CODIFI2: Randomised controlled trial of swab versus tissue sampling for infected diabetic foot ulcers, and comparison of culture versus molecular processing techniques
A 730-patient trial will determine whether a simple wound swab works as well as a tissue sample for diagnosing and treating infected diabetic foot ulcers. Diabetic foot infections are a leading cause of lower-limb amputation. Clinicians currently disagree on whether to take a swab from the ulcer surface or cut out a piece of tissue, and whether standard culture or faster molecular testing gives better results. No large trial has directly compared these approaches. If the trial shows that swabs are non-inferior to tissue samples, thousands of patients each year could avoid a painful, invasive procedure. If molecular testing proves superior to culture, microbiology results could arrive in hours rather than days, allowing clinicians to target antibiotics sooner and reduce the risk of amputation. The trial also includes a health economic analysis to determine whether molecular techniques are cost-effective for the NHS, and a virtual clinic study to explore how clinicians would change prescribing based on molecular results. A prognostic model will link bacterial profiles to patient outcomes, potentially identifying who needs more aggressive treatment.
View original technical description
Trial Design: Multi-centre, phase III, parallel group, randomised controlled trial in diabetic foot ulcer infection (DFU), including a 12month internal pilot phase (study 1) & study 2: diagnostic technique agreement study, virtual clinic study, prognostic model development, qualitative study and value of information analysis Setting:UK multi-disciplinary DFU clinics Inclusion criteria:age 18+, Type 1 or 2 diabetes; clinically suspected DFU infection;consent Exclusion criteria:suspected osteomyelitis/severe infection; ulcer duration>2 years;previously randomised *Study 1:Sampling Techniques Randomised (1:1 allocation) to policy of swab or tissue sampling for index & subsequent ulcers, using minimisation algorithm Blinding: Independent clinical assessor to conduct outcome assessments with blinded central photography review of the healing endpoint Technologies being evaluated: Intervention 1:Wound swab with analysis by culture & antibiotic susceptibility (C&S) Intervention 2:Tissue sample with analysis by C&S Primary outcome:Time to healing of index ulcer Secondary outcomes: Compliance with randomisation; healing status at 26 & 52 weeks (& at 104 weeks in people recruited in year1), area change at week 4, time on antibiotics, SAEs (amputation, osteomyelitis, hospital admission) & QoL (Diabetic Foot Ulcer Scale-Short Form & EQ-5D-3L) Health Economic outcomes:Health resource utilisation (incl. antibiotics, cost of health & social services) and QALYs. Follow-up visits: At week 4, 26 and healing. Record review, health resource & QoL at 4/8/12/26/39/52 and (for recruits in first year) 104 weeks Sample Size: 730 participants are required to have 90% power to detect a minimum clinically important difference of 12.5% in the proportion healed at 12 months (centres on 45% healed at 12 months), assuming 2-sided, 5% significance level & 10% loss to follow-up Analysis: Primary analyses on intention-to-treat patient population Primary endpoint analysis: Cox proportional hazards regression model fitted to time to healing with covariates for minimisation factors. Deaths and amputations will be considered competing risks and treatment effect will be estimated with respect to time to death and amputation-free healing. Revascularisation will be considered as a time dependent covariate Health economic analysis: within-trial cost-effectiveness, cost per QALY, using EQ-5D utilities, NHS & PSS costs. Value of information of trial comparing molecular and culture techniques During the trial we will also investigate the impact of modifying letters sent with patient questionnaires improves response rates (a study within a trial). *Study 2:Processing Techniques - A second wound sample (swab in those allocated to swab sampling at randomisation; tissue from those allocated to tissue sampling) will be sent to a central laboratory for molecular testing, thus we will have both swab and tissue samples with analysis by molecular methods and culture. We will compare microbiology results obtained via C&S and molecular methods to assess agreement. - Results from C&S and molecular methods will be presented to clinicians in a ‘virtual clinic’ to ascertain their proposed antibiotic prescribing decision-stop/amend/continue. - We will refine prognostic models of the relationship between clinical data plus bacterial profiles (via molecular methods), and patient outcomes. - We will explore with clinicians the aspects of microbiology reports required to permit molecular techniques to replace, rather than be an adjunct, to culture (e.g. time to results, comprehensiveness) - We will estimate the value of further research aimed at reducing the likely uncertainty surrounding the economic model’s parameters associated with molecular techniques vs plating and culture using a Value of information analysis approach (VOIA).
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know