Targeted Muscle Reconditioning with HIFEM Technology for Low Back Pain: A Retrospective Study on Pain, Function, and Disability

Abstract

Background Context

Low back pain (LBP) affects over 619 million people globally and is the leading cause of disability. Most cases are non-specific and linked to muscle impairment, yet effective, non-invasive treatments remain limited.

Purpose

To evaluate the impact of back-targeted electromagnetic muscle stimulation (HIFEM) therapy on pain, disability, and function in patients with LBP.

Design

Uncontrolled retrospective cohort study.

Patients and methods

Patients (n=25; 72% female) with LBP who received a study device treatment were 33–76 years old. Because treatment was tailored to individual needs, many patients received therapies to other body regions. Medical records of patients who received at least one study device treatment on low back between April 1, 2024 and April 30, 2025 and completed Visual Analog Scale (VAS) before and after the last treatment were analyzed, with a 1-month (±14 days) interval between baseline and post-treatment assessments.

Outcome Measures

The outcome measures included VAS, Pain Disability Questionnaire (PDQ), Oswestry Disability Index (ODI), Magnetic resonance imaging (MRI) findings, and iliopsoas muscle strength measurements.

Results

Significant absolute and percentage reductions were observed in VAS (mean change: −5.2 points; −75.7%, 95% CI [−83.4, −63.1]), PDQ (−49.3 points; −66.1%, 95% CI [55.1, 80.9]), and ODI (−11.9 points; −61.0%, 95% CI [44.7, 78.7]) scores (all p < 0.0001). The reductions in VAS, PDQ and ODI scores surpassed established minimal clinically important differences (MCID). Iliopsoas muscle strength improved by 124.8% post-treatment in the subgroup treated for the core.

Conclusion

HIFEM therapy was associated with a significant reduction of pain, disability, and the impact of LBP on daily functioning while improving core muscle strength, offering a non-invasive option for managing LBP.

Plain Language Summary

Millions of people around the world live with LBP, and many face ongoing muscle weakness that current medications and treatments often fail to address, focusing instead on masking symptoms. The author wanted to find out whether actively strengthening the back and core muscles, using a specialized technology, could offer people meaningful relief. The author reviewed medical records for 25 people who were treated for LBP using a non-invasive electromagnetic muscle stimulation (HIFEM) treatment. This technology stimulates and strengthens deep core and back muscles. After treatment, people reported a 75.7% drop in their pain levels, described improvements in their daily functioning, and experienced less disability in daily life. Muscle strength tests also showed a substantial increase in core muscle strength following treatment. These findings suggest that strengthening core muscles with HIFEM technology is associated with reduced back pain. For people looking for a non-surgical option, this approach may help rebuild physical strength and support daily functioning. Further research involving a comparison group is needed to confirm these findings.

Introduction

Low back pain (LBP) is the leading cause of disability worldwide, with an estimated 619 million people affected as of 2019, a staggering increase from 387 million in 1990.1,2 Projections indicate that this number will surge to 843 million by 2050, creating a public health challenge with profound implications for individuals, healthcare systems, and the global economy.1–3 LBP is especially prevalent among working-age adults, driving absenteeism, reduced productivity, and early retirement.1 In the United States alone, LBP accounts for approximately 2.3% of all physician visits, representing not just a clinical challenge but also a significant financial strain due to repeated treatments, loss of work capacity, and comorbid mental health issues,4,5 highlighting the need for more effective prevention and treatment.

Despite its widespread prevalence, LBP remains a heterogeneous and diagnostically challenging condition. In 90% of cases, the etiology is non-specific or mechanical, with a clear pathoanatomical cause identified in only a small proportion.6 While disc degeneration accounts for an estimated 40% of chronic cases,7 other contributors include facet joint pathology, sacroiliac dysfunction, muscle weakness/impairment, and ligamentous changes.6,7 Notably, muscle dysfunction and atrophy in lumbar and pelvic regions have emerged as key factors in both the development and persistence of LBP, especially in chronic presentations. Poor lumbar-pelvic stability, motor control deficits, and fat infiltration of lumbar musculature are all strongly correlated with increased pain severity and disability, particularly in older adults, where sarcopenia and muscle weakness further compound the issue.8–10 Despite LBP being multifactorial, when comparing the strength level of healthy people and those suffering from LBP, findings show lower strength levels in individuals with LBP and indicate a gradual loss of strength due to recurrent LBP.11 Moreover, a statistically significant decrease in pain severity was observed in the exercise intervention groups relative to the control group.12 Additionally, prolonged physical inactivity, whether due to occupational factors, lifestyle choices, or injury, can lead to muscle disuse, also resulting in atrophy, neuromuscular impairment, and biomechanical imbalances.13–15

