Objectives: The goal of this meta-analysis was to estimate the diagnostic accuracy as measured by sensitivity and specificity of FDG-PET in the diagnosis of primary bone and soft tissue sarcomas. . Subjects and Methods: Several databases including PubMed, Embase, Cochrane library and Web of Science were searched. In addition to sensitivity and specificity, the diagnostic accuracy region for detection and grading of sarcomas were pooled using Hierarchical Summary Receiver Operating Characteristic (HSROC) models. Subgroup analysis included pooling soft tissue and bone sarcoma separately, and sensitivity analysis included the high-quality studies. The quality of eligible studies was assessed using QUADAS-2. Results: Of the 1258 articles screened, 21 studies satisfied the inclusion criteria. The pooled estimate of the sensitivity and specificity of FDG-PET combined with CT for the detection of sarcomas were 89.2% (95% CI: 83.6-93.1%) and 76.3% (95% CI: 69.8-81.9) respectively. These diagnostic accuracy measures were higher when combined with CT than those of PDG-PET alone. Diagnostic accuracy for bone and soft tissue lesions were comparable, but slightly better for soft tissue tumors. Pooling only the high-quality studies with low risk of bias yielded a sensitivity of 88.5% and specificity reduced to 65.6%. There was no evidence for publication bias, but significant heterogeneity (P<0.001) among the studies was apparent. This study also showed that FDG-PET can efficiently differentiate between benign and malignant tumors, with weighted mean SUVmax of 2.52 units in benign and 6.81 units in malignant tumors. Conclusion: Our findings indicate FDG-PET combined with CT can efficiently differentiate between benign and malignant bone and soft tissue tumors, being a potentially useful screening tool for soft tissue sarcomas and combining with CT improves the diagnostic accuracy.