Impact of Intramuscular Electrical Stimulation on Multifidus Muscle Function in Chronic Low Back Pain: A Randomized Controlled Trial

Introduction

Chronic low back pain is a significant public health problem affecting millions of people worldwide [1]. Moreover, low back pain leads to high clinical and economic burdens in high-income countries, with annual direct and indirect costs ranging from €2.3 billion to €2.6 billion and €0.24 billion to $8.15 billion, respectively [2]. Chronic low back pain, defined as pain that persists for more than 3 months, can severely limit mobility and daily activities, leading to increased disability and related healthcare costs [1]. This condition is often associated with impaired function of the multifidus muscle, a key stabilizer of the lumbar spine, which weakens or atrophies and is a source of sensitization in patients with chronic pain [3].

Traditional methods of treating chronic low back pain include pharmacological therapies, physiotherapy, and invasive surgical procedures, which are not always effective and may be associated with the risk of adverse effects or complications [4–6]. In recent years, evidence has appeared in the scientific literature for the greater effectiveness of a multimodal approach in treating chronic low back pain [4–6]. Popular interventions used by physiotherapists include dry needling [7], manual trigger point therapy [8], kinesitherapy [9], transcutaneous electrical nerve stimulation [10], and acupuncture [11,12]. In addition, percutaneous minimally invasive electrotherapeutic needle treatments, such as percutaneous electrical nerve stimulation [13] and intramuscular electrical stimulation (IMES) [14], which use needles to deliver low-frequency (5–100 Hz) and low-intensity (0.1–2 mA) electrical current to trigger points in muscles or nerves, are also reported to be effective in treating chronic low back pain [7].

Among the muscles involved in chronic low back pain, the multifidus has received particular attention due to its essential role in segmental spinal stability and postural control [15,16]. Targeting this muscle can be clinically relevant, as dysfunction of the multifidus has been associated with persistent low back pain [17]. Furthermore, the multifidus can be a source of nociceptive input due to structural or functional impairments, potentially contributing to central sensitization – a mechanism underlying the persistence of chronic pain [17]. IMES, by delivering electrical impulses directly into the muscle tissue, may offer an advantage in reactivating inhibited motor units, improving local circulation, and reversing disuse-related degeneration. However, there is scarce evidence supporting that therapies like IMES solely address the multifidus muscle [18], and the specificity of such interventions for this muscle needs clearer documentation. While much attention has been given to the multifidus in relation to chronic low back pain, there remains limited evidence on how specific therapies, like IMES, can isolate and target this muscle effectively in isolation from other structures of the spine [19]. Understanding the physiological mechanisms behind multifidus function is essential to developing effective treatment strategies for low back pain [20]. Dysfunction of the multifidus is characterized by loss of neuromuscular control, leading to muscle atrophy and degeneration, contributing to low back pain [21]. An increasing number of studies are beginning to show the potential of various therapeutic interventions. IMES is emerging as a promising method for treating pain by targeting muscle dysfunction and improving neuromuscular function [14]. However, despite initial findings indicating the efficacy of IMES in treating pain, there is still a significant gap in the literature regarding its specific applications to the multifidus muscle to relieve low back pain [22].

Interventions specifically targeting the multifidus muscle are essential. Sharma and Chaudhary [19] showed that IMES targeting the multifidus muscle significantly reduced pain levels and disability in a patient with chronic LBP. Similarly, Chu et al [23] found that electrical twitch-obtaining intramuscular stimulation in paraspinal muscles at the bilateral T10-S1 levels provided greater immediate pain relief than traditional muscle and skin stimulation methods, although the long-term significance was not statistically established. Thus, recent clinical trials have shown positive results when IMES is applied to the multifidus, indicating reduced pain and improved functional aspects of daily living in patients with low back pain [20]. Furthermore, the results suggest that this method can facilitate proprioceptive feedback and neuromuscular re-education, key components of rehabilitation of the affected muscle [22]. A systematic review by Linzmeyer et al [14] found that neuromuscular electrical stimulation improved muscle function in chronic LBP patients, improving strength and endurance. Despite promising results, the literature has significant gaps that warrant further investigation. For example, the long-term effects of IMES on the multifidus muscle remain understudied, with few longitudinal studies documenting outcomes beyond the initial phases of treatment [24]. While the literature suggests the importance of evaluating stability characteristics, tissue responses, and other functional outcomes, much of the focus has been on pain reduction (eg, Numeric Rating Scale, NRS) [21]. More comprehensive studies are needed to assess the broader therapeutic effects of interventions on muscle function, stability, and tissue reactions. Restoring spinal stability, particularly by rehabilitation of the multifidus muscle, is crucial not only for pain relief but also for preventing future episodes of low back pain and improving overall function [16]. As such, assessing the effects of interventions on stability parameters, such as trunk control and postural alignment, will provide a more holistic understanding of their long-term benefits in individuals with chronic low back pain. Therefore, there is a pressing need for high-quality randomized controlled trials to evaluate the clinical efficacy and neuromuscular effects of IMES in individuals with chronic low back pain. In particular, interventions that specifically target the multifidus muscle and employ objective measures of muscle and functional outcomes are critical to advancing evidence-based practice in chronic low back pain management.