Current first-line treatment options for back pain include physiotherapy and pharmacotherapy. While pharmacological interventions such as non-steroidal anti-inflammatory drugs (NSAID), muscle relaxants and opioids offer symptomatic relief, they do not address the underlying cause and can be associated with adverse effects and even risk of dependency.16,17 Based on guidelines, exercise therapy such as strength and resistance training, core stabilization and muscle reconditioning remains the main conservative treatment to address back pain; however, it relies completely on patient adherence and specialist skills.12,18,19 After conservative interventions, injection-based approaches and surgery can be considered, but those, due to their invasive character, carry increased risks and require longer recovery time.18,20 Comorbidity of LBP exceeds the physical domain, with data showing that individuals with LBP are at least twice as likely to suffer from mental conditions such as depression, anxiety, stress, psychosis and sleep deprivation when compared to those without LBP, making this condition even more complex. Therefore, guidelines also recommend considering psychological support to overcome obstacles to recovery when previous treatments have not been effective.18

At the same time, research on passive rehabilitation options in patients with documented lumbar degenerative pathology is limited.21 Novel energy-based modalities such as Emsculpt NEO (BTL Industries, Inc.) utilize basic principles of exercise, but induce muscle contractions beyond typical voluntary exercise capacity through supramaximal contractions via HIFEM with the addition of synchronized radiofrequency, which delivers gentle warming associated with symptom relief and supports tissue regeneration.21–23 HIFEM, a trademarked electromagnetic stimulation technology, has previously been used to restore muscle strength, coordination, and spinal stability and addresses the neuromuscular underpinnings of LBP. This modality has already emerged as a useful approach in the functional wellness field by improving muscle function and reducing pain in the extremities, and an increase in strength of deep core musculature.24–27 It can also potentially serve as an alternative treatment for individuals who are not able to perform physiotherapy exercises due to pain or the mental burden of the LBP condition.21

Back pain has many contributors such as facet joint pathology, sacroiliac dysfunction, muscle ligamentous changes and weakness/impairment. Since muscle weakness is one of them, this study aimed to evaluate the effectiveness of a study device utilizing HIFEM in reducing the severity of LBP. Therefore, this approach is hypothesised to reduce the severity of LBP in the population with lumbar degenerative pathology.

Methods

In this single-center, single-arm, open-label retrospective cohort study, the medical records of existing patients from the investigator’s practice who met the eligibility criteria were analyzed. All individuals who met the inclusion criteria were enrolled and evaluated in the study. The inclusion criteria required age ≥ 22 years, presence of active back pain/ache/discomfort, and/or sarcopenia-induced and/or sarcopenia-predisposed back pain/ache/discomfort, and/or age-related functional limitations of the back. Patients had to complete at least one treatment with the study device applied to the low back between April 1, 2024, and April 30, 2025, and complete the Visual Analog Scale (VAS) questionnaire before and after the last treatment in order to be included in this retrospective cohort study. Patients with incomplete demographic data or concurrent noninvasive back pain treatments during the study period were excluded. Selection bias was minimized by consecutively including all eligible patients treated during the study period, regardless of outcome. The study protocol was submitted for review to an Advarra Institutional Review Board and was granted an exemption status under 45 CFR 46.104(d)(4) from IRB oversight (Pro00087455), as it involved secondary analysis of abstracted data. This retrospective study was not registered in a clinical trials registry, as it involved analysis of pre-existing medical records rather than a prospective interventional trial. This study was conducted in accordance with the ethical principles of the Declaration of Helsinki.

Treatment Schedule

In accordance with the study protocol, all enrolled subjects received a minimum of one 30-minute treatment session targeting the low back. While the inclusion criterion required only one treatment session, this reflects a minimum eligibility threshold rather than the treatment actually delivered. The reported mean of four sessions (range 3–6) describes the treatment received by the analyzed cohort. Treatment sessions were scheduled within a 1-month period (±14 days). Depending on individual clinical needs, medical history, and in line with a holistic approach, patients received additional treatments targeting other regions of the body, including the abdomen, quadriceps, hamstrings, gluteal muscles, calves, and/or the pelvic floor muscles. On average, subjects received 16 HIFEM treatments in total.

An exploratory subgroup of six patients underwent comprehensive body core therapy, which included treatments to the back, abdomen, and pelvic floor. Treatments to the back and abdomen were delivered using the EMSCULPT NEO device, while pelvic floor muscle (PFM) stimulation was performed using the EMSELLA device (BTL Industries, Inc.), which also utilizes HIFEM technology.

