Early Percutaneous Transluminal Angioplasty in Diabetic Foot Syndrome (PTA-DFS)

Overview

The planned study is a Randomized Controlled Monocentric Trial, which will provide evidence on whether early angiography in percutaneous transluminal angioplasty (PTA) readiness ("immediate" treatment, within 48h) has advantages over the "standard of care", i.e., an elective procedure ("elective PTA") for the treatment of diabetic foot ulcer (DFU). The primary study endpoint is to investigate the impact of the "early PTA" within 48 hours on wound-healing assessed by wound area changes after PTA using a 3D-camera with artificial intelligence (AI)-based wound-analysis-system. The secondary endpoint is the effect of early PTA on the combined occurrence of major adverse limb (MALE) and cardiac events (MACE) over 12 months post-angioplasty using time-to-event analysis. Data will be collected at baseline, 24 hours, 1, 2, 3, 6, and 12 months after PTA. Diabetic kidney disease, distal symmetric polyneuropathy, retinopathy, cardiomyopathy, laboratory analyses, clinical scores, AI-based fundus photography, echocardiography, duplex sonography, and pulse oscillography will be assessed. Explanatory variables for wound healing are wound microbiome changes using whole-genome sequencing and oxygen saturation of the wound environment measured using near-infrared spectroscopy. Altered microbiome composition in ulcers can lead to severe local and systemic infections and complications, including major amputations. Nevertheless, the specific significance of the wound microbiome composition in chronic ischaemic ulcers in type 2 diabetes and the impact of PTA on the wound microbiome in type 2 diabetes is unclear. The exact timing for treating peripheral arterial disease (PAD) by revascularization in DFU after initial diagnosis is unknown and has yet to be fully understood.

The planned study is a monocentric, randomized, and controlled clinical trial. Study participants will be randomized in a 1:1 ratio into two groups: one receiving early PTA ("immediate PTA", within 48 hours after diagnosing a hemodynamically relevant chronic leg artery stenosis) and the other receiving elective PTA (standard of care). Baseline and all follow-up examinations will be performed identically in both groups, except for the timing of PTA. Although the study is unblinded for both participants and physicians due to practical feasibility, blinding will be applied during the data analysis phase. This will involve separate analysis of primary and secondary endpoints by interventionists (physicians or clinicians) and data analysts to minimize bias. Aside from the timing of PTA, baseline and follow-up examinations will be conducted uniformly across both groups. These examinations will focus on the severity and healing process of the foot ulcer. If complications in wound healing occur in the elective group, a crossover to the "immediate" PTA group will be implemented. Thus, for each participant, the treatment of peripheral arterial disease (PAD) and diabetic foot ulcers (DFU) will follow established guidelines and will be monitored and documented by specialized staff during clinical follow-ups and study visits. During the visits, investigators will not only focus on the diabetic foot ulcers but will also consider each participant's overall health. In this context, the investigator will first aim to improve renal retention parameters through sufficient hydration, taking cardiac function into account. In cases of severe renal insufficiency, guideline-based selective angiography with PTA standby using CO2 angiography may be performed, as this method does not affect kidney function. Baseline characterization The participant's complete medical history, as well as information on current and previous medications, is obtained during the initial clinical evaluation. Additionally, data on lifestyle factors (such as smoking, alcohol consumption, diet, physical activity, mental health, erectile function, symptom severity, pain, self-management, diabetes distress, erectile dysfunction, and diabetes burden), as well as social status, are collected. A physical examination is performed, which includes determining whether venous or neuropathic triggers should be considered in addition to the ischemic etiology of the ulcers. Anthropometric measurements, retinopathy screening, wound swabs, and sampling of deep tissue material from the wound bed for microbiome DNA sequencing are also carried out. Additional proteomic and metabolomic analyses of the wound biofilm are planned. Routine laboratory chemistry testing will be performed regularly during the study visits, and the duration of diabetes and ulcer will also be recorded. Wound documentation Wound documentation and clinical measures are performed to monitor existing ulcerations using a 3D camera with an AI-based wound analysis system (Curevision®, Munich, Germany) to objectively assess wound healing progress. The system analyzes wound dimensions (length, width, area, and depth) and wound bed tissue (granulation, fibrin, and necrosis), providing consistent, non-operator-dependent data. The wound is assessed according to the WIfI (Wound, Ischemia, and foot Infection) and Wagner (Armstrong) classification criteria, as recommended by current interdisciplinary guidelines. Tissue involvement, including possible osteomyelitis, is investigated by X-ray, computed tomography, or magnetic resonance imaging if clinically indicated. Wound documentation will be conducted at each follow-up visit until wound closure is confirmed. Perfusion measurement Duplex/Doppler ultrasound of the leg arteries, along with Doppler frequency analysis, is performed alongside ankle-brachial and toe-ankle index measurements. Segmental pulsoscillography is used to locate the site of vascular occlusion and evaluate perfusion pressure, serving as a monitor for critical limb ischemia. Measurement of arterial perfusion at rest and perfusion reserve is achieved using occlusion plethysmography. In the subsequent duplex ultrasound, the anatomy of the peripheral arteries is assessed. Later in the course, these examinations determine the patency of the revascularized vessel segment, the degree of stenosis by peak flow velocity determination, and intimal hyperplasia. Furthermore, flow-dependent dilatation of the femoral artery is also examined using duplex sonography. Of note, in cases of severe renal insufficiency, carbon dioxide angiography may be employed for selective angiography. Oxygen saturation of the foot will be analyzed using a spectroscopic camera, and control measurements will be performed by assessing the hands. Assessment of Microvascular Complications Diabetic Sensorimotor Polyneuropathy (DSPN) is commonly prevalent in Diabetic Foot Syndrome (DFS) and is assessed using the Neuropathy Disability Score (NDS), based on the modified Toronto Consensus criteria. An NDS score ≥ 3 points is considered pathological. AI-based fundus photography is performed using a non-mydriatic camera (DRS Plus, Centervue®, Padua, Italy) to screen for diabetic retinopathy and macular degeneration. Utilizing confocal imaging technology, the camera provides high-quality retinal images without the need for pupil dilation. Additional complications, such as diabetic kidney disease, are assessed through urinary albumin and creatinine levels, and the estimated glomerular filtration rate (eGFR) is calculated based on cystatin C and creatinine levels. Wound swab Specimen collection is performed according to the Levine method using a commercially available flocked swab (FLOQSwab®, COPAN Diagnostics Inc., Murrieta, USA). Swabs are taken from each study participant simultaneously to obtain material for determining the microbiome and the corresponding resistogram. Specific transport media are used depending on the downstream process. Swabs are collected after thorough antiseptic-free or neutral wound cleansing. The first swab is taken after initial wound care, the second at four weeks, the third at eight weeks, the fourth at 12 weeks, the fifth at 24 weeks, and the last at 48 weeks. Biosample collection, processing, and storage follow established Standard Operating Procedures (SOPs). Access to biosamples is regulated by a formal request process reviewed by the PTA-DFS Study Principal Investigators (PIs), ensuring ethical compliance. Separate ethical approval for the CARDIAMET biobank (#2023-2525) guarantees adherence to national and international standards, ensuring sample integrity and participant confidentiality. Blood collection This will be performed in the morning under fasting conditions. Blood samples are collected as part of the clinical routine, immediately before PTA and during follow-up visits for DFS, and do not constitute study-specific collections. On these occasions, additional blood samples may be obtained without requiring extra study-related punctures. Blood will be collected in so-called stretch tubes for testing cell-free (cf)DNA. The time points include routine follow-up examinations on the initial day of examination (V0), on day 1 (V1), after one month (V2), two months (V3), three months (V4), six months (V5), and 