Transcutaneous electrical nerve stimulator of 5000 Hz frequency provides better analgesia than that of 100 Hz frequency in mice muscle pain model

Abstract

Transcutaneous electrical nerve stimulators (TENSs) have been proved to be effective in muscle pain management for several decades. However, there is no consensus for the optimal TENS program. Previous research demonstrated that a 100 Hz TENS (L‐TENS) provided better analgesia than a conventional TENS (< 5 Hz). However, no research compared a higher‐frequency (> 100 Hz) TENS with a 100 Hz TENS. We used a 5000 Hz (5 kHz) frequency TENS (M‐TENS) and an L‐TENS to compare analgesic effect on a mice skin/muscle incision retraction model. Three groups of mice were used (sham, L‐TENS, and M‐TENS) and applied with different TENS programs on Day 4 after the mice skin/muscle incision retraction model; TENS therapy was continued as 20 min/d for 3 days. Mice analgesic effects were measured via Von Frey microfilaments with the up–down method. After therapy, mice spinal cord dorsal horn and dorsal root ganglion (DRG) were harvested for cytokine evaluation (tumor necrosis factor‐α and interleukin‐1β) with the Western blotting method. Our data demonstrated that the M‐TENS produced better analgesia than the L‐TENS. Cytokine in the spinal cord or DRG all expressed lower than that of the sham group. However, there is no difference in both cytokine levels between TENSs of different frequencies in the spinal cord and DRG. We concluded that the M‐TENS produced faster and better mechanical analgesia than the L‐TENS in the mice skin/muscle incision retraction model. Those behavior differences were not in accordance with cytokine changes in the spinal cord or DRG.

Introduction

Transcutaneous electrical nerve stimulators (TENSs) have been proved to be effective in muscle pain management for several decades [[1], [2]]. In addition to the management of muscle pain, several other types of pain, including visceral pain [3], joint pain [4], and neuropathic pain [[5], [6]], have also been reported to be alleviated after TENS therapy. Recently, TENSs have also been reported to be helpful for nerve regeneration [7] and chronic low back pain [8]. However, there is no consensus for the optimal TENS program in terms of frequency or other parameters (amplitude, duration, impendence, and wave form). Sluka et al [9] proved that a 100 Hz TENS but not a 4 Hz TENS reduced spinal glutamate and aspartate in animals with joint inflammation. A TENS at 80 pulses per second (80 Hz) has been shown to increase the pain threshold related to pressure algometry in healthy participants [10]. Furthermore, a recent randomized, double‐blind crossover study also indicated that 100 Hz and 3 Hz TENSs all increased the pain thresholds of young individuals [11]. All the above‐mentioned results demonstrated that a higher‐frequency TENS is better than a conventional low‐frequency TENS. Meanwhile, TENSs have been reported to reduce spinal neurotransmitter content in rats experiencing pain [12], and have also been shown to inhibit glial activation, mitogen‐activated protein (MAP) kinase activation, protein kinase C gamma (PKC‐gamma), phospho‐cAMP response element‐binding protein (p‐CREB) expression, and proinflammatory cytokine expression, as well as to contribute to the maintenance of spinal opioid receptors [6]. In a muscle pain model, a 100 Hz TENS was also reported to suppress spinal substance P and proinflammatory cytokines [13]. However, there are still few data indicating whether a higher‐frequency (> 100 Hz) TENS provides a better analgesic effect as compared with a 100 Hz TENS and whether it contributes to a lack of cytokine response after higher‐frequency applications. Either 2 kHz or 5 kHz peripheral stimulation was reported to produce antinociception in rats, as measured with tail flick latency, but no comparison was made with a 100 Hz TENS [14]. In this study, we compare the analgesic effect of a 100 Hz TENS with that of a 5000 Hz (5 kHz) TENS and assessed the changes in proinflammatory cytokines within the spinal cords of mice in our muscle pain model. Our data demonstrated better pain relief with a 5 kHz TENS.

Methods

Digital recordable middle‐frequency TENS

A commercially available 5 kHz TENS (M‐TENS) was purchased from Hometech (HT66B; Home Care Technology Co., Ltd, Tainan, Taiwan). With the help of a computer technician, an M‐TENS with a digital recordable function was added to the original machine. Different frequencies plus different wave forms, amplitudes, and durations could be set for testing. A 100 Hz TENS (L‐TENS) (Trio 300; Ito Co., Tokyo, Japan) was also commercially available.