Among the subgroup receiving core-focused therapies, each patient underwent an average of four treatment sessions for the back (ranged from 3 to 5), four for the abdomen (ranged from 2 to 6), and four for the PFM (ranged from 3 to 8).

Data Collection & Evaluation

The retrospective chart analysis involved a systematic review and abstraction of existing medical records for patients previously treated with the investigated device at the clinical site. Only pre-existing data documented in the pre-specified period were considered for analysis. Data were abstracted using a standardized filling template, also reviewed by the IRB. Each record that passed the eligibility screening was assigned a unique identifier to maintain subject anonymity. A total of 25 subjects (n=25) were screened and were included in the study. Abstracted data included: demographics information, baseline diagnosis, baseline magnetic resonance imaging (MRI) lumbar spine pathology grading (Pfirrmann, Goutallier, Schizas, Lee, Lateral Recess Stenosis (LRS), relevant comorbidities, number of treatment sessions, dynamometer muscle strength scores (pre- and post-treatment), VAS scores (pre- and post-treatment), Pain Disability Questionnaire (PDQ) scores (pre- and post-treatment), Oswestry Disability Index (ODI) scores (pre- and post-treatment), and safety data, such as adverse events, complaints associated with the treatment, and other clinical observations. While the ODI was used to evaluate condition-specific physical impairment related to spinal mechanics, the PDQ was utilized to assess global pain disability, including its broader psychosocial and emotional impact. Assessments were performed at baseline and directly after the last treatment was completed with 1-month (±14 days) period in between assessments.

Baseline structural and degenerative severity were assessed on spine MRI using the Pfirrmann, Goutallier, Schizas, LRS, and Lee grading systems by the treating physician, however, formal inter-rater reliability testing was not conducted, which we acknowledge as a limitation. The Pfirrmann grading system was designed to evaluate lumbar intervertebral disc degeneration. It comprises five grades, with increasing severity reflecting progressive structural deterioration of the disc.28 The Goutallier classification assesses fatty infiltration within muscle tissue, also using a five-grade scale in which higher grades correspond to greater degrees of fatty degeneration.29 The Schizas classification was established for the assessment of lumbar central canal stenosis. It incorporates seven categories (A1-D) determined by cerebrospinal fluid (CSF) visibility and cauda equina morphology, with higher grades indicating advanced canal compromise.30 The Lee grading system was developed to characterize lumbar foraminal stenosis. It defines four grades of foraminal narrowing, whereby increasing grades correspond to progressive obliteration of perineural fat and eventual compression of the exiting nerve root.31 Finally, the LRS grading system was introduced to evaluate impingement of the traversing nerve root within the lateral recess. This system includes four grades, with higher categories reflecting greater displacement and compression of the nerve root.32 These morphologic assessments were performed exclusively at baseline, as they reflect chronic anatomical changes that are not expected to remodel or alter significantly over the study period.

The questionnaire-related outcomes were based on averaging the pre-treatment score and comparing it to the post-treatment score, while observing the difference between those two scores for the whole population treated. The proportion of patients falling into each pain/disability severity category was calculated both before and after treatment, based on the classifications defined below.

The VAS assesses the intensity of pain using an 11-point scale ranging from 0 to 10, where higher scores indicate greater pain severity. Pain levels are grouped as follows: The “0” corresponds to no pain present, scores of 1 to 3 indicate mild pain, 4 to 6 moderate pain, and scores equal to or greater than 7 severe pain.33

The PDQ evaluates the extent of a disability associated with pain. It is a 15-item tool with a score ranging from 0 to 150. Higher scores reflect a more significant disability level. Severity is classified as: 0–70 indicates mild to moderate severity, 71–100 reflects severe cases, and 101–150 represents extreme severity.34

The ODI consists of 10 questions that assess the impact of LBP on daily life and functional capacity. Based on the total score, the following categories were developed: 0–20, minimal disability; 21–40, moderate disability; 41–60, severe disability; 61–80, crippling back pain; 81–100, bed-bound or patient symptom exaggeration.35

Dynamometer (microFET® 2, HogganScientific, LLC) strength results are reported for the subjects and particularly for subjects who received treatment targeting all three core regions - the abdomen, back, and pelvic floor - given that the strength of an important core stabilizer,32 the iliopsoas muscle, was measured and evaluated. A “make” test protocol was utilized, where investigator held the dynamometer stationary while the subject exerted maximal force against the sensor. Subjects were seated in an upright position on the edge of a stable examination table with their hips and knees flexed to 90 degrees and their legs hanging freely. To isolate the targeted musculature and minimize compensatory pelvic or trunk movements, subjects were instructed to grip the edge of the table with both hands and maintain a neutral, upright posture. The dynamometer, fitted with a curved transducer pad, was positioned on the anterior aspect of the distal thigh, approximately 5 centimetres proximal to the superior pole of the patella. Subjects were instructed to perform a maximal isometric hip flexion contraction by driving their knee vertically upward directly into the device. Moreover, the Symmetry Index (SI) was calculated for each subject pre- and post-treatment following the formula proposed by Robinson, Herzog W and Nigg,36 as it is the most commonly used index:37

where F_R represents the strength of the right iliopsoas muscle, and F_L represents the strength of the left iliopsoas muscle.

