Ten Years Follow-up After the Arthroscopic Repair of Osteochondral Lesions of the Talus Via Autologous Matrix-induced Chondrogenesis (AMIC)

The target population for this investigation consists of patients treated for osteochondral defects of the talus at the IRCCS Ospedale Galeazzi Sant'Ambrogio within the specified timeframe from April 2013 to April 2016. To be eligible for inclusion, patients must have undergone the repair procedure via the AMIC technique utilizing a specific porcine collagen I/III membrane, commercially known as Chondro-Gide. Conversely, the study excludes individuals with irrelevant post-operative diagnoses, such as pigmented villonodular synovitis, as well as cases where it is impossible to retrieve the necessary follow-up data.The study protocol relies on the retrospective and longitudinal analysis of clinical and functional outpatient data gathered at four distinct time points. These checkpoints consist of a pre-operative baseline assessment followed by three post-operative follow-up evaluations conducted at one year, five years, and ten years. Throughout these intervals, patient pain levels are quantified using the Visual Analog Scale. Functional and clinical outcomes are measured through the American Orthopaedic Foot and Ankle Score to determine overall functionality, and the 17-item Italian Foot Function Index to evaluate disability and functional limitations. Additionally, the European Foot and Ankle Score is administered specifically at the five and ten-year follow-ups to provide further validation aligned with the most recent international standards. Safety and procedure survival are also strictly monitored by recording all complications, clinical failures, and surgical revisions that occurred over the ten-year span. Finally, demographic and biometric data-including age at surgery, sex, BMI, smoking habits, and lesion characteristics such as localization and volume-are gathered to perform correlation analyses.The statistical approach is designed around the primary objective of evaluating clinical parameters across the four established time points. This will be achieved by measuring within-patient effects using a one-way repeated measures ANOVA. The sample size calculation was performed assuming a medium effect size of Cohen's $d = 0.3$, which corresponds to $\eta^2 = 0.059$. With a Type I error ($\alpha$) set at 0.05, a statistical power of 0.9, and a within-subject correlation of measures of 0.5, the minimum required sample size was determined to be 30 subjects. To safely account for a potential 20% rate of patient dropouts and missing data, this requirement was increased to 38 subjects, which aligns with the planned cohort of approximately 40 patients. All sample size calculations were executed using R software, version 4.5.2, utilizing the pwrss package, version 1.0.0.