Animal model for muscle pain

Male Swiss CD1 mice (8 weeks old, from the Animal Centre of National Cheng Kung University, Tainan, Taiwan) arrived 7 days before the experiments. All the subsequent animal experiments and procedures were approved by National Cheng Kung University, college of medicine, and performed following the Animal Care Guidelines of the college. A skin/muscle incision retraction (SMIR) model was applied as previously reported [15]. Briefly, mice were anesthetized with inhalation isoflurane (1–2%), and the hair on their medial left thighs was shaved. A 1 cm incision was made in the skin of the medial thigh approximately 4 mm medial to the saphenous vein to reveal the muscle of the thigh. An incision (2–5 mm long) was then made in the superficial (gracilis) muscle layer of the thigh, approximately 2 mm medial to the saphenous nerve. The superficial muscle was then parted further, and a microdissecting retractor was inserted into the incision site. The skin and superficial muscle of the thigh were then retracted by 1 cm, revealing the fascia of the underlying adductor muscles, and this retraction was maintained for 1 hour. Tissues were closed with silk 4.0 and 5.0 Vicryl sutures. The SMIR surgery evoked significant sensory hypersensitivity in the ipsilateral hind paw by Postoperative Day 3 and increased to maximal levels, then reversed to baseline at about Day 12.

Behavioral tests

The behavior was tested as before. Briefly, the mice were habituated for at least 2 days before baseline testing, and the room temperature and humidity were consistently maintained. Mechanical nociceptive responses were evaluated by measuring with a Von Frey filament. The paw was pressed with one of a series of Von Frey hairs and logarithmically incremented (0.02–2.56 g; Stoelting Co., Wood Dale, IL, USA) stiffness presented perpendicular to the plantar surface, as in our previous work. The 50% withdrawal threshold was determined using Dixon's up–down method [16].

Effect of M‐TENS and L‐TENS

On Day 4 after surgery, after confirming the pain response behavior, the animals were lightly anesthetized with 1% isoflurane when administered M‐TENS or L‐TENS treatment for the subsequent 3 days. The left medial thigh was shaved, and two small pre‐gelled adhesive electrodes were applied on the proximal part close to the hip joint and proximally close to the knee joint. In addition, gel and adhesive tape were used to hold the electrodes in place. The TENS program was set with different frequencies at a sensory amplitude, as determined by inducing a muscle contraction and then reducing the intensity to just below this level. The pulse duration was maintained, and the treatments lasted for 20 minutes (Table 1). On the days of TENS treatment, behavioral tests were conducted before the TENS treatment to avoid the disturbance of any immediate effect.

Table 1.

Parameters of TENS (L‐TENS and M‐TENS) treatment.

	L‐TENS	M‐TENS
Frequency (Hz)	100	5000
Pulse width (μs)	100	50–500
Intensity (V)	9 (1.0 mA)	0.0032
Time (min)	20	20

L‐TENS = 100 Hz transcutaneous electrical nerve stimulator; M‐TENS = 5 kHz transcutaneous electrical nerve stimulator; TENS = transcutaneous electrical nerve stimulator.

Western blot analysis

On Day 8, after behavioral confirmation, the animals were rapidly sacrificed after deep anesthesia with 4% isoflurane, and the lumbar spinal cord (L3–L5) and dorsal root ganglions (DRGs; L3–L5) were quickly removed. As before [17], the samples were homogenized mechanically and then incubated on ice for 40 minutes, prior to centrifugation at 7878 g for 20 minutes. Following centrifugation, supernatant was collected and stored at −80°C. Protein samples were separated in sodium dodecyl sulfide–polyacrylamide gel electrophoresis gel and transferred to nitrocellulose blots. The blots were blocked with 5% milk and incubated overnight at 4°C with primary antibodies against tumor necrosis factor (TNF)‐α and interleukin (IL)‐1β. These primary antibodies were used as an anti‐TNF‐α rabbit polyclonal antibody (1:1000, cat# ab66579; Abcam, Cambridge, MA, USA) and IL‐1β (1:1000, cat# ab9722; Abcam; active IL‐1b, 17 kD). Endogenous β‐actin was used as the loading control with the antihuman b‐actin mouse monoclonal antibody (cat# MAB1501; EMD Millipore, Billerica, CA, USA), as described previously. These blots were then further incubated with horseradish perioxide secondary antibody, developed in enhanced chemiluminescence (ECL) solution, and exposed on Hyperfilm (Amersham Biosciences, Little Chalfont, Buckinghamshire, UK) for 1–10 minutes. Subsequently, specific bands were evaluated by apparent molecular size, the intensities of which were captured and analyzed using Image J software (NIH, Bethesda, MD, USA).