For subgroup analysis, acute and chronic LBP groups were identified and analyzed. The acute group was defined as subjects with pain or condition onset in less than 3 months pre-treatment, and the chronic group included subjects with pain or condition that persisted more than 3 months.38

Minimal Clinically Important Difference (MCID)

To contextualize the clinical relevance of observed changes beyond statistical significance, percentage and/or absolute changes in each outcome measure were compared against previously established minimal clinically important difference (MCID) thresholds. For VAS, an improvement of 30–33% from baseline was considered the MCID.39 For ODI, both a relative improvement of 7–15%, as established in post-lumbar-surgery populations, and an absolute reduction of 10 points, a threshold commonly used to define clinically meaningful improvement in LBP, were used as reference benchmarks.35,40 For PDQ, an absolute reduction of 20 points was used as the MCID threshold.34 The proportion of subjects meeting or exceeding the respective MCID was calculated for each outcome measure and reported as a responder analysis.

Statistical Considerations

To present the results, mean values together with a standard deviation and 95% confidence intervals were calculated and reported. To compare pre- and post-treatment results, the % change in the score/parameter was calculated. When applicable, the proportion of subjects was calculated by dividing the number reporting the condition or characteristics by the total number who completed the evaluation. Statistical testing was conducted using specialized statistical software. Firstly, the normality of the sample was determined using the Shapiro–Wilk test. Secondly, the pre- and post-treatment results were analyzed using the appropriate paired statistical test, either paired t-tests for normally distributed samples, or Wilcoxon matched-pairs signed rank tests for non-normally distributed samples. The lateral muscle strength comparison was also performed with a paired sample test. Moreover, a correlation analysis was performed to investigate the relationship between the changes in pain severity, disability levels, and pain-related influence on daily life and functioning post-treatment. The Spearman’s coefficient is reported as a result of the correlation analysis, due to the character of the investigated data. The significance threshold for all statistical testing was set to α = 5%. Because this was a retrospective analysis of existing medical records, the sample size was determined by the number of eligible subjects treated within the defined study period rather than an a priori statistical power calculation.

The subjects with incomplete PDQ and/or ODI questionnaires at either of the two investigated time points were excluded from the final analysis.

Results

A total of 25 subjects (54.6±12.0 years old, 7 males, 18 females) were screened, met the inclusion and none of the exclusion criteria were retrospectively evaluated. The chart review included subjects 33–76 years old, with BMI ranging from 18.6–41.2 kg/m²; the average BMI for the evaluated cohort equalled 28.1±5.2 kg/m². The first-listed diagnosis for all (n=25) subjects was LBP. As per the comorbidities, muscle weakness and muscle spasms of the gluteal muscles and/or lower extremities were the most commonly observed alongside the LBP, in over half of the cases (see Figure 1).

Figure 1.

Comorbidities observed alongside low back pain at baseline in the analyzed patients (n=25).

44% (11/25) of subjects presented with acute LBP, whereas the remaining 56% (14/25) suffered from a chronic condition.

Baseline MRI Lumbar Pathology

The grading systems (Table 1) suggested a predominance of advanced disc degeneration, with most subjects graded at the higher end of the Pfirrmann scale. Paraspinal muscle fatty infiltration was also frequent, with the majority classified in the moderate-to-severe range according to the Goutallier system. In contrast, central canal stenosis was uncommon, as most subjects remained in the lower Schizas grades, with no cases of severe stenosis observed. On the other hand, lateral recess and foraminal narrowing were more prevalent, with a substantial proportion of subjects demonstrating higher-grade changes on both the LRS and Lee classifications.

Table 1.

