Radioiodine ablation of thyroid tissue after subtotal thyroidectomy has been shown to decrease recurrence in certain subsets of patients with well-differentiated thyroid cancer. In a substantial percentage of cases (20-30%), initial ablation of the thyroid remnant fails, necessitating a second treatment. The factors associated with ablation failure are not fully understood. In particular, it is not certain whether the use of doses higher than 3.70 GBq would result in any additional benefit, or whether there is a 'stunning' effect of the diagnostic dose of 131I on the subsequent ablation rate. A retrospective analysis was performed of all patients (n=389) with well-differentiated thyroid cancer treated at our institution between 1992 and 2001. Remnant ablation success was determined by a whole-body radioiodine scan. The following factors, thought to be associated with thyroid remnant ablation, were studied by logistic regression analysis: age, gender, tumour histology, stage, pre-therapy neck uptake of 131I, diagnostic dose, ablation dose, time between diagnostic and therapeutic dose (T1), time between therapeutic administration and the first follow-up whole-body scan (T2) and the thyroid-stimulating hormone (TSH) level measured at the time of therapy. Follow-up whole-body scans were available in 214 patients. We found no association with age, gender, histology, TSH level, neck uptake, diagnostic dose and successful ablation. The therapeutic dose was the only variable found to be associated with success (odds ratio, 1.96 per 1.85 GBq increment; 95% confidence interval, 1.11-3.46). Our results confirm the presence of a significant percentage of ablation failures (24.4%) despite the use of high ablative doses (3.70-7.40 GBq). Higher therapeutic doses are associated with higher rates of successful ablation, even when administered to patients with more advanced stages. Using our protocol, higher diagnostic doses were not associated with higher rates of ablation failure.