Statistical analysis

The results are expressed as mean ± standard error of the mean. Two‐way analysis of variance, Student t tests (2 groups), and multiple comparisons were used for statistical evaluation of the differences among means (a Newman–Keuls test was used for the post hoc analysis). A p value < 0.05 was considered statistically significant.

Results

Table 1 presents the two different TENS parameters for the SMIR mice. A 100 Hz stimulating program was used for L‐TENS and 5 kHz stimulation was used for M‐TENS. When a TENS was applied to the animals, all the parameters were recorded via the commercially available TENS with the help of a computer. Figure 1 presents the pain behavior after the use of the SMIR model and different TENS applications. SMIR evoked significant allodynia after 4 days of surgery compared with the presurgery response, and the symptoms persisted for several days and went back to normal on Day 12. Both L‐TENS and M‐TENS produced an analgesic effect on the animals in pain when applied on Day 4. In addition, a significant difference between the L‐TENS and M‐TENS groups was found on Day 4 after the M‐TENS was applied to the animals. The M‐TENS group experienced a better and faster analgesic effect than the L‐TENS group and even returned to the normal behavioral response level in the paw withdrawal latency after the application of the M‐TENS for 3 days. After stopping the TENS therapy (Day 7), analgesic effects of the L‐TENS and M‐TENS began to reduce but not the effects of the M‐TENS, which still showed a better analgesic effect (Figure 1). Figure 2 represents the proinflammatory cytokine response (Day 8) in the dorsal horn of a mice spinal cord after the SMIR. It clearly shows the SMIR‐induced overexpression of cytokine from 1.00 ± 0.07 to 1.29 ± 0.15 (TNF‐α) and from 1.00 ± 0.06 to 1.58 ± 0.20 (IL‐1β). After the TENS application, it dropped to the following levels: TNF‐α, 0.63 ± 0.12 (L‐TENS) and 0.50 ± 0.04 (M‐TENS); IL‐1β, 0.95 ± 0.03 (L‐TENS) and 0.97 ± 0.05 (M‐TENS). Similarly, Figure 3 represents the proinflammatory cytokine response (also on Day 8) in the mice DRG after the SMIR. It also demonstrates the change in cytokine from 1.00 ± 0.02 to 1.41 ± 0.07 (TNF‐α) and from 1.00 ± 0.07 to 1.19 ± 0.16 (IL‐1β). After the TENS application, it dropped to the following levels: TNF‐α, 1.02 ± 0.08 (L‐TENS) and 0.97 ± 0.10 (M‐TENS); IL‐1β, 1.02 ± 0.08 (L‐TENS) and 0.97 ± 0.10 (M‐TENS). These two figures indicate that the SMIR elevated the proinflammatory cytokines in the spinal cord and DRG, whereas the TENS reduced the cytokine elevation in both the spinal cord and the DRG. However, there was no difference between the 5 kHz TENS and the 100 Hz TENS.

Figure 1.

Behavioral response for mice in SMIR model after M‐TENS or L‐TENS treatment. (Sham: animals with SMIR but no TENS application; L‐TENS: animals with SMIR and 100 Hz TENS application; M‐TENS: animals with SMIR and 5 kHz TENS application.) Arrow indicates the time period for TENS application. Values are mean ± SEM, n = 4–6 per group. * p < 0.05 versus SMIR sham group (Student t test). p < 0.05 versus L‐TENS group (Student t test). L‐TENS = 100 Hz transcutaneous electrical nerve stimulator; M‐TENS = 5 kHz transcutaneous electrical nerve stimulator; SEM = standard error of the mean; SMIR = skin/muscle incision retraction; TENS = transcutaneous electrical nerve stimulator.