In addition to evaluating the mechanical and neuromuscular aspects of treatment, assessing transcutaneous hyperaemic tissue reactions may provide information into the underlying physiological processes that occur during therapeutic interventions for chronic low back pain. Hyperemia is often used as an indicator of tissue metabolism and inflammatory processes [25]. Following prolonged muscle dysfunction, tissues surrounding the lumbar spine may exhibit impaired circulation, contributing to pain, stiffness, and delayed healing [26]. Tissue perfusion can be associated with muscle health, and the removal of metabolic waste products, both important for tissue recovery [27]. Impaired blood flow can exacerbate muscle tension and stiffness, which are commonly seen in individuals with chronic low back pain [28]. Therefore, integrating tissue perfusion assessments with other biomechanical and sensory outcomes can provide a more holistic understanding of the effectiveness of treatments like IMES in improving tissue health and functional recovery.

Given the above, this study compared the efficacy of a 6-week physical therapy program incorporating IMES therapy targeting the multifidus muscle versus transcutaneous electrical nerve stimulation (TENS) in patients with chronic low back pain in terms of stability parameters and transcutaneous hyperemic tissue response variables. We hypothesized that IMES would lead to greater improvements in pain relief and neuromuscular function compared to conventional physiotherapy with TENS in patients with chronic low back pain. IMES uses medium-frequency currents to deeply stimulate motor neurons, possibly enhancing muscle recruitment, coordination, and tissue healing, which can be important for restoring neuromuscular function and reducing the muscle weakness often seen in chronic pain conditions.

Material and Methods

STUDY DESIGN AND PROCEDURES:

This study employed a single-blind, randomized design. Following baseline assessment, participants underwent a 6-week experimental intervention. A second assessment was conducted immediately after the intervention, followed by a final follow-up assessment 4 weeks later, during which no treatment was provided. Participants were randomly assigned to 1 of 2 groups using simple randomization using opaque, sealed envelopes, ensuring a 1: 1 allocation ratio. The control group comprised 15 participants, while the experimental group included 15 participants. To maintain allocation concealment, randomization was performed before baseline measurements were taken. The intervention was administered once per week over 6 weeks. Outcome measurements were conducted at 3 time points: at rest before the intervention (baseline), immediately after completing the 6-week treatment, and 4 weeks after the conclusion of the treatment to assess longer-term effects.

To ensure blinding and impartial allocation, the randomization process was managed by a researcher who was not involved in the subsequent assessments. Additionally, specific researchers were assigned exclusively to the evaluation phases, while others were responsible for administering the treatments. Participants were randomly assigned to treatment groups in a 1: 1 ratio using a simple randomization method. The assignments were then placed in sequentially numbered, opaque, sealed envelopes. Before their initial assessments, participants were allocated by drawing the next numbered envelope. This method maintained allocation concealment from both researchers and participants until the point of intervention. The randomized allocation remained unchanged throughout the course of the study. To minimize bias, evaluators were blinded to participant treatment assignments throughout the study. Therapists who delivered the interventions and participants were not blinded, as the nature of the treatment precluded this.

The measured parameters included blood tissue perfusion described in arbitrary units (AU), muscle tension in Hertz (Hz), muscle stiffness in Newtons per meter (N/m), and pain intensity using the numerical rating scale (NRS, 0–10). Additionally, the pressure pain threshold was recorded in Newtons per centimeter (N/cm), along with the forward flexion range of motion assessed using the finger-to-floor test in centimeters. Functional and biomechanical assessments included maximum voluntary contraction in kilograms and postural stability measured in square millimeters (mm²).

This study was approved by the Ethics Committee of the Polish Physiotherapy Association (approval number 01/01/25, granted on 15/01/2025) and was registered as a clinical trial under the following DOI: https://doi.org/10.1186/ISRCTN16484644. The study was conducted according to the principles outlined in the Declaration of Helsinki. All participants were fully informed about the study procedures, potential risks, and their right to withdraw at any time without providing a reason. Before participation, each individual provided written informed consent, completed a health questionnaire, and received detailed information regarding possible side effects and adverse events.

