Introduction: Discriminating among benign chondroid tumors, low-grade chondrosarcomas, and grade 2/3 chondrosarcomas is frequently difficult with standard imaging modalities. We systematically reviewed the literature to determine the performance of PET-CT in making this distinction.
Methods: A systematic review was performed identifying 811 PubMed- and Embase-indexed articles containing combinations of "chondrosarcoma," "enchondroma," "chondroid," "cartilage" and "PET/CT," "PET," "positron." Eight articles including 166 lesions were included. Age, gender, tumor size, histologic grade, and SUVmax values were extracted for individual lesions when possible and otherwise recorded as aggregated data. Comparisons in SUVmax among benign, low-grade, and intermediate-/high-grade chondroid neoplasms were made.
Results: Individual SUVs were available for 101 lesions; 65 additional lesions were reported as aggregated data. There were 101 malignant and 65 benign tumors. Benign tumors were seen more frequently in females (p = 0.04, Fischer's exact test), but malignancy was not associated with age or lesion size. SUVmax was lower for benign (1.6 ± 0.7) than malignant tumors (4.4 ± 2.5) (p < 0.0001, t-test). SUVmax was lower for grade 0/1 (2.0 ± 0.7) than grade 2/3 (6.0 ± 3.2) (p < 0.0001, t-test). Increasing SUVmax correlated with higher grade chondroid tumors (Spearman's rank, ρ = 0.78). SUVmax ≥4.4 was 99% specific for grade 2/3 chondrosarcoma.
Conclusions: SUVmax correlates with histologic grade in intraosseous chondroid neoplasms; very low SUVmax supports a diagnosis of benign tumor, while elevated SUVmax is suggestive of higher grade chondrosarcoma.
Keywords: Chondroid; Chondrosarcoma; Enchondroma; FDG; Pet/Ct; SUVmax.