Figure 2.

Representative Western blots of TNF‐α and IL‐1β in dorsal horn of the spinal cord represent the densitometric analysis for the amount of protein relative to β‐actin (as loading control) in the lumbar spinal cord (L3–L5) for the SMIR model. (Naive: animals without SMIR; Sham: animals with SMIR but no TENS application; L‐TENS, M‐TENS: animals with SMIR, and L‐TENS and M‐TENS application.) Values are mean ± SEM, n = 4–6 per group. * p < 0.05 versus expression of sham group (Student t test). IL‐1β = interleukin‐1β; L‐TENS = 100 Hz transcutaneous electrical nerve stimulator; M‐TENS = 5 kHz transcutaneous electrical nerve stimulator; SEM = standard error of the mean; SMIR = skin/muscle incision retraction; TENS = transcutaneous electrical nerve stimulator; TNF‐α = tumor necrosis factor‐α.

Figure 3.

Representative Western blots of TNF‐α and IL‐1β in dorsal root ganglion represent the densitometric analysis for the amount of protein relative to β‐actin (as loading control) in the dorsal root ganglion (L3–L5) for the SMIR model. (Naive: animals without SMIR; Sham: animals with SMIR but no TENS application; L‐TENS, M‐TENS: animals with SMIR, and L‐TENS and M‐TENS application.) Values are mean ± SEM, n = 4–6 per group. * p < 0.05 versus expression of sham group (Student t test). IL‐1β = interleukin‐1β; L‐TENS = 100 Hz transcutaneous electrical nerve stimulator; M‐TENS = 5 kHz transcutaneous electrical nerve stimulator; SEM = standard error of the mean; SMIR = skin/muscle incision retraction; TENS = transcutaneous electrical nerve stimulator; TNF‐α = tumor necrosis factor‐α.

Discussion

This is the first paper, to our knowledge, presenting a comparison between 5 kHz TENS and 100 Hz TENS with respect to their analgesic effect in an SMIR muscle pain model. It not only indicates that the 5 kHz TENS is superior to the 100 Hz TENS in terms of pain behavior, but also demonstrates the proinflammatory cytokine change after the SMIR in the spinal cord and DRG when a TENS program is applied for 20 minutes each day. Although we failed to demonstrate the cytokine difference between different frequencies of a TENS, we can claim that the proinflammatory cytokines in the spinal cord and DRG decreased when a 100 Hz or 5 kHz TENS was applied. This reveals the unique contribution of this manuscript.

Regarding the possible mechanism for the superiority of the 5 kHz TENS over the 100 Hz TENS, there are several possible explanations. One possibility is that the Western blotting method for cytokine evaluation is not sensitive enough to detect the differences among TENSs of different frequencies. The other possibility is that the tissue sampling time is not right for cytokine evaluation. Certainly, we never exclude the possibility that other molecules, such as substance P, glutamate, or chemokine, may be the key to these differences [13]. Cannabidiol and endogenous opioid peptide‐mediated mechanisms may also be involved in TENS analgesic effects [18]. Glial activation, MAP kinase activation, PKC‐γ, and p‐CREB expression have also been mentioned before to explain the possible analgesic mechanism of a TENS [6]. The real mechanism requires further evaluation.

In conclusion, this research presented a unique result when a TENS was applied for 20 min/d for 3 days. A TENS (either 5 kHz or 100 Hz) produced analgesia in an SMIR mice muscle pain model, and it was observed that a 5 kHz TENS was superior to a 100 Hz TENS. Furthermore, the proinflammatory cytokines (TNF‐α and IL‐1β) within the spinal cord and DRG were all suppressed after different‐frequency TENS therapy, although differences in cytokines within different TENSs could not be observed using the Western blotting detection method. Further investigation is suggested to evaluate the possible mechanism required to observe such differences.

Acknowledgments

The research is funded by the E‐Da and National Cheng Kung University Co‐promotion Research Program in 2014 (10358) and also funded by Innovative Exploration Program from School of Medicine, National Cheng Kung University, in 2012. We also want to thank the digital record method of our TENS from the Industrial Technology Research Institute, Tainan Branch.