Background Foot problems associated with systemic sclerosis (SSc; scleroderma) are common and disabling. To-date there have been no formal studies of interventions directed at the foot problems experienced in people with SSc.
Objectives The main objective of this randomized controlled trial was to evaluate whether foot pain in SSc can be improved through the routine provision of a simple pressure-relieving and insulating insole.
Methods A multicentre RCT conducted in four UK participating centres. A total of 141 consenting patients with confirmed SSc (ARA/ACR 1980 criteria) and plantar foot pain were randomised to receive either a commercially available pressure-relieving thermally insulating insole (Prothotic™ Duoform, RRP £8.55) or a sham insole identical in appearance to the active intervention but with minimal cushioning or thermal properties. Randomisation on a 1:1 basis was performed centrally using a 24-hour, automated randomisation system at a UKCRC registered clinical trials research unit.
The primary end point was a reduction in pain measured using the 100mm pain subscale of the Foot Function Index, after 12 weeks of intervention. Sample size was determined a priori on a requirement to detect a 15mm difference (SD 25.7, alpha =0.05 and power =90%). In a subset of 49 patients at the lead centre (Leeds), plantar pressure measures (maximum mean pressure at heel and forefoot, with and without insoles in situ) were also obtained for entry into an exploratory analysis.
Results Analysis of baseline characteristics indicated excellent comparability between the active intervention and sham group for centre, gender, age, SSc subset and Rodnan skin score. One hundred and thirty patients provided valid data for the primary endpoint. In both groups there was a systematic improvement in FFI pain subscale scores from baseline to 12 weeks (Active group -13.1mm, 95%CI -18.66 to-7.55 ; Sham group -10.7, 95%CI -16.17 to -5.28). An ANCOVA model adjusting for centre, gender and baseline FFI score confirmed no difference in effect between the intervention and sham groups (difference=-2.4, 95%CI -7.70 to 2.94, p=0.3778). Compared to a shoe-only baseline measure, pressure was lowered in the heel region by use of the active insole (median difference (range) -24kPa (-75,22)) compared to the sham -4.6kPa ((-23,10), (p<0.01)), but in the forefoot the difference between insole types was not significant; pressure change -17.2 kPa (-61,19) for active insoles and -10.8kPa (-39,0) for sham insoles (p=0.31).
Conclusions The study compared a simple therapeutic insole with a sham device and found that over 12 weeks both produced a clinically worthwhile improvement in patient reported foot pain but with no difference between intervention arms. The active device produced a significantly greater reduction in pressure than the sham at the heel although not at the forefoot. This exploratory analysis suggests therefore, that despite careful selection the sham device introduced some unintended physical effect. The difficulty with employing a true sham as a control in physical intervention trials such as this may suggest the need for a zero intervention arm for similar studies in future.
Acknowledgements Funding: Arthritis Research UK Clinical Studies Group for Autoimmune Rheumatic Diseases (award number 18826). Trial regn. ISRCTN 02824122
Our thanks to all TSC and TMG contributors and to the clinical staff at each site.
Disclosure of Interest: None Declared