MRI Pathology in the Lumbar Region, Collected Before Treatment, According to Several Grading Systems

Proportion of evaluated patients reporting the grade (the highest grade reported)
Pfirrmann grading system
Grade	I	II	III	IV	V
Share of patients	0% (0/25)	12% (3/25)	8% (2/25)	56% (14/25)	24% (6/25)
Goutallier grading system
Grade	0	I	II	III	IV
Share of patients	0% (0/25)	4% (1/25)	56% (14/25)	20% (5/25)	20% (5/25)
Schizas grading system
Grade	None	A	B	C	D
Share of patients	24% (6/25)	64% (16/25)	4% (1/25)	8% (2/25)	0% (0/25)
Lateral Recess Stenosis (LRS) grading system
Grade	0	1	2	3
Share of patients	20% (5/25)	28% (7/25)	20% (5/25)	32% (8/25)
Lee grading system
Grade	0	1	2	3
Share of patients	12% (3/25)	44% (11/25)	20% (5/25)	24% (6/25)

Notes: Black shaded cells indicate that the category is not applicable, as the Lee and Lateral Recess Stenosis (LRS) grading systems comprise only four grades.

Visual Analog Scale (VAS)

All subjects (n=25) completed the VAS assessment, with an average baseline score of 6.9 ± 1.8, indicating moderate to severe pain. 60% (15/25) of subjects reported severe pain (VAS ≥ 7) at baseline. After treatment, the average pain score dropped significantly to 1.7 ± 1.7 (p < 0.0001), reflecting only mild pain. This represents a 75.7% reduction in back pain based on VAS scores (95% CI [−83.4%, −63.1%]). Overall, 96% (24/25) of the cohort met or exceeded the MCID threshold of 33% improvement. When stratified by symptom duration, subjects with acute pain saw their average baseline scores decrease from 6.5 ± 2.2 (moderate-to-severe pain) to 1.6 ± 2.0 (mild pain) following treatment. Similarly, subjects presenting with chronic pain reported a substantial reduction from a baseline average of 7.3 ± 1.4 (severe pain) to 1.7 ± 1.4 post-treatment, successfully transitioning them into the mild pain category.

All subjects who initially reported pain showed improvement, with their post-treatment pain levels reduced to moderate, mild, or none (see Figure 2). Over 80% of subjects reported mild pain or no pain following treatment.

Figure 2.

Visual Analog Scale - A proportion of patients (n=25) reporting a score of corresponding pain severity.

Pain Disability Questionnaire (PDQ)

Three subjects did not complete the post-treatment PDQ assessment and were excluded from the analysis. Among the 22 who completed both evaluations, the average baseline PDQ score was 74.6 ± 35.3, indicating severe disruption of daily life and well-being. At the outset, 54.5% (12/22) of these subjects fell within the severe-to-extreme disability range (PDQ ≥ 71). Following treatment, average scores dropped to 25.3 ± 22.8 (p < 0.0001, 95% CI [16.3, 34.4]), indicating only mild to moderate disability—a 66.1% (95% CI [−80.9, −55.1]) reduction, reflected in decreased pain-related functional and psychosocial disability. Overall, 81.8% (18/22) of the cohort successfully met the MCID of 20 points. When stratified by symptom duration, the acute pain subgroup demonstrated a 72.2% improvement, as their average baseline score decreased from 69.4 ± 39.0 (upper end of mild-to-moderate disability) to 19.3 ± 24.2 (mild disability) after treatment. The chronic pain subgroup exhibited a comparable 61.7% improvement, with baseline scores falling from an average of 79.0 ± 33.1 (severe disability) down to 30.3 ± 21.4 (mild-to-moderate disability) post-treatment.

Notably, every subject who began with extreme disability improved. Post-treatment, their scores no longer met the threshold for severe disability, reducing to mild or moderate (Figure 3).

Figure 3.

Pain Disability Questionnaire - A proportion of patients (n=22) reporting a score corresponding to the functional and psychosocial disability level.

Oswestry Disability Index (ODI)

Three subjects did not complete the post-treatment ODI assessment and one did not complete either the pre- or the post-treatment assessment; all 4 subjects with incomplete assessments were excluded from the analysis. Among the 21 who completed both evaluations, the average baseline ODI score was 19.5 ± 11.0. At the outset, 47.6% (10/21) of these subjects fell within the moderate-to-severe disability range (ODI ≥ 21). Following treatment, average scores dropped to 7.6 ± 7.5 (p < 0.0001, 95% CI [4.6, 11.0]), corresponding to a 61.0% (95% CI [−78.7, −44.7]) reduction.

A clinically meaningful improvement, defined by the 10-point MCID, was achieved by 52.4% (11/21) of the cohort. When evaluating outcomes by symptom duration, the acute pain subgroup demonstrated a 75.7% reduction in mechanical disability, with average scores decreasing from 20.1 ± 12.8 at baseline to 4.9 ± 5.8 post-treatment. The chronic pain subgroup exhibited a comparable trend, achieving a 49.3% improvement as their average baseline score of 19.2 ± 10.0 was reduced to 9.7 ± 8.2. Following treatment, both subgroups successfully transitioned into the minimal disability category.

