Which kinds of lymph node metastases can FDG PET detect? A clinical study in melanoma.

UNLABELLED

The purposes of this study were to establish the diagnostic accuracy of FDG PET for lymph node metastases and to determine the smallest detectable volume of disease.

METHODS

Using FDG PET, we preoperatively studied 56 lymph node basins in 38 patients with a clinical or instrumental diagnosis of lymph node metastases from melanoma. All lymph node basins underwent node dissection. The FDG PET results were compared with the postoperative histopathology results. PET images were obtained using a GE 4096 WB scanner, after injection of a mean activity of 496 MBq (range, 366-699 MBq) of FDG.

RESULTS

The efficacy of FDG PET in the diagnosis of involved lymph node basins was good. Sensitivity was 95% (35/37); specificity, 84% (16/19); accuracy, 91% (51/56); positive predictive value, 92% (35/38); and negative predicative value, 89% (16/18). Metastases were shown histologically in 114 of 647 surgically removed lymph nodes. FDG PET detected 100% of metastases > or = 10 mm, 83% of metastases 6-10 mm, and 23% of metastases < or = 5 mm. Moreover, FDG PET had high sensitivity (> or = 93%) only for metastases with more than 50% lymph node involvement or with capsular infiltration.

CONCLUSION

Our study shows that FDG PET has a reasonable sensitivity and specificity for detecting the presence or absence of lymph node metastases in patients with melanoma. However, even if able to detect small volumes of subclinical macroscopic disease, FDG PET cannot detect subclinical microscopic disease with acceptable sensitivity. The specificity of FDG PET is good, but some false-positive results may occur.