PARTICIPANTS:

Participants for this study were recruited from outpatient clinics, community centers, and online advertisements. A convenience sampling method was employed, targeting individuals with chronic low back pain. To determine the appropriate sample size for the study, a power analysis was conducted using G*Power (version 3.1). The analysis was based on an ANCOVA mixed design, which included both a within-subjects factor and a between-subjects factor, with a continuous covariate. The effect size for the primary comparison was estimated based on previous studies using the numeric rating scale [29], which suggested a medium effect size (f=0.811) for the anticipated main effects and interactions. The alpha level was set at 0.05, and a power of 0.95 was aimed to detect a significant effect. The numerator degrees of freedom for the ANCOVA were calculated based on the number of levels of the between-subjects factor and within-subjects factor. Specifically, the sample size estimation was conducted for the main effects of the between-subjects factor, the within-subjects factor, and their interaction, considering the covariate’s influence on the dependent variable. The sample size calculation determined that 22 participants in total (ie, not per group) were needed to ensure adequate statistical power for the study.

Thirty volunteers (n=30) with chronic bilateral low back pain of non-specific origin, lasting for more than 3 months, and exhibiting characteristic symptoms of myofascial syndrome (eg, palpable tight bands and hypersensitive tender points) were included in the study (Figure 1). However, 1 participant was excluded due to a history of surgical intervention in the lumbar spine. The participant group comprised 21 men and 8 women, with mean age 39.2±10.9 years. The mean body mass index (BMI) was 25.9±2.4 kg/m2, the mean body weight was 80.2±13.5 kg, and the mean height was 176.1±7.8 cm. Participants were randomly allocated, resulting in a control group of 14 (3 women, 11 men) and an experimental group of 15 (5 women, 10 men). Participant adherence to the intervention was monitored using an attendance log to track the presence of all participants at each session.

INCLUSION AND EXCLUSION CRITERIA:

To be eligible for participation in the study, individuals had to meet several key criteria. First, participants had to have bilateral low back pain of non-specific origin, lasting for a minimum of 3 months. The pain had to have a variable character (eg, intermittent, constant, aching, or sharp), with no clear underlying diagnosis such as disc herniation or radiculopathy. Secondly, participants had to report a pain score of at least 4 on the Numeric Rating Scale (NRS), indicating moderate to severe pain. Additionally, participants had to present with palpable tight bands in the lower back muscles, indicative of muscle hypertonicity or myofascial trigger points. A hypersensitive tender point had to be palpable within the tight band(s) in the lower back, and this point had to reproduce the participant’s pain upon pressure. Furthermore, participants had to experience local or referred low back pain when pressure was applied to the identified tender points. Lastly, participants had to exhibit limited forward flexion mobility, with a measurable reduction in range of motion in the lumbar spine compared to age- and sex-matched normative values. Regarding the sample, although myofascial syndrome has been reported more frequently in women at a 2: 1 ratio, the study did not exclude men and recruited participants based on clinical criteria, not sex. The composition of our sample reflects the availability and eligibility of participants meeting the diagnostic criteria during the recruitment period, not a predetermined demographic balance.

Certain conditions disqualified individuals from participation in the study. Participants currently receiving pharmacological treatment for low back pain, such as opioids, muscle relaxants, or corticosteroids, were excluded, as these treatments could have confounded the results. Those who presented with radicular disease, as confirmed by clinical tests such as the Lasègue test or imaging (eg, MRI indicating radiculopathy), were also excluded. Furthermore, individuals with a history of an acute lumbar spine injury within the past 6 months or those with neurogenic claudication – manifesting as pain or numbness in the legs exacerbated by standing or walking and relieved by sitting – were not eligible. Participants who had undergone any type of lumbar spine surgery, including procedures such as discectomy or spinal fusion, were excluded due to the potential for altered tissue or neural pathways. Individuals with systemic connective tissue diseases like rheumatoid arthritis, ankylosing spondylitis, or systemic sclerosis were excluded, as these conditions could have complicated the interpretation of results. In addition, participants with a known nickel allergy or fear of needles (eg, a history of severe anxiety or syncope related to needling) were not included. Individuals diagnosed with epilepsy or other seizure disorders were also excluded, as electrical stimulation could have posed a risk. Participants with metal implants in the lumbar spine, such as rods, plates, or screws, or those with any electronic devices like a pacemaker, were excluded due to potential interference with electrical stimulation. Lastly, pregnant or breastfeeding women, as well as individuals with unspecified or active skin lesions or moles in the lumbar region, were not eligible, as these conditions could have interfered with the acupuncture or electrical stimulation procedures.