Notably, almost all subjects reported minimal mechanical disability, as indicated by their ODI scores post-treatment (Figure 4).

Figure 4.

Oswestry Disability Index - A proportion of patients (n=21) reporting a score corresponding to the mechanical disability level.

Correlation Analysis

A correlation analysis was performed on the percentage changes in pain-related VAS, PDQ, and ODI scores. For the VAS vs PDQ comparison, data from 22 subjects with complete records were analyzed. For the VAS vs ODI and PDQ vs ODI comparisons, data from 21 subjects were used. Spearman correlation coefficients were as follows: r = 0.85 for VAS vs PDQ, r = 0.75 for VAS vs ODI, and r = 0.78 for PDQ vs ODI. All p-values for these correlations were p < 0.0001, indicating strong and statistically significant associations.

Higher baseline LRS grades demonstrated a statistically significant moderate negative correlation with percentage improvements in both VAS (r = −0.534, p = 0.006) and PDQ (r = −0.439, p = 0.041) scores. A similar directional trend was observed between LRS and ODI improvement, though it did not reach statistical significance (r = −0.354, p = 0.115). Furthermore, higher Lee foraminal stenosis grades exhibited a trending, but non-significant, association with greater VAS score reduction (r = −0.358, p = 0.079). A negative correlation indicates that higher initial radiological grading is associated with a greater reduction (negative percentage change) in clinical scores, denoting more substantial symptomatic improvement.

Conversely, other baseline morphological markers did not significantly correlate with clinical recovery trajectories. Disc degeneration (Pfirrmann), paraspinal muscle fatty infiltration (Goutallier), central canal stenosis (Schizas), and subcutaneous fat thickness demonstrated weak and statistically insignificant associations across all three outcome measures (all p > 0.05). Notably, paraspinal fatty infiltration (Goutallier) displayed a weak positive correlation with disability outcomes, indicating a slight trend toward poorer PDQ (r = 0.293, p = 0.186) and ODI (r = 0.279, p = 0.221) improvements, though these findings remained statistically insignificant.

Iliopsoas Muscle Strength

A total of 23 subjects completed both pre- and post-treatment assessments of iliopsoas muscle strength. At baseline, the mean strength was 17.1 ± 4.5 lb on the right side and 16.3 ± 4.3 lb on the left. Following treatment, mean muscle strength increased by +95.9% on both sides (p < 0.0001; 95% CI: 77.1, 116.8), reaching 32.8 ± 7.6 lb on the right and 32.6 ± 7.8 lb on the left. The mean pre-treatment symmetry index was 7.7 and decreased to 4.8 post-treatment, corresponding to a 37.0% reduction in inter-limb asymmetry.

Subgroup Analysis - Body Core Treatments

Six subjects received body core treatments (65.8±6.8 years old, 3 males, 3 females, with an average BMI of 26.6±4.9 kg/m²). The subgroup comprised individuals aged 57 to 76 years, with BMI values ranging from 18.6 kg/m² to 33.1kg/m^2. Before treatment, a statistically significant strength imbalance between the right and left iliopsoas muscles was detected (p < 0.05), with an average symmetry index of 12.6. Following treatment, the index decreased to 4.4, representing a 65.5% reduction in asymmetry. No significant lateral imbalance was observed post-treatment (p > 0.1). In terms of absolute strength, the iliopsoas muscle showed, on average, an increase of +124.8% post-treatment (p < 0.0005).

Illustrative Case

A 60-year-old male subject (BMI: 26.2 kg/m²) presented with acute LBP, muscle spasms, and facet joint effusion. Imaging revealed a collapsed intervertebral disc space at L5–S1 and stage II fatty infiltration of the paraspinal musculature, as classified by Goutallier. The subject underwent three treatment sessions targeting the low back, four on the abdomen, and four additional treatments directed at the pelvic floor muscles. At baseline, he reported severe LBP with a VAS score of 8. After the last treatment, his pain resolved completely, with a post-treatment VAS score of 0. He also demonstrated a 90.7% reduction in the PDQ score and an 87.5% reduction in the ODI score, indicating substantial improvements in pain-related disability and overall functional status.

Safety

No adverse events, complaints associated with the treatment, and other clinical observations were documented in the analysed records.

