OBJECTIVE Accurate detection of lymph node metastases in prostate cancer has important implications for prognosis and approach to treatment. We investigated whether preoperative [18F]fluorocholine combined in-line positron emission tomography-computerized tomography and intraoperative laparoscopic radioisotope guided sentinel pelvic lymph node dissection can detect pelvic lymph node metastases in patients with clinically localized prostate cancer as reliably as extended pelvic lymph node dissection. METHODS A total of 20 patients (mean age 63.9 +/- 6.7 years, range 52 to 75) with clinically localized prostate cancer, prostate specific antigen greater than 10 ng/ml, and/or a Gleason score sum of 7 or greater and negative bone scan were enrolled in the study. [18F]fluorocholine combined in-line positron emission tomography-computerized tomography was performed before surgery. Sentinel pelvic lymph node dissection preceded extended pelvic lymph node dissection including the area of the obturator fossa, external iliac artery/vein and internal iliac artery/vein up to the bifurcation of the common iliac artery. Laparoscopic radical prostatectomy was performed afterward. RESULTS In 10 of the 20 patients (50%) lymph node metastases were detected, and were exclusively found outside the obturator fossa in 62%. These metastases would not have been identified with standard lymph node dissection of the obturator fossa only. [18F]fluorocholine combined in-line positron emission tomography-computerized tomography was true positive in 1, false-positive in 2, false-negative in 9 and true negative in 8 patients. The largest lymph node metastasis not seen with [18F]fluorocholine combined in-line positron emission tomography-computerized tomography was 8 mm. Laparoscopic sentinel guided lymph node dissection revealed lymph node metastases in 8 of 10 patients. In the other 2 patients sentinel lymph node dissection was not conclusive. In 1 patient normal nodal tissue was completely replaced by cancer and, therefore, there was no tracer uptake in the involved pelvic sidewall/node, and the other patient had no tracer activity at all in the involved pelvic sidewall. Extended pelvic lymph node dissection missed 1 lymph node metastasis (2 mm diameter near pudendal artery) which was detected by sentinel pelvic lymph node dissection only. CONCLUSIONS Extended pelvic lymph node dissection reveals a higher number of lymph node metastases as described for obturator fossa dissection only. [18F]fluorocholine combined in-line positron emission tomography-computerized tomography is not useful in searching for occult lymph node metastases in clinically localized prostate cancer. Sentinel guided pelvic lymph node dissection allows the detection of even small lymph node metastases. The accuracy of sentinel pelvic lymph node dissection is comparable to that of extended pelvic lymph node dissection when the limitations of the method are taken into consideration.