Objective To evaluate the value of 18F-FDG PET for finding of bone marrow infiltration in patients with lymphoma. Methods Data on the diagnostic performance of 18F-FDG PET in bone marrow infiltration of lymphoma from MEDLINE and EMBASE database were combined quantitatively across eligible studies. The weighted summary sensitivities, specificities and accuracy were separate analyses according to various subgroups. Bone marrow biopsy (BMB) was used as the reference standard. Results Twenty-seven studies were identified, thirteen of them were eligible, which enrolled a total of 587 patients, were included in the meta-analysis. The independent random-effects weighted estimates of sensitivity and specificity against BMB were 51% (95% confidence interval [CI], 38%–64%) and 91% (95% CI, 85%–95%) respectively. The weighted positive likelihood ratio (LR_) was 5.75 (95% CI, 348–9.48) and the negative likelihood ratio (LR_) was 0.67 (95% CI, 0.55–0.82). Subgroup analyses showed better sensitivity in patients with Hodgkin’s disease and in aggressive histological types of non-Hodgkin’s lymphoma than in patients with less aggressive histological types. Conclusion This meta-analysis showed that 18F-FDG PET might be good, but not excellent, concordance with the results of BMB for the detection of bone marrow infiltration in patients with lymphoma. 18F-FDG PET may be complement to the results of BMB and its performance may vary with the type of lymphoma.