12 months (V6) post-PTA. Over the study's duration, no more than 500 ml of blood will be drawn per participant. Examination of cell-free deoxyribonucleic acid (cfDNA) To evaluate the incorporation of microbial components from revascularized tissues, the investigators aim to quantify the amount of microbial cfDNA. A sufficient amount of the collected material will be analyzed within a maximum period of three years. Outcome Monitoring Over the 12-month follow-up period, the composite secondary endpoints, major adverse limb events (MALE) and major adverse cardiac events (MACE), are monitored through interviews focusing on vascular events (cardiac, cerebral, peripheral), amputations, infections, sepsis, hospitalizations, and all-cause mortality. For participants unable to attend on-site visits, telephone interviews are conducted to assess complications, including cardiovascular or all-cause mortality. Statistical Analysis The primary endpoint focuses on the difference in changes in wound area (delta wound area: first visit - last visit) in T2D participants after "early PTA" compared to individuals with "standard PTA." For the primary endpoint-change in wound area after early PTA-and the secondary endpoint-combined incidence of MALE and MACE-a power calculation was performed. Assuming a medium effect size (Cohen's d = 0.4), corresponding to 0.4 standard deviations of the difference in the delta wound area and the difference in time-to-event for MALE/MACE from baseline to the last visit between groups, a sample size of n=100 per group (total n=200) is required to achieve 80% power at α=0.05. One explanatory variable evaluates the effect of "early PTA" on wound microbiome dynamics. This objective aims to measure the beta-diversity changes of the wound microbiome, assessed via whole-genome sequencing and analyzed using Bray-Curtis distances. Taxonomic profiles from whole-genome sequencing of the wound microbiome will be generated using bioinformatics scripts. Sequencing data produced via the Oxford Nanopore platform will be analyzed with Kraken2 and minimap2, using a custom database that includes protozoan, fungal, and viral sequences. The secondary endpoint includes the combined incidence of MALE and MACE post-angioplasty. Oxygen saturation in the wound environment is measured as another explanatory outcome potentially linked to wound healing. Additional outcomes include evaluating risk factors such as comorbidities (severity of anemia and stages of chronic kidney disease), severity of insulin resistance, DSPN, duration of hospital stay, and interactions with medications, including antibiotics and antithrombotic therapies. Taxonomic profiling of whole-genome sequencing data of the wound microbiome is performed using bioinformatics scripts developed by the Department of Algorithmic Bioinformatics at Heinrich-Heine-University (HHU), as previously described. Sequencing data generated with the Oxford Nanopore platform are used for taxonomic profiling with Kraken2 and a comprehensive custom database of protozoan, fungal, viral, and plant sequences. Visualizations comparing metagenomic composition at different taxonomic levels are generated using custom scripts. To evaluate the clustering of samples, principal coordinate analysis based on the Bray-Curtis distance is performed. Using the Conda version management system, the entire workflow is provided as a reproducible Snakemake workflow. Associations between wound microbiome composition and metabolic phenotypes are analyzed using omnibus, blockwise, and traitwise statistical models. Variables are compared using the Wilcoxon matched-pairs signed-rank test, the unpaired two-sided Student's t-test, and the two-sided Mann-Whitney U-test to detect differences over time and between groups. Nominal variables are compared using the chi-square test or Fisher's exact test, as appropriate. Relationships between variables (e.g., associations between changes in microbiome composition and wound healing or complications) are assessed using partial Pearson correlation coefficients, with and without adjustment for confounders and multiple testing using the Bonferroni method. A standardized mean difference (Cohen's d) was used for power analyses because preliminary estimates for the mean and standard error of the mean of wound microbiome changes after PTA were not available in the literature. All statistical analyses are performed using SPSS for Windows 23.0 (SPSS Inc., Chicago, IL, USA), and all graphs are generated using GraphPad Prism, version 7.01 (GraphPad Software, Inc., La Jolla, CA, USA).