Pulsed Electromagnetic Field Therapy (PEMF)

Electrical and electromagnetic stimulation in medicine

Reviews of the field place pulsed electromagnetic fields within a longer history of electrical and electromagnetic stimulation in medicine. A 1989 review in the Journal of Biomedical Engineering describes the history and development of electric and electromagnetic fields used to treat a variety of medical conditions, and a 2000 review in Neurosurgery sets out the history and scientific basis of electrical stimulation to aid bone fusion. Only the abstracts of these two papers were read for this page, so specific dates and named pioneers are not given here.

Laboratory rationale and clinical adoption

The Cochrane review of electromagnetic field stimulation for delayed union or non-union of long bone fractures states that laboratory studies have shown electromagnetic fields can stimulate the formation of new bone, which indicated a potential role in fractures that had failed to unite. A 2018 narrative review in Research in Veterinary Science, covering PEMF history, device taxonomy and mechanisms, reports accumulating clinical evidence in animals and humans for indications including bone healing, wound healing, osteoarthritis and post-operative pain and edema.

That 2018 paper is a narrative overview, not a systematic review, and its authors' favorable framing should be read alongside the systematic reviews summarized under Research and Evidence, which are more cautious for several of the same indications.

PEMF among physical agent modalities

Pulsed electromagnetic field (PEMF) therapy refers to delivering pulsed electric and magnetic fields to tissue through inductive coils, as the 2018 narrative review describes. Systematic reviews treat it as one member of a wider group of physical agent modalities. A 2025 network meta-analysis of fracture care lists electrical stimulation, PEMF stimulation, ultrasound, low-level laser, magnetic stimulation, extracorporeal shock wave therapy and capacitively coupled electric field stimulation as separate interventions.

Overlapping labels in reviews

Review authors do not use one label consistently. The Cochrane review of non-union grouped three trials of pulsed electromagnetic fields with one trial of capacitive coupled electric fields under the heading electromagnetic field stimulation, and a 2014 meta-analysis pooled PEMF with low-intensity pulsed ultrasound as bone growth stimulation. A finding reported for a combined group should not be read as a finding for PEMF alone unless the review separates them.

This page keeps PEMF separate from static magnet products such as magnetic bracelets or insoles, which are covered on the Magnet Therapy page and have a different evidence base. Evidence for one should not be transferred to the other.

Devices and delivery

In the sources reviewed, PEMF is delivered by a device that generates pulsed fields through a coil placed near the treated body part. The 2026 knee osteoarthritis meta-analysis reports that trials differed in field amplitude and frequency and that this heterogeneity in treatment protocols limits clear conclusions. This page does not give device settings, schedules or treatment protocols, and the sources read do not describe consumer whole-body mats, which are therefore not characterized here.

Settings, comparators and what is not described

The trials summarized in the reviews mostly compared an active device with a sham or placebo device, so participants could not always tell whether a field was being delivered. The abstracts read for this page do not describe session sequences, practitioner roles, hygiene, consent processes or training. No professional register or training standard specific to PEMF was found in the sources read, and the evidence comes from clinical research settings rather than from wellness settings.

Uses examined in clinical research

Randomized trials and systematic reviews have examined PEMF for delayed union and non-union of long bone fractures, for acute fractures, for osteoarthritis of the knee and other joints, and more recently for fatigue in multiple sclerosis. The 2018 narrative review also lists wound healing, inflammation and post-operative pain and edema among indications for which its authors describe accumulating clinical evidence.

General wellness claims

Promotional descriptions of PEMF often refer to cellular energy, tissue repair or general wellbeing. The sources opened for this page do not evaluate those general wellness claims; the systematic reviews cited here concern patients with specific conditions and specific outcomes. Findings for fractures, osteoarthritis or multiple sclerosis fatigue should not be carried across to wellness uses, or from one condition to another.

Delayed union and non-union of long bones

The Cochrane review included four blinded, placebo-controlled trials with 125 participants, three of pulsed electromagnetic fields and one of capacitive coupled electric fields, mostly in non-union of the tibia. The pooled risk ratio for union at a fixed time point was 1.96 (95% CI 0.86 to 4.48), not statistically significant, with substantial heterogeneity (I2 = 58%). The authors judged the evidence inconclusive and insufficient to inform practice, and no pain reduction was found in two trials.

Acute fractures

A 2014 meta-analysis of 13 trials with 737 adults with acute fractures pooled PEMF with low-intensity pulsed ultrasound and found no significant difference in the proportion of non-union, with some heterogeneous time-to-union findings in subgroups. A 2024 systematic review update of three randomized trials (197 patients) found no significant effect of PEMF on acute bone healing and did not support its use for that purpose, noting scant and heterogeneous literature.

A 2025 network meta-analysis of 39 studies (2,379 participants) of seven physical agent modalities reported that the modalities overall were associated with shorter healing time, but most direct and indirect comparisons between modalities had confidence intervals including the null, and it does not isolate PEMF as a distinct finding.

Osteoarthritis

A 2026 meta-analysis of nine randomized trials (457 patients with knee osteoarthritis) found no significant improvement in VAS pain or total WOMAC score at one month. WOMAC pain improved at 18 to 21 days but not at one month, stiffness and daily activity improved at one month, objective functional measures did not improve, and risk of bias was high. The authors note the changes may not reach thresholds for clinical meaningfulness.

An earlier 2018 meta-analysis of 12 sham-controlled trials reported less pain and better function with PEMF in knee and hand osteoarthritis but not cervical osteoarthritis, while cautioning that the hand and cervical evidence was limited. Results differ between reviews, and the more recent one reports less consistent benefit.

Fatigue in multiple sclerosis

A 2025 meta-analysis of seven sham-controlled trials with 327 participants found a small reduction in fatigue severity with PEMF (SMD -0.23, 95% CI -0.45 to -0.01), with the interval close to no effect. Quality of life and depressive symptoms did not differ from sham. The authors call for larger trials, and this finding applies to fatigue in multiple sclerosis only.

Adverse events reported in trials

In the 2018 osteoarthritis meta-analysis, three trials reported adverse events and the combined results showed no significant difference between PEMF and sham. The Cochrane non-union review noted that one trial reported two minor complications resulting from treatment. Most other abstracts read do not report harms, and the trials were small, so the sources do not allow the frequency of rare or serious events to be estimated.

Contraindications and care boundaries

The sources read do not state contraindications or precautions, such as questions about implanted electronic devices or pregnancy, and this page does not supply them from outside the sources. Such questions belong with a clinician and the device manufacturer's labeling, which was not reviewed here. Nothing in the sources supports delaying assessment or replacing necessary care; a fracture that is not uniting, or persistent joint or neurological symptoms, calls for clinical assessment.

Limits of the safety evidence

Several reviews describe small trials, high risk of bias and heterogeneous protocols. The reviews concern clinical research devices; they do not assess the output, quality or labeling of consumer-market products, and their findings should not be assumed to apply to them.