INTERVENTION:

In the control group, TENS was applied using a PhysioGo.Lite device (Poland, 2020) with 5×5 cm silicone electrodes placed bilaterally over the lumbar paraspinal muscles, targeting the erector spinae and multifidus muscle groups. The electrodes were placed approximately 2–3 cm from the midline, at L3/L4 and L4/L5 levels based on the typical anatomical landmarks for low back pain management. These segments are commonly affected in patients with low back pain, making them ideal for targeting pain relief and muscle relaxation. The stimulation parameters were set to a frequency range of 50–100 Hz to engage A-beta fibers for pain modulation, with an intensity of 20–50 mA to produce a strong but comfortable sensation (sensory threshold without muscle contraction). The total duration of TENS application was 15 minutes per session. This modality was intended to reduce pain and promote relaxation of the paraspinal musculature through sensory nerve stimulation.

Myofascial trigger point therapy was administered according to standardized guidelines for low back pain management [30,31]. Palpation of myofascial trigger points in the lumbar paraspinal muscles, particularly in the erector spinae and multifidus, was performed, followed by manual ischemic compression (applied at the most tender points identified) for 30–60 seconds per point. This therapy aims to reduce muscle tension and release myofascial trigger points, thereby decreasing pain and improving muscle flexibility. The total duration of this therapy was 15 minutes, with focus on bilateral lumbar regions.

A structured exercise program, which lasted for 15 minutes per session, was implemented. The exercise program was designed to target lumbar stability and strengthening, including the following exercises (Figures 2, 3), doing 4 sets of 12 repetitions per exercise. The total intervention time for the control group was approximately 45 minutes per session.

The experimental group followed the same protocol as the control group with 1 key modification: instead of TENS, participants received IMES (Figure 4), administered through percutaneous acupuncture needles. All IMES procedures were performed by a licensed physical therapist with over 5 years of clinical experience in electrotherapy and certified training in ultrasound-guided needling techniques. IMES was applied via acupuncture needles (0.3 cm in diameter, 5 cm in length) inserted under ultrasound guidance (SonoScape, 5–20 Hz line head, China, 2022) into the multifidus muscles bilaterally at the L4/L5 and L3/L4 levels. The precise placement of the needles was confirmed by real-time ultrasound imaging to ensure that they were placed within the muscle belly of the multifidus, as these muscles play a critical role in spinal stability and postural control. The ultrasound guidance aimed to reduce the risk of improper needle placement and improve safety during the procedure. The electrical stimulation was delivered at an intensity of 1–2 mA, within a frequency range of 50–100 Hz. These settings were chosen to engage motor fibers and sensory afferents for a therapeutic effect on muscle tissue. The stimulation duration was set to 15 minutes per session, with the goal of promoting muscle relaxation, improving muscle perfusion, and reducing myofascial pain. The device used was the Enraf-Nonius TensMed S84-1727911 (Spain, 2023), specifically equipped with needle attachments for percutaneous application. This device allows for precise control over stimulation parameters and was used for direct muscle activation through the acupuncture needles.

A detailed register of adverse events was kept for all patients during IMES application. Participants were closely monitored for any adverse reactions, including needle-site infection, excessive pain, or skin irritation using an ad hoc form. If adverse effects occurred, such as pain exacerbation or discomfort during needle insertion, treatment modifications were implemented. These included adjustment of needle depth, reduction in stimulation intensity, or reapplication at a different anatomical site to avoid further discomfort. In cases of significant adverse effects, the IMES application was discontinued for that session, and the patient was provided with alternative pain management before being excluded from the group.

Both groups underwent therapy sessions twice per week for 6 weeks, totaling 12 sessions, with a 3-day interval between sessions (eg, Monday–Thursday or Tuesday–Friday). This schedule was implemented to ensure adequate recovery time between sessions and to avoid overloading the tissues.

In addition to the therapy sessions, participants were instructed to perform the prescribed exercise program (outlined above) at least 3 times per week for 15 minutes between sessions, in line with best practices for rehabilitation of chronic low back pain, doing 4 sets of 12 repetitions per exercise. To ensure consistency in treatment effects, participants were advised to refrain from excessive physical activity between sessions.

