Introduction - There is an increasing evidence on efficacy of drug coated balloons(DCB) in reducing restenosis and reocclusion after endovascular treatment of peripheral arterial disease(PAD), especially in the femoro-popliteal district but DCB use is discouraged by its high cost. The aim of the study was to identify subgroups of patients with chronic limb threatening ischemia (CLTI) that can benefit more from DCB than others and thus, perform a better resources management. Methods - A propensity score matched analysis was performed on a prospectively maintained database, including all patients treated for PAD between January-2014 and December-2017. Inclusion criteria were: CLTI, infrainguinal PAD and endovascular treatment. Patient's demographics and clinical characteristics were assessed. Limbs were stratified according to WIfI clinical stages. Clinical and Duplex follow-up was performed at 3,6 and every 6 month thereafter. A propensity score analysis was used to match all DCB treated patients to balloon angioplasty(BA) patients in a 1:1 nearest neighbour matching method. The matching variables were: age, diabetes, Rutherford stage, treated arterial segment (femoral, tibial and femoro-tibial) and type of lesion (primary vs. restenosis/reocclusion). Endpoints were: Limb Salvage(LS), Primary(PP), Assisted(AP) and Secondary Patency(SP), Patient Survival(S), wound healing (WH) and identification of subgroups susceptible to DCB on the basis of Kaplan-Meier analysis and Log-Rank test. Results - A total of 211 patients satisfying the inclusion criteria were treated in the study period. DCB was used in 60 cases (median age 72 years, male 76.7%). Sixty (median age 73 years, male 66.7%) out of 151 cases treated by POBA were selected, based on the propensity score matching. DCB and POBA groups were similar in terms of Coronary artery disease (47% Vs 50%. P=0.43), insulin-dependent diabetes(IDD) (40% Vs 27%. P=0.09), kidney disease (45% Vs 60%. P=0.07), Rutherford stages 5/6 (87% Vs 88%. P=0.41), and WIfI clinical stages (stage 3: 37% Vs 42%; stage 4: 31% Vs 27%; P=0.92). In the overall population, femoral, tibial and femoro-tibial lesions were treated in 44%, 49% and 7%, respectively. Lesion type was primary in 73% and restenosis/reocclusion in 27%. The mean follow-up was 21.5 months. WH was 54%, 75% and 85.7% at 6,12 and 24-month, respectively. LS was 87.2%, 86.1% and 84.7% at 6,12 and 24-month, respectively. PP, AP, SP, and S are reported in table 1. At Cox-regression analysis, IDD was an independent negative predictor of LS (HR=3.4, P=0.011). Within IDD patients, at 1-year, those treated by DCB had significantly higher rates of LS (81.1% Vs 56.3%, P=0.05), AP (80% Vs 50%, P=0.04) and SP (85% Vs 56%, P=0.029).Table 1endpoints at 6,12 and 24-monthPP (%)AP (%)SP (%)S (%)6-month8081.885.389.812-month73.575.183.185.624-month6872.481.174.6 Open table in a new tab Conclusion - The study results suggest a more beneficial effect of DCB in the treatment of infrainguinal PAD associated with CLTI in insulin-dependent diabetes patients. In a limited resources assets, insulin-dependent patients could represent a reasonable target for DCB technology.