Objectives: We aim to describe the efficacy, safety, and characteristics of the Amplatzer Vascular Plug (AVP) II and IV "off-label" use for multiple cardiovascular occlusions in children under 10 years. Methods: Observational retrospective multicenter (2007-2020, 6 centers) review of paediatric procedures using AVP II or IV. Results: A total of 125 children (49.6% aged ≤ 1 year, 147 lesions) underwent 136 successive procedures (success rate: 98.5%) using 169 devices (109 AVP IV, 60 AVP II). The mean device diameter was 7.7 ± 3.2 mm (4-20 mm). The median AVP size to vessel diameter ratio was 1.3 (0-2). The median age and weight at implantation were 1.0 year (0.01-9.98) and 8.4 kg (1-69). Procedures were heterogeneous (55 patent ductus arteriosus (PDA), 28 collaterals, 18 sequestrations, 22 arteriovenous/veinovenous/coronary fistulas, 6 vertical veins, 6 conduits, 5 ventricular septal defects, 7 miscellaneous). Day 1 and 6-month occlusion rates were respectively 94.8% and 98.5%. Major adverse events (MAE) occurred in 5.2% of cases (no procedure-related deaths), and more frequently in weight ≤ 5 kg (p = 0.01), younger patients (p = 0.03) during PDA closure (p = 0.02) of tubular types (p = 0.02) using larger devices (p = 0.03) and AVP II (p = 0.003). Independent predictor of MAE risk was a higher AVP diameter to patient weight ratio (Odds-ratio: 2.33, 95% confidence interval 1.31-4.13, p = 0.004, optimal cut off: 1.45). Conclusions: Both AVPs are safe and effective for percutaneous occlusions in children under 10. Such devices represent an alternative "off label" use for well selected paediatric patients.