MEASUREMENTS:

To maintain consistency, all measurements and treatments were administered by qualified physiotherapists and randomly chosen for each participant. Examinations took place between 10: 00 AM and 12: 00 PM at the Provita Medical Center in Żory, Poland, located at Zjednocznej Europy Street 37. The study period was from January to March 2025. Environmental conditions were kept uniform, with a temperature of 21°C and relative humidity at 40–45%, monitored automatically using the HVAC system.

TISSUE PERFUSION: Tissue perfusion was assessed using a Perimed device (Sweden, 2004). Laser Doppler flowmetry measurements were performed at a skin tissue volume of 1 mm3 and a depth of 2.5 mm, using 2 contact laser probes placed on labeled manual trigger point therapy. Laser Doppler flowmetry uses emitted radiation that penetrates deep into the tissue, where it encounters moving blood cells, causing a change in their oscillation frequency due to the Doppler effect, as previously explained [32]. The test was conducted in standard positions according to the protocol described in the scientific literature [33]. The unit of measurement used was the perfusion unit, which is an arbitrary unit representing the relative blood flow in the measured tissue volume. Higher perfusion unit values indicate greater blood flow, suggesting better tissue perfusion, while lower perfusion unit values suggest reduced blood flow and potentially compromised tissue health.

MUSCLE TENSION AND STIFFNESS: After determining the manual trigger points and marking the measurement site on the upper part extensor spine with a marker, measurements of the biomechanical properties of the muscle, specifically muscle tension and muscle stiffness, were performed using the MyotonPRO myotonometer (AS, Myoton Ltd, Estonia 2021). The MyotonPro is a digital device with a body and a push-in probe (3 mm diameter). The scientific literature confirms its reliability and repeatability [34]. Initially, an initial pressure of 0.18 Newtons (N) is applied to the surface through the probe, compressing the tissue beneath. Subsequently, the device releases a mechanical impulse (0.4 N, 15 milliseconds), deforming the tissue for a brief period [35]. The device evaluates resting muscle tension based on the frequency of muscle vibrations at rest, while muscle stiffness is determined based on the tissue’s resistance to deformation, calculated using a logarithmic scale. Measurements were taken at 4 spinal extensors, 2 on the left and 2 on the right sides, in the projection of the transverse processes of the L4 and L5 spinal vertebrae. Muscle tension was measured in Hertz (Hz), where higher Hz values indicate greater muscle tension, representing the frequency of muscle vibrations at rest. Muscle stiffness was measured in Newtons per meter (N/m), where higher N/m values indicate greater muscle stiffness, representing the tissue’s resistance to deformation.

PRESSURE PAIN THRESHOLD: The PPT was measured using an algesimeter (FDIX, Wagner Instruments, Greenwich, CT, USA 2013). Participants underwent 3 pressure tests using a probe with a 4-mm radius in a specific tissue area. The force value, PPT, was calculated as the average of the 3 measurements and displayed digitally. The device signaled the need to repeat the test if there was a significant deviation among the measured values. The pressure was gradually increased until the stimulus became unpleasant for the participant. This tool has been validated as a reliable method for both diagnosis and treatment assessment of myofascial pain syndromes and demonstrates high reliability in repeated measures [36,37]. Measurements were performed in the treatment position. The PPT was measured in Newtons per square centimeter (N/cm2), representing the amount of pressure required to elicit pain. Higher PPT values indicate a higher pain tolerance, meaning more pressure is needed to elicit pain, while lower PPT values suggest a lower pain tolerance, indicating that less pressure is needed to elicit pain.

NUMERIC RATING SCALE: The NRS was used in assessing low back pain. The NRS has been validated in various contexts, including its use in patients using pain medications, where it can provide insight into the perceived effects of pain management strategies [38]. The NRS is a scale from 0 to 10, where 0 represents ‘no pain’ and 10 represents ‘the worst pain imaginable’. Higher NRS scores indicate greater pain intensity, while lower scores indicate less pain.

RANGE OF MOTION:

The spinal flexion test, often called the finger-to-floor test, is a clinical assessment of lumbar flexibility and spinal mobility. This test measures the distance from the fingertips to the floor during forward bending, providing insight into spinal function and potential chronic low back pain issues. Participants were asked to bend forward twice, holding the bend for 3 seconds, and then the measurements were averaged. The finger-to-floor distance was measured in centimeters (cm), representing the distance from the fingertips to the floor during maximum forward flexion. Smaller finger-to-floor values indicate greater spinal flexibility, as the fingertips are closer to the floor, while larger finger-to-floor values suggest reduced spinal flexibility.

