Evaluating Extracorporeal Shockwave Therapy for Diabetes Related Foot Ulcer Healing

Introduction
Diabetes related foot ulcers (DFUs) are associated with prolonged healing, predisposing patients to infection, hospitalisation and limb loss, despite evidence-based care. Extracorporeal shockwave therapy (ESWT) involves delivering high frequency soundwaves directly to the ulcer bed. It is proposed to improve healing through the stimulation of angiogenesis and re-epithelisation. Currently there is a paucity of evidence to support the routine use of ESWT in clinical practice as an adjunct to DFU care. In particular, there is conflicting evidence on the efficacy of ESWT and the optimal dose. This study aimed to investigate ESWT as a DFU healing aid by:
• Evaluating the current literature to assess the impact of ESWT on DFU healing
• Assessing the deliverability, cohesiveness of trial components and up-scalability of a randomised control trial of ESWT for DFU healing
• Developing theories behind patients’ and clinicians’ perceptions, ideas, concerns and experiences of ESWT for DFU healing
• Assessing whether including ESWT in the treatment for DFUs represents value for money
Methods
A systematic review searched databases for randomised controlled trials of ESWT plus standard of care, to standard care alone in adult patients with DFUs. The primary outcome was time to healing. Bias was assessed using the Cochrane Risk of Bias 2 tool and certainty in the evidence was assessed using the GRADE criteria.
A pilot three arm double blinded randomised controlled trial (RCT) comparing high dose (500 shocks/cm2) ESWT, low dose (100 shock/cm2) ESWT and sham (0 shocks/cm2) ESWT, in addition to standard care. ESWT was delivered 3 times over 7±2 days and patients were followed up for 24 weeks. The primary outcome was deliverability of the trial. Secondary outcomes included time to healing, quality of life and healthcare resource use.
Qualitative studies assessed patients’ and healthcare professionals’ engagement in the trial and DFU research. The patient study was nested in the pilot trial, whilst the clinicians’ study ran parallel to the trial. Maximum variable sampling was used to recruit patients from one tertiary centre and clinicians nationally. Data was collected using semi-structured interviews, which were audio-recorded and transcribed verbatim. An inductive exploratory approach using reflexive thematic analysis was used to analyse data.
A four state Markov model was used to undertake deterministic, probabilistic and value of information analysis. Input parameters were estimated from the pilot trial and the literature. High dose, low dose and standard of care alone were investigated using a 1000 patient cohort with one DFU. The perspective of the analysis was from the UK’s National Health Service and Personal Social Services. The cycle length was one day, and the time horizon was one year.
Results
The systematic review included six RCTs (471 participants). Every trial used a different treatment regime for ESWT. Risk of bias ranged from unclear to high across the domains. Time to healing was probably shorter with ESWT (GRADE: low certainty).
In the pilot RCT 15.6% (141/904) of screened patients were eligible and 52.5% (74/141) of screened patients were recruited. Overall compliance with treatment arm allocation was 94.6% (70/74). Follow-up attendance at 6-, 12- and 24-weeks was 97.3% (72/74), 93.2% (69/74) and 87.8% (64/74). Amount of missing data ranged from 0% to 20%. The median DFU healing time in the high dose ESWT arm was 54.0 (IQR 119.0) days, 78.5 (IQR 61.0) days in the low dose arm and 83.0 (IQR 85.0) days in the sham arm. The mean EQ-5D-5L utility value at 24 weeks was 0.621 (95% CI 0.438 to 0.804; effect size 0.12) in the high dose arm, 0.779 (95% CI 0.683 to 0.876; effect size 0.34) in the low dose arm and 0.806 (95% CI 0.717 to 0.895; effect size 0.06) in the sham arm. Healthcare resource use was lowest in the low dose ESWT arm.
The patient qualitative study included 18 participants who completed the trial, one participant who dropped out of the trial, and three patients who were eligible but declined trial participation. Drivers for participants to take part in the pilot trial included desperation for healing and gaining something extra. Barriers for trial participation included a feeling of alienation from healthcare services based on poor prior experiences, discrimination at work and poor health. The clinician qualitative study included ten clinicians. Clinician engagement in research was influenced by the presence of a supportive multidisciplinary team and how the research was conducted. Adoption of new interventions was based upon whether there were clear intervention development steps, generalisability of research findings and available resources.
The early economic modelling demonstrated that high dose ESWT was the dominant treatment strategy. The deterministic analysis found a mean difference between high dose ESWT and standard of care alone of 0.022 QALYs and £922. In the probabilistic analysis, the probability of high dose ESWT being cost effective was 60.2%, at a willingness to pay threshold of £20,000. The Expected Value of Perfect Information (EVPI) per patient was £24.56. The partial EVPI per patient for high dose ESWT efficacy was £7.87, £3.56 for health state utilities and £0 for healthcare resource use costs.
Conclusion
ESWT remains a promising intervention to aid DFU healing. A future trial is deliverable and should be adapted to optimise patient engagement and meet clinicians’ needs. A two-arm trial comparing high dose ESWT to standard of care assessing the clinical and cost effectiveness of ESWT’s efficacy in augmenting time to DFU healing is now required.