We previously demonstrated that (1) most papillary microcarcinomas can be followed without surgical treatment and (2) when surgery is performed, patients with lateral lymph node metastasis detected on preoperative ultrasonography (US) are more likely to develop recurrence. In this study, we further investigated the application of these strategies. To date, we have observed 211 patients (average follow-up 47.9 months). In more than 70% of these patients the tumor size did not increase during the follow-up period. There were no clinicopathologic features linked to tumor enlargement except in tumors > or = 7 mm, which tended to enlarge in patients followed for 4 years. To evaluate not only whether observation can continue but also how to dissect the lymph nodes optimally at surgery, US diagnosis for lateral node metastasis is essential because the presence of US-diagnosed lateral metastasis is an even stronger predictive marker for recurrence than the presence of pathologically confirmed node metastasis. The positive predictive value (PPV) was 80.6% for US but reached 100% if fine-needle aspiration biopsy (FNAB) of nodes or FNAB-thyroglobulin measurement is added. Furthermore, carcinomas occupying the upper region of the thyroid more frequently showed US-diagnosed and pathologically confirmed lateral metastasis, and those measuring > or = 7 mm were more likely to show pathologically confirmed lateral metastasis. These findings suggest that, for papillary microcarcinoma: (1) US-diagnosed lateral metastasis is a strong marker predicting a worse relapse-free survival; (2) FNAB of nodes and FNAB-thyroglobulin measurement are useful tools for evaluating lymph node metastasis; and (3) careful US evaluation for lateral metastasis is necessary in patients with a tumor measuring > or = 7 mm or that is located in the upper region of the thyroid both during observation and preoperatively.