POSTURAL STABILITY: Stylometric platforms were used to assess postural stability in individuals with low back pain. These platforms quantitatively measure the area of displacement of the center of gravity, providing reliable insight into the relationship between pain intensity and postural control [39]. Our research used the E.P.S./R2 pedobarograph (Italy 2019) with Biomech Studio software. The volunteers held still for 20 seconds for the measurement. Two measurements were taken each time, and the results were averaged. The pedobarograph measured the Center of Pressure (CoP) displacement, typically expressed in millimeters (mm) or centimeters (cm), and sway velocity, measured in millimeters per second (mm/s) or centimeters per second (cm/s). Larger CoP displacement and higher sway velocity values indicate greater instability and poorer postural control, reflecting a wider range of movement and faster shifts in the body’s center of gravity.

MAXIMAL VOLUNTARY CONTRACTION: The Kinvent K-Force Push v3 (Italy, 2022) dynamometer used in our studies provides detailed maximal voluntary contraction measurements and a repeatable and reliable measurement method [40]. The measurements were taken in a standardized forward lying position. To determine the maximal voluntary contraction of the lumbar extensors, participants adopted a standardized forward lying position on a plinth or examination table. The Kinvent K-Force Push v3 dynamometer was positioned to apply resistance directly against the lower back, specifically targeting the lumbar extensor muscles. Each participant performed 3 maximal voluntary isometric contraction trials. For each trial, the participant was instructed to lie prone (face down) on the examination table with their hips extended and feet unsupported or lightly supported to ensure isolation of the lumbar extensors. The dynamometer was firmly secured against their lower back, ensuring it was perpendicular to the direction of force application. The participant was given clear instructions to exert their maximal effort to push against the dynamometer. Upon a verbal command (“Push!”), the participant was instructed to gradually increase their force against the dynamometer over approximately 2–3 seconds until they reached their maximal effort, and then hold this maximal contraction for a sustained period of 3–5 seconds. Verbal encouragement was provided during the contraction to elicit maximal effort. After the sustained contraction, the participant was instructed to slowly relax. A rest period of 60 seconds was provided between each of the 3 trials to minimize fatigue and ensure maximal effort in subsequent trials. The Kinvent K-Force Push v3 dynamometer automatically recorded the peak force in Newtons (N) achieved during each maximal contraction. The highest force value attained across the 3 trials was taken as the participant’s maximal voluntary contraction (MVC) for the lumbar extensors. The lumbar extensor strength was measured in Newtons (N), representing the maximum force exerted by the lumbar extensor muscles during voluntary contraction. Higher N values indicate greater lumbar extensor strength, suggesting better muscle function, while lower N values suggest reduced strength and potential muscle weakness.

STATISTICAL ANALYSIS:

A mixed-design ANCOVA was conducted to assess the effects of time (baseline, after treatment, and 4 weeks after treatment) and group (control vs experimental) on the dependent variable. Baseline values were included as a covariate to account for initial differences in each outcome measure, thereby controlling for potential baseline disparities and ensuring a more accurate comparison between groups over time. The assumptions of ANCOVA, including normality (assessed via Shapiro-Wilk tests), homogeneity of variances (Levene’s test), homogeneity of regression slopes, and sphericity (Mauchly’s test), were verified. In cases where sphericity was violated, Greenhouse-Geisser corrections were applied. Partial eta squared (np2) was reported as a measure of effect size. Significant main effects and interactions were further explored using Bonferroni-adjusted pairwise comparisons. Statistical analyses were conducted using IBM SPSS Statistics (Version 28.0., Armonk, NY: IBM Corp), with significance set at P<0.05.

Results

MUSCLE STIFFNESS:

The interaction between time and baseline muscle stiffness of the right leg was not statistically significant (F_(1,26)=0.508, P=0.482), with a small effect size (partial η2=0.019). There was no statistically significant interaction between time and group (experimental vs control) on muscle stiffness of the right leg (F_(1,26)=3.234, P=0.084). The effect size was medium (partial η2=0.111).

The interaction between time and baseline muscle stiffness of the left leg was not statistically significant (F_(1,26)=1.016, P=0.323), with a small effect size (partial η2=0.038). There was no statistically significant interaction between time and group (experimental vs control) on muscle stiffness of the left leg (F_(1,26)=2.060, P=0.163), with a small to medium effect size (partial η2=0.073).

