We retrospectively analyzed the regions and perioperative outcomes associated with lymph node dissection in patients with prostate cancer. Of 543 patients who underwent robot-assisted radical prostatectomy for prostate cancer with or without lymph node dissection according to the modified D'Amico criteria, 333 (61.3%), 128 (23.6%), and 82 (15.1%) were classified into the non-dissection, limited dissection, and extended dissection groups, respectively. Lymph node metastasis was identified in eight patients : one in the limited dissection group and seven in the extended dissection group. Notably, all eight biopsies showed Gleason scores of 4＋4 or higher, and the initial prostate-specific antigen (PSA) concentration was ≥10 ng/ml in seven of these patients. Lymph node metastasis was detected in areas other than the obturator lymph nodes in five patients (62.5%). Although there was no significant difference in the rate of Clavien- Dindo grade ≥II complications among the three groups, six patients (7.3%) in the extended dissection group developed infectious lymphoceles. In the extended dissection group, the PSA progression-free survival (PSA-PFS) was significantly shorter in patients with than in those without lymph node metastasis (p＜0.001). Because lymph node metastases were rare in the limited dissection group in our cohort of patients with a high risk of localized prostate cancer, achieving a diagnosis seems difficult with limited dissection. By contrast, in the extended dissection group, the PSA-PFS prognosis was significantly worse in lymph node-positive cases. Therefore, considering the high complication risk of lymphoceles, extended dissection should be performed in patients with a high likelihood of lymph node metastasis.