Abstract
BACKGROUND: Diabetic foot is among the most frequent complications of diabetes mellitus (DM), with potentially dramatic consequences ranging from chronic wounds to major lower-limb amputations. The Tardivo Algorithm is a simple prognostic scoring system designed to support risk stratification and structured longitudinal reassessment in routine clinical care. OBJECTIVE: To describe the real-world implementation and feasibility of dynamic risk reassessment using the Tardivo Algorithm in a prospective observational cohort of patients with diabetic foot managed in a vascular surgery outpatient setting, and to explore associations between baseline risk stratification and clinical outcomes. METHODS: This prospective observational cohort study was conducted in a routine outpatient clinic for complex wounds. Adult patients with diabetic foot were classified according to the Tardivo Algorithm at baseline and underwent structured serial reassessments at each follow-up visit as part of usual multidisciplinary care. No comparator group was included. Patients were followed for 6-18 months, and outcomes were descriptively recorded as minor amputation, major amputation, wound in process of healing, or complete healing. RESULTS: A total of 42 patients were followed for up to 18 months. Mean initial Tardivo score was 7.6 ± 4.8, with 19% classified as high risk (≥12 points). Limb preservation was observed in 94.3% of participants, and complete healing occurred in 57%, with a mean healing time of 5.05 ± 1.95 months. Higher baseline Tardivo scores were positively associated with peripheral arterial disease (r = 0.740; p < 0.001), while healing time correlated with both PAD (r = 0.547; p = 0.006) and prior amputations (r = 0.523; p = 0.009). These correlations were not independent in multivariable models. Findings reflect associations observed in a real-world, structured outpatient care model. CONCLUSION: In this prospective real-world cohort, structured application of the Tardivo Algorithm was feasible and allowed dynamic clinical monitoring. Clinical outcomes observed during follow-up are described within the context of this non-controlled design and should be interpreted as observational associations rather than indicators of therapeutic effect. Controlled studies are required to determine the independent impact of algorithm-guided reassessment and adjunctive therapies.