Coil embolization of pulmonary arteriovenous malformations (PAVMs) has a high re-canalization/re-perfusion rate. Embolization with Amplatzer plugs has been previously described, but the long-term efficacy is not established. This study reports the experience of a referral medical center with the use of coils and Amplatzer plugs for treating PAVMs in patients with hereditary hemorrhagic telangiectasia. The study was approved by the Institutional Review Board with waiver of informed consent. The cohort included all patients who underwent PAVM embolization in 2004-2014 for whom follow-up imaging scans were available. The medical files were retrospectively reviewed for background data, embolization method (coils, Amplatzer plugs, both), and complications. Re-canalization of treated PAVMs was assessed from intrapulmonary angiograms (following percutaneous procedures) or computed tomography angiograms. Fisher's exact test and Pearson Chi-squared test or t test were used for statistical analysis, with significance at p<0.05. 16 patients met the study criteria. Imaging scans were available for 63 of the total 110 PAVMs treated in 41 procedures. Coils were used for embolization in 37 PAVMs, Amplatzer plugs in 21, and both in five. Median follow-up time was 7.7years (range 1.4-18.9). Re-canalization was detected in seven vessels, all treated with coils; there were no cases of re-canalization in plug-occluded vessels (p=0.0413). The use of Amplatzer plugs for the embolization of PAVMs in patients with hemorrhagic telangiectasia is associated with a significantly lower rate of re-canalization of feeding vessels than coils. Long-term prospective studies are required to confirm these findings.