Objective: To test the efficacy and safety of therapeutic magnetic resonance (TMR) in the management of diabetic foot ulcers (DFU), the authors designed a prospective randomised controlled trial in three highly specialised diabetic foot clinics. Method: All the patients consecutively visited in a period of 18 months were screened according to the inclusion (presence of an ulcer >1 cm 2 in the foot lasting at least 6 weeks; ABPI>0.6; consent to participate in the study) and exclusion (Charcot's foot; local or systemic infections; chronic renal failure; any wearable electrically-driven life-supporting device) criteria. Patients, who were treated according to international guideline protocols, were randomised into two groups: group A received for four weeks the sham application of TMR, while group B received the active TMR for the same period. People were followed-up to 10 weeks and healing rate (HR), healing time (HT), rate of granulation tissue on wound bed (% GT), reduction of the area of the lesion (∆AL) and a score (0–3) evaluating erythema, oedema, pain and tenderness, respectively, were measured. Adverse events (AE) were registered and monitored throughout the study. Results: No differences were observed in HR, HT and ∆AL between the two groups during follow-up, while % GT and the scores for erythema, oedema and pain at 10 weeks showed significant (p<0.05) improvements in group B compared with group A and versus baseline. When restricted to non-ischaemic patients (ABPI>0.8), ∆AL was significantly (p<0.05) more pronounced in group B than in group A. No difference in AE occurrence was observed between the two groups. Conclusion: Our study, despite not being able to demonstrate the effectiveness of TMR on healing rate at 10 weeks, with 4 weeks of active treatment in neuro-ischaemic DFUs, shows positive effects on clinical aspects of the DFU and is associated with a significant increase of GT in the wound bed. Declaration of interest: The study has been fully sponsored by Thereson S.p.A., manufacturer of TMR devices.