MUSCLE TENSION:

The interaction between time and baseline muscle tension of the right leg was statistically significant (F_(1,26)=12.949, P=0.001), with a large effect size (partial η2=0.332). There was no statistically significant interaction between time and group (experimental vs control) on muscle tension of the right leg (F_(1,26)=1.721, P=0.201), with a small effect size (partial η2=0.062).

The interaction between time and baseline muscle tension of the left leg was not statistically significant (F_(1,26)=0.001, P=0.972), with a trivial effect size (partial η2=0.000). There was no statistically significant interaction between time and group (experimental vs control) on muscle tension of the left leg (F_(1,26)=0.044, P=0.836), with a trivial effect size (partial v2=0.002).

TISSUE PERFUSION:

The interaction between time and baseline tissue perfusion of the right leg was statistically significant (F_(1,26)=10.602, P=0.003), with a large effect size (partial η2=0.290). There was a highly statistically significant interaction between time and group (experimental vs control) on tissue perfusion of the right leg (F_(1,26)=39.712, P<0.001), with a very large effect size (partial η2=0.604).

The interaction between time and baseline tissue perfusion of the left leg was statistically significant (F_(1,26)=15.572, P=0.002), with a very large effect size (partial η2=0.545). There was a highly statistically significant interaction between time and group (experimental vs control) on tissue perfusion of the left leg (F_(1,26)=55.029, P<0.001), with an exceptionally large effect size (partial η2=0.679).

PRESSURE PAIN THRESHOLD:

The interaction between time and baseline PPT of the right side was not statistically significant (F_(1,26)=0.644, P=0.429), with a small effect size (partial η2=0.024). There was no statistically significant interaction between time and group (experimental vs control) on PPT of the right side (F_(1,26)=0.503, P=0.484), with a small effect size (partial η2=0.019).

The interaction between time and baseline PPT of the left side was statistically significant (F_(1,26)=7.287, P=0.012), with a large effect size (partial η2=0.219). There was no statistically significant interaction between time and group (experimental vs control) on PPT of the left side (F_(1,26)=3.441, P=0.075), although it approached significance. The effect size was medium (partial η2=0.117).

NUMERIC RATING SCALE:

The interaction between time and baseline NRS scores was not statistically significant (F_(1,26)=0.783, P=0.384), with a very small effect size (partial η2=0.029). There was no statistically significant interaction between time and group (experimental vs control) on NRS scores (F_(1,26)=0.395, P=0.535), with a very small effect size (partial η2=0.015).

RANGE OF MOTION:

The interaction between time and baseline ROM was not statistically significant (F_(1,26)=0.425, P=0.520), with a small effect size (partial η2=0.016). There was no statistically significant interaction between time and group (experimental vs control) on ROM (F_(1,26)=0.040, P=0.843), with a negligible effect size (partial η2=0.002).

MAXIMAL VOLUNTARY CONTRACTION:

The interaction between time and baseline MVC was not statistically significant (F_(1,26)=1.477, P=0.235), with a small effect size (partial η2=0.054). There was no statistically significant interaction between time and group (experimental vs control) on MVC (F_(1,26)=3.094, P=0.090), with a medium effect size (partial η2=0.106).

POSTURAL STABILITY:

The interaction between time and baseline postural stability was not statistically significant (F_(1,26)=0.433, P=0.516), with a very small effect size (partial η2=0.016). There was a statistically significant interaction between time and group (experimental vs control) on postural stability (F_(1,26)=6.800, P=0.015), with a large effect size (partial η2=0.207).

ADVERSE EFFECTS:

In the experimental group, the following adverse effects were reported during the treatment protocol: 3 cases of increased pain, 2 cases of burning, 1 case of bleeding, and 2 cases of fainting.

Discussion

This randomized controlled trial aimed to compare the effectiveness of a 6-week IMES protocol targeting the multifidus muscle in patients with chronic low back pain, relative to TENS, when both were combined with standard therapies. The results suggest that IMES was more effective in significantly reducing muscle tension on the right side, increasing tissue perfusion bilaterally at post-treatment, and improving the pressure pain threshold on the left side. Additionally, IMES led to a greater reduction in pain intensity, as well as improvements in postural stability and lumbar flexion range of motion, compared to conventional physiotherapy with TENS. While IMES and TENS were not applied in isolation, and therefore no causal relationship can be inferred since they were part of a combined therapy, the findings suggest potential benefits of IMES over TENS when used as a complement to standard therapy focusing on the multifidus muscle.

