The BRAF T1799A mutation is a heterozygous point mutation and its reported prevalence in papillary thyroid carcinoma (PTC) has varied from 29 to 83%, with an overall mean of 44%. In Korea, the reported mutation rate reached 83% in PTC and 52% in micropapillary carcinoma. We hypothesized that the differences in prevalence may be influenced by the methods of mutation analysis, the sizes of tumor and ethnic differences. Three types of DNA samples from the same PTC mass (0.4-1.6 cm, mean 0.83 cm sized) per each patient (n=17) were isolated. The first type was obtained from frozen PTC tissues using laser-captured microdissection (Frozen-laser, n=17), the second was obtained from frozen tissue by manual tumor margin dissection using a blade (Frozen-blade, n=17) and the third was obtained from formalin-fixed, paraffin-embedded tissue by manual margin dissection (Paraffin-blade, n=15, 2 failed). The mutation rates of the three-matched DNA samples were compared by the SNP mode and AQ mode of pyrosequencing, and direct DNA sequencing. Both the AQ mode of pyrosequencing and the direct DNA sequencing detected the BRAF T1799A mutation in 100% of the 'Frozen-laser' samples, but the mutation was omitted in 1/17 of the 'Frozen-blade' samples and in 5/15 of the 'Paraffin-blade' samples, while the former was more rapid and objective than the latter. The SNP mode of pyrosequencing variably detected the mutation from 40 to 100%, and it showed the lowest sensitivity. Our results indicate that the reported prevalence of the BRAF T1799A mutation in PTC can be underestimated due the mutation analysis methods, and especially in small PTCs. The BRAF T1799A mutation may be an early and essential carcinogenic event in nearly all Korean PTCs, and even in micro-PTCs. For the accurate detection of the BRAF T1799A mutation from small PTCs, fresh or frozen tissues and more cautious microdissection are required, and the AQ mode of pyrosequencing assay is preferred.