Prostate cancer: To scan or not to scan for lymph node involvement?

Abstract

Objective. When diagnosing lymph node involvement in prostate cancer (PC), the available imaging techniques are considered to be of doubtful value but are still in common use in many institutions because of a lack of better alternatives. To help us choose between methods while we wait for improvements in technologies, we evaluated the performance of ultrasonography (US), CT and MRI. Material and methods. We reviewed the records of all PC patients who were potential candidates for a surgical staging procedure prior to treatment with intent to cure at Skejby Sygehus, Denmark between 1996 and 2004. In total, 227 patients met our inclusion criteria and 149 were examined with US, 80 with MRI and 71 with CT. The performance of the three modalities was analysed. Results. A total of 50 patients (22%) had lymph node metastases (pN1) and only 12 of these cases (24%) were detected by imaging. US identified all 12 pN1 patients, MRI five and CT did not identify any. Moreover, US missed 28 patients with pN1 metastases, MRI 18 and CT 13. Consequently, US, MRI and CT had sensitivities of 0.3, 0.22 and 0, respectively, while all three modalities had a specificity of 1. The 12 patients identified by imaging had clinically advanced and pathologically aggressive cancers. Conclusions. Available imaging techniques showed moderate to very low sensitivity. US proved superior to MRI and CT, but should only be applied in high-risk patients. In particular, CT can be considered unsuitable for diagnosing pelvic lymph node metastases in PC.

• Lymph nodes
• metastases
• prostate cancer
• ultrasonography
• CT
• MRI