The improvement in muscle tone, particularly in the right-side multifidus, both immediately post-treatment and at the 4-week follow-up, is consistent with previous research on the efficacy of electrotherapeutic modalities in modulating hypertonicity and improving neuromuscular function [18,41]. These findings are particularly important since the multifidus muscle plays a key role in spinal stability. The reduction in muscle tension, particularly on the right side, may be aligned with the gate control theory of pain, where sensory-level stimulation may interfere with the transmission of nociceptive signals at the spinal cord level, thus possibly providing analgesic effects [42].

The significant improvements observed in tissue perfusion and pain sensitivity in the experimental group also suggest that IMES can induce local microvascular adaptations, resulting from increased muscular activity and vasodilation induced by electrical stimulation, potentially contributing to enhanced healing and pain relief [43,44]. The increase in PPT supports this notion, with the assumption that IMES may decrease nociceptor excitability and modulate central sensitization mechanisms, thus possibly offering both peripheral and central analgesic effects [45].

An interesting finding from this study was the differential effects observed between the right and left sides. The experimental group showed a greater reduction in muscle tension on the right side, but similar benefits were not seen on the left side for muscle stiffness and tension. One potential explanation could be the dominance of right-sided dysfunction or asymmetry in the study population, which may have influenced the response to treatment [46].

The lack of significant changes in muscle stiffness and maximal voluntary contraction between the groups suggests that IMES exerts a more profound effect on neuromuscular control and sensory modulation than on structural adaptations or peak force production. While maximal voluntary contraction values increased slightly post-treatment in both groups, the fact that IMES was applied at a sensory level may explain the lack of substantial muscle strengthening. Sensory-level stimulation primarily focuses on modulating pain and improving proprioception rather than increasing strength. This finding aligns with previous research suggesting that while IMES may improve recruitment patterns, its role in improving muscle strength is limited [47,48]. Therefore, IMES should be viewed primarily as an adjunctive therapy for pain relief and neuromuscular control, rather than a direct muscle-strengthening intervention.

The improved range of motion observed in the IMES group is consistent with the notion that pain reduction and muscle relaxation facilitate greater flexibility and functional movement [49]. Despite the absence of a direct stretching component in the IMES protocol, the sensory-level stimulation may have contributed to a reduction in muscle hypertonicity, which in turn improved flexibility [50,51]. However, the lack of sustained improvements in tissue perfusion and PPT at the 4-week follow-up suggests that longer or more frequent treatment protocols may be necessary to achieve sustained therapeutic effects.

The adverse effects reported in the IMES group, including increased pain, burning, bleeding, and fainting, highlight the need for appropriate patient screening and gradual dose progression. While these adverse effects were self-limiting, they emphasize the importance of clinician expertise in the application of electrotherapy protocols, especially when working with high-risk patient populations.

This study has several limitations that must be addressed in future research. A limitation of this study is the use of convenience sampling for participant recruitment, which may affect the generalizability of the findings. The small sample size limits also the generalizability of the findings. Additionally, the inclusion of multiple interventions in both treatment groups precludes the ability to exclusively attribute outcomes to the IMES intervention. The results should thus be interpreted with caution, as the observed benefits may also be influenced by the concurrent use of exercise and myofascial trigger point therapy. Future studies should include larger sample sizes, longer-term follow-up assessments to evaluate the durability of treatment effects, and electromyographic analyses to better understand the neuromuscular mechanisms underlying the IMES intervention. Additionally, studies comparing IMES with other electrotherapeutic modalities and exploring its integration with strength training protocols may further improve its clinical application. Finally, side-specific responses to treatment, especially the observed differences between the right and left sides, should be investigated further to understand any potential asymmetries in treatment efficacy.

Conclusions

This randomized controlled trial suggests that a 6-week IMES protocol targeting the multifidus muscle, when combined with conventional physiotherapy, offers benefits over treatment with transcutaneous electrical nerve stimulation (TENS) for managing chronic low back pain. While it is not possible to definitively assign a causal effect to IMES alone, given that it was applied in conjunction with standard treatment, the comparison with TENS suggests that IMES, as part of a comprehensive rehabilitation program, led to significant improvements. Specifically, there were significant reductions in muscle tension on the right side, increases in bilateral tissue perfusion, and improvements in PPT on the left side. Additionally, participants reported greater reductions in pain intensity and improvements in postural stability and lumbar flexion range of motion when compared to conventional physiotherapy combined with TENS. However, no significant changes were observed in muscle stiffness or maximal voluntary contraction. These findings suggest that IMES, in conjunction with other rehabilitation techniques, may offer advantages over TENS in enhancing pain modulation, neuromuscular function, and mobility in individuals with chronic low back pain.