Discussion

The findings of this study provide exploratory evidence supporting an association between HIFEM therapy, as a non-invasive intervention, and improvements in pain, function, and disability among subjects with LBP, a highly prevalent musculoskeletal condition with increasing global incidence.1

This study was designed to focus on a mechanism-driven strategy to engage and strengthen musculature. The results showed a significant and clinically meaningful improvement across key measures: VAS pain scores dropped by 75.7%, more than twice the widely accepted MCID of 30–33%.39,41 ODI scores fell by 61.0%, exceeding the MCID range from 7 to 15% established for post-lumbar surgery subjects.42 The average absolute change in ODI was −11.9 points, surpassing the 10-point threshold typically used to define meaningful improvement in LBP cases.43 Regarding the additional outcome, a 66.1% reduction in PDQ scores was observed. Moreover, all subjects with pain (VAS), pain-related physical disability (ODI), and psychosocial and emotional disability (PDQ) observed improvements post-treatment, all suggesting broader improvements in pain severity, functional outcomes, and disability associated with LBP. An additional consideration relevant to interpretation of these findings is the potential contribution of factors unrelated to the intervention itself, particularly within the acute LBP subgroup (44.0% of the cohort). Acute LBP is well recognized to have a high rate of spontaneous improvement independent of treatment, with many subjects experiencing meaningful symptom resolution within weeks to months regardless of intervention. Regression to the mean is a further plausible contributor, as subjects were enrolled based on the presence of active pain or discomfort, meaning enrollment often coincided with a symptomatic flare; some of the observed improvement may reflect a return toward each subject’s own baseline rather than a true treatment effect. Placebo and expectation effects cannot be excluded either, given the absence of a sham comparator and subjects’ awareness that they were receiving an active, technology-based intervention. Notably, however, the chronic LBP subgroup (56% of the cohort) - a population less prone to spontaneous resolution - demonstrated comparable improvements across VAS, PDQ, and ODI. This pattern argues against spontaneous natural history as the sole driver of the observed improvement and is consistent with a genuine contribution from the treatment itself; however, it does not exclude regression to the mean or placebo/expectation effects, which could plausibly operate similarly in both subgroups.

To contextualize these findings against the exercise-based rehabilitation literature, meta-analyses of exercise interventions for chronic low back pain typically report small-to-moderate standardized effect sizes for pain reduction relative to control conditions.12 More broadly, a recent systematic review and meta-analysis of placebo-controlled trials found that only around one in ten commonly used non-surgical and non-interventional treatments for low back pain - including many exercise-based approaches - showed reliable evidence of a clinically meaningful analgesic effect beyond placebo.44 A direct numerical comparison between these figures and the results of the present study is not appropriate: the cited meta-analyses report between-group standardized effect sizes derived from randomized, placebo-controlled trials, whereas the present study reports within-group percentage change from a single-arm, uncontrolled retrospective cohort - different statistical quantities addressing different questions, compounded by substantial heterogeneity in design, population, and outcome measurement across the exercise-rehabilitation literature. This literature context is provided not to claim a favorable comparison, but to note that the broader evidence base for non-surgical, non-interventional LBP treatments shows modest and inconsistent effects, underscoring the need for the prospective, controlled study of HIFEM-based treatment.

The statistically significant positive correlations among improvements in VAS, PDQ, and ODI scores (Spearman r = 0.75–0.85, p <0.0001 for all comparisons) point to a coherent clinical effect: as pain decreases, so does disability, and subjects regain functional and psychosocial capacity. These outcomes are in agreement with previous findings45–47 and support the hypothesis that improving paraspinal muscle activation can mitigate both back pain and its related disability and impact on daily functioning. While strong correlations between VAS, ODI, and PDQ are expected given their overlapping roles in assessing the pain experience, utilizing both disability measures revealed important clinical nuances. Notably, a subset of subjects demonstrated severe global disability on the PDQ despite reporting only mild spine-specific physical impairment on the ODI. This divergence highlights the multidimensional nature of chronic pain: while a subject’s mechanical lumbar function and ability to perform specific physical tasks may remain relatively intact (captured by the ODI), the broader psychosocial impact, emotional distress, and generalized interference of chronic pain can still be profoundly debilitating (captured by the PDQ).

Iliopsoas muscle strength analysis adds further weight to this mechanism. Subjects experienced an average 95.9% increase in iliopsoas strength (124.8% increase in the body core treatment cohort), alongside a marked reduction in lateral asymmetry, confirmed by a 37.0% reduction in the symmetry index (65.5% asymmetry reduction in the body core treatment group).37 These improvements likely reflect enhanced lumbopelvic stability and reduced compensatory strain on adjacent structures. Balanced activation of the core musculature not only improves postural control but may also reduce the risk of further degeneration or recurrence.48,49 Additionally, the enhanced results of the body core treatment subgroup support the integration of pelvic floor muscle activation, induced by HIFEM, as a relevant component of comprehensive core rehabilitation. Given the role of pelvic muscles and sacral ligaments in stabilizing the sacroiliac joint,50 this approach could have significant implications for a broader subject population experiencing both axial and referred pain.

MRI data from selected subjects revealed the multifactorial nature of their LBP, not just structural degeneration, but also significant muscle tissue impairment, evidenced by elevated Goutallier grades indicating fatty infiltration of the paraspinal muscles. In this context, the non-invasive technology-based core muscle reconditioning emerges as a promising and underutilized tool. Prior functional wellness studies targeting lower extremity muscle systems24,51 have similarly shown benefits in strength recovery, functional mobility, and pain reduction, supporting the biological plausibility of the observed outcomes. Correlation analysis showed LRS severity was significantly correlated with the magnitude of clinical improvement in pain (VAS, p = 0.006) and disability (PDQ, p = 0.041). A non-significant trend was also observed for Lee foraminal stenosis grades regarding VAS reduction (p = 0.079). Pathologically, LRS involves the focal compression of traversing nerve roots, frequently leading to radicular symptoms that are traditionally addressed via conservative stabilization or decompressive techniques.52 The preferential response in subjects with focal lateral recess pathology may relate to the intervention’s proposed mechanism: utilizing HIFEM and radiofrequency energy to dynamically stabilize the motion segment and mechanically offload the compressed nerve root. HIFEM technology utilizes alternating magnetic fields to depolarize motor neurons, inducing brain-independent supramaximal contractions that strengthen and remodel deep muscle tissue; simultaneously, synchronized radiofrequency recruits myosatellite cells and raises tissue temperature to promote muscle regeneration and improve local perfusion.53

In contrast, generalized degenerative markers, such as the Pfirrmann classification for global disc degeneration or generalized central canal narrowing did not significantly correlate with outcomes in this cohort, which aligns with standard MRI grading evaluations of structural spine interventions.54 This suggests that diffuse spinal degeneration may be less responsive to this specific non-invasive modality than isolated, mechanically driven impingements. However, these findings remain exploratory and the precise in vivo biomechanical impact of HIFEM on lateral recess dimensions remains speculative without dynamic imaging. Consequently, further investigation in larger, controlled cohorts is necessary to validate whether LRS grading can reliably guide subject selection for this therapy. Despite these promising results, this study has several limitations. The relatively small sample size (n=25), without a formal statistical power calculation, formal inter-rater reliability testing alongside the retrospective design, and absence of a control group limit the generalizability of the findings. Because a subset of subjects also received treatment to other body regions, the possibility that improvements in back-related outcomes were partly influenced by non-lumbar treatment cannot be fully excluded, even within the core-focused analysis presented here. This study was not designed to formally assess a dose-response relationship. The number of treatment sessions (range 3–6) was determined by individualized clinical judgment rather than randomized allocation, and the narrow range together with the limited sample size precluded a meaningful dose-response analysis. The lack of long-term follow-up data prevents conclusions about the durability of the treatment effects. Additionally, the intervention was not tested against existing standard-of-care protocols, which would be necessary to assess relative efficacy. However, the retrospective study design offers real-world insights and clinical applicability of the results. The correlation analysis performed among the investigated outcomes and baseline MRI using validated self-reported questionnaires supports the strength of this investigation, showing internal consistency of the results and offering a subject perspective. Future randomized controlled trials (RCTs) with larger, stratified populations and standardized treatment protocols are essential to validate these results. Incorporating longer follow-up periods and combining this intervention with established rehabilitation programs could further elucidate its role in chronic LBP management.

Conclusion

In this retrospective cohort, this study suggests that HIFEM technology is associated with a significant reduction in pain severity, functional and psychosocial disability, and its impact on the ability to manage everyday life in subjects with LBP, including those with age-related muscle decline. While short-term findings suggest this non-invasive intervention addresses key structural and muscular deficits, prospective controlled studies with longer follow-up periods are needed to determine long-term efficacy, durability, and overall clinical utility.

Funding Statement

This study was sponsored by BTL Industries. The sponsor provided the equipment used in this study. However, no financial support was received for the authorship or publication of this article.

Data Sharing Statement

The datasets generated and analyzed during the current study are not publicly available due to the proprietary nature of the research.

Disclosure

The author serves as a clinical investigator and speaker for BTL Industries, the sponsor of the study. The sponsor had no role in the study design, data collection, analysis, interpretation, or the decision to publish these findings. The author is solely responsible for the content and writing of this article, and this relationship did not influence the representation or interpretation of the reported research results. The author reports no other conflicts of interest in this work.