baseline (first visit), before and after each EAS treatment (2nd-19th visit), and 2 weeks after the last treatment (20th visit).

In the first week of treatment, patients in the EAS group received placebo analgesics. Before EAS treatment, patients were asked to remove any conducting metal accessories, watch, mobile, and socks. Each patient had their own EAS.

During treatment, if the patient experienced any discomfort, indicated by symptoms such as extreme LBP, spasm in both lower limbs, paralysis, tachycardia, or dizziness, the assessors recorded the symptoms and their duration. If the symptoms were slight, EAS treatment was ceased for 5-10 minutes and then continued after the symptoms were relieved. If the symptoms were severe, EAS treatment was stopped at once, and the assessors recorded the reasons for stopping the management and the treatment duration. After each treatment, the patient rested for 3-5 minutes, relaxed the joints of the lower limbs, and then underwent the efficacy assessment.

In the first week of treatment, patients in the NSAID group received 400 mg ibuprofen 3 times a day. Subsequently, they received EAS treatment without electric current 3 times a week for 6 weeks, as in the EAS group.

The population evaluated in the intention-to-treat (ITT) analysis included those who met the inclusion and exclusion criteria and who received treatment at least once without violating the protocol. The population evaluated in the per-protocol (PP) analysis was included as those receiving treatment at least 12 times without violating the protocol.

Eighty-three patients were enrolled in this study between April 2009 and January 2012. One patient was excluded because of not meeting the inclusion criteria, and 10 patients were excluded because they received analgesics. The CONSORT (Consolidated Standards of Reporting Trials) (Multimedia Appendix 1) flow diagram for the study is shown in Figure 2. Patient characteristics, including age, sex, weight, height, and duration of pain, were similar between the groups, except for age in the ITT analysis, which differed significantly (EAS group, 45.7 years and NSAID group, 41.1 years, P=.04) (Table 1).

No severe adverse effects were reported in this study. Moderate and mild adverse effects such as pain in the extremities, back pain, hypoesthesia, and arthralgia were recorded. As shown in Table 2, no significant differences were observed between both the groups. All adverse effects occurred during the period of EAS (or sham) treatment and were relieved after the completion of EAS (or sham) treatment without sequelae.

We conducted a trial to compare the effects of EAS and conventional treatment (NSAIDs) for CLBP. In the ITT and PP analyses, the change in the VAS score from baseline in the EAS group in each visit was greater than that in the NSAID group, and significant differences were observed in the 5th visit and 9th visit in the ITT analysis (P=.048 and P=.048, respectively). In the ITT and PP analyses, the treatment success rate in each visit in the EAS group was higher than that in the NSAID group, and significant differences were observed only in the 16th visit (P=.04). In both the ITT and PP analyses, there were no significant differences in the VAS score before and after each treatment between the groups. No significant differences were observed in the time required for achieving pain remission, range of motion, SF-36 scores, and RMDS scores between the groups, but significant differences were observed in the left rotation at the 2nd and 4th visits.

EAS are a novel medical device, and it combine the properties of electroacupuncture and TENS and are different from acupoint TENS [8-10]. Electrodes of TENS are placed on the painful area of the body, that is, close to the lesion or near the nerve bundles proximal to the painful area [8]. Instead of local treatment, EAS are worn on the feet with electrical stimulation over 2 acupoints (KI1 and shimian) on the soles. Moreover, the administration of both conventional TENS and acupuncture-like TENS is required to achieve physiological intentions (paresthesia and muscle twitch, respectively) to confirm its effectiveness, thereby making patient blinding impossible [8-10]. By contrast, EAS outputs complex waveforms composed of low-frequency and middle-frequency waves. Furthermore, the amplitude of the current is so low that patients wearing EAS barely notice they are under treatment.

In clinical practice, needling on yongquan (KI1) is usually employed for treating conditions of disturbance of consciousness [20]. Few studies have evaluated the effect of electroacupuncture on KI1 [21]. Shimian is an extra acupoint, not located on the meridian, and it is employed to treat insomnia. According to the theory of traditional Chinese medicine, the flow of defensive qi (wei qi) travels around the outer side of the body (yang) through a special pattern and streams into the inner side of the body (yin) through the kidney meridian [22-24]. Moreover, in traditional Chinese medicine, one mechanism underlying insomnia is that the defensive qi cannot stream from the yang to yin [22,23]. Thus, many acupoints of the kidney meridian are believed to regulate the flow of defensive qi and are used to treat insomnia [25,26]. According to the theory of traditional Chinese medicine, chronic pain is a result of disharmony or depletion in the supply of qi [27,28]. For these reasons, we suspected that the regulation of the circulating defensive qi might relieve chronic pain. Therefore, we chose the yongquan and shimian as our treatment acupoints and applied electric current to these areas.

Yang et al [12] conducted an animal study, wherein the results revealed that the application of TENS on KI1 could produce analgesic effects; this was demonstrated through the increased response time of hind paw licking to thermal stimuli induced by complete Freund’s adjuvant. The results of immunostaining and western blotting of the brain and spinal tissue of rats revealed that the application of TENS on KI1 inhibited ERK2 activation and c-Fos expression, which are associated with pain perception. In that study, TENS was not applied as a conventional method to the painful lesion or nearby nerve bundles but on an acupoint distal to the painful site. Stimuli were transferred through the peripheral afferent fiber to the central nervous system and they caused changes in the brain network, thereby affecting pain perception and modulation [13]. These findings can explain that EAS may alleviate CLBP through a central pathway.

Furthermore, we think that EAS also work possibly through a peripheral pathway. Application of acupuncture on acupoints remote from the lesion or site of pain to alleviate disease or uncomfortableness is a fundamental concept based on the meridian theory in traditional Chinese medicine. A study demonstrated that acupuncture on distal acupoints could increase the blood flow of the meridian distribution area that the acupoints belong to [14]. A few studies found that CLBP is correlated with insufficient blood flow in the lumbopelvic region [15,16]. To summarize these theories and findings, we speculate EAS may alleviate CLBP by improving the blood flow.

The strengths of our study are as follows. First, we designed a randomized, double-blinded, placebo-controlled, dual-intervention (real EAS with placebo analgesic and sham EAS with ibuprofen) crossover trial to investigate the effect of EAS compared with that of conventional NSAIDs. Achieving participant blinding is very difficult in the design of an electroacupuncture or TENS study unless patients have little knowledge of electroacupuncture or TENS [8]. In this study, all patients were unaware of EAS because EAS are innovative medical devices. Even if patients were aware of EAS, they still could not differentiate between the real and sham treatment because the applied current was too small to be felt. Second, we employed rigorous methodology in sample size calculation and analytical paradigms (ITT and PP) to evaluate the effects of the intervention [29]. Third, we obtained various outcome measurements to compare the effects of EAS and NSAIDs; we focused not only on the pain itself but also on the mental status and the range of motion affecting the quality of life of the patients.

Our study has several limitations. First, we did not test the success of blinding by using an assessment tool such as that developed by Deyo et al [30], although patients were not knowledgeable about EAS and the applied current was extremely low. Second, the duration of ibuprofen intake was only 1 week in order to prevent the side effects such as gastrointestinal, cardiovascular, renal, hepatic, and other systemic adverse effects [31-35]; therefore, we could not determine whether a longer intake of ibuprofen would achieve a higher efficacy. In clinical practice, the duration of NSAID use in patients with CLBP should be examined and evaluated regularly [36]. Treatments for CLBP include pharmacological, interventional, and surgical strategies [36]. Oral NSAIDs are widely used as first-line therapy for CLBP [36]. However, for specific populations with CLBP, the risk of adverse effects and drug-drug interactions [37-41] should be considered before prescribing oral NSAIDs. In this situation, developing an alternative therapy with similar efficacy to NSAIDs is crucial. The effect of TENS on CLBP is still debatable [42]; although the effectiveness of acupuncture is highly recommended [43], it depends on the performer.

EAS might serve as a reliable alternative therapeutic tool. Additional studies for evaluating the effect of EAS on other sites or categories of pain such as neck pain or pain caused by cancer should be conducted. EAS are more than just a tool for treating LBP. If sensors can be integrated to monitor physiological parameters and then shared on the internet, physicians will be able to personalize the amount of electrical stimulation through remote control [44]. This system will make a great contribution to mobile health, and in this era of new infectious diseases such as COVID-19, this feature will enhance the applicability of EAS [45].

In this prospective double-blinded randomized controlled study, we demonstrated that EAS had a better treatment success rate and analgesic effect compared to NSAIDs in patients with CLBP during some of their visits for treatment, with partial improvement in the range of motion. EAS could be considered as a potential noninferior and reliable alternative therapeutic tool for patients with CLBP who are contraindicated for oral NSAIDs.

There are restrictions on the availability of data for this study owing to the signed consent agreements regarding data security, which only allow access to external researchers for research monitoring purposes. Requestors wishing to access the trial data to replicate or check our analyses can apply to Protech Pharmaservices Corporation (contact@ppccro.com) after receiving permission from the East Bamboo Company Limited (ingo@eastbamboo.com.tw) and principal investigator Doctor Yu-Sheng Chen (cusp01@cgmh.org.tw).