Objective: In patients with differentiated thyroid cancer (DTC), the stimulated serum thyroglobulin (Tg) level at radioiodine ablation is a known predictive factor of persistent disease. This prognostic value is based on data obtained after thyroid hormone withdrawal (THW), but little is known about this prognostic value after recombinant human TSH (rhTSH) stimulation and about the relationship between the stimulated Tg level and the burden of persistent tumor. We aimed to assess the impact of both radioiodine preparation modalities and persistent tumor burden on stimulated Tg levels. Design and methods: The stimulated Tg level was measured at radioablation in 308 consecutive DTC patients without serum Tg antibodies. Of these, 123 (40%) were prepared with rhTSH and 185 with THW. Post-ablation scintigraphy included total-body scan and neck and thorax single photon emission computed tomography with computed tomography (SPECT-CT). During a mean follow-up of 43 months, persistent/recurrent disease (PRD) was found in 56 patients (18%). PRD was considered structural in the presence of lesions O1 cm and nonstructural otherwise. Results: Nonstructural PRD was more frequent in the rhTSH group than in the THW group (64 vs 26%, PZ0.01). Stimulated Tg levels were lower after rhTSH than after THW in patients with (13.5 vs 99.5 ng/ml, P!0.01) and without (1.2 vs 3.2 ng/ml, P!0.001) PRD. Also, Tg levels were lower in nonstructural disease than in structural disease in both rhTSH (3.8 vs 127.0 ng/ml, P!0.01) and THW (13.0 vs 143.5 ng/ml, P!0.0001) patients. The best Tg cutoff to predict PRD was 2.8 in rhTSH and 28 ng/ml in THW patients. Conclusion: Both radioiodine preparation modalities and the burden of persistent tumor affect the stimulated Tg level at ablation.