Background and study aim: Argon plasma coagulation (APC) has been used to ablate dysplastic and nondysplastic Barrett's esophagus. We determined the longer-term efficacy of APC ablation within two randomized controlled trials of APC versus surveillance for Barrett's esophagus in patients in whom gastroesophageal reflux was controlled by either surgery or proton pump inhibitors. Patients and methods: 129 patients (surgical trial 70, medical trial 59) with Barrett's esophagus (nondysplastic or low grade dysplasia) were randomly allocated to either ablation using APC or to continuing endoscopy surveillance. Outcomes were determined at three time points: short-term (12 months), mid-term (42 - 75 months) and long-term (> 84 months). Results: In the APC groups, initial ablation of > 95 % of the Barrett's esophagus was achieved in 61 of 63 patients; the > 95 % ablation persisted in 47 of 56 patients at short-term follow-up, in 33 of 49 at mid-term and in 21 of 32 at long-term follow-up. In the surveillance groups, the length of Barrett's esophagus reduced from a mean of 4.2 cm to 2.7 cm at long-term follow-up. High grade dysplasia (HGD) developed in 1 patient in the APC groups and in 3 in the surveillance groups. Low grade dysplasia developed in 1 APC patient and in 6 surveillance patients. Conclusions: APC ablation reduced the extent of Barrett's esophagus, and this reduction was maintained in some patients at longer-term follow-up. However, progression to HGD can still occur despite APC ablation, suggesting endoscopic surveillance is still required. Clinical trial registration: ACTRN012607000293460 and ACTRN12607000292471 (Australian Clinical Trials Registry).