The impact of rituximab (R) on the incidence of central nervous system (CNS) relapse of diffuse large B-cell lymphoma (DLBCL) is unclear. We performed a meta-analysis to explore the risk factors and assess the association between R-chemotherapy (R-chemo) and CNS relapse. We searched MEDLINE, PubMed, EMBASE and OVID for eligible studies. Published group statistics were extracted from each study for analysis; individual patient data from each study were not accessed. Fixed- or random-effects models were used to estimate the odds ratio (OR) with a 95% confidence interval (CI). Eight studies were identified. The OR for R-chemo compared with identical chemotherapy was 0.70 (95% CI 0.54-0.91). Stage III/IV (OR 2.25, 95% CI 1.64-3.08), International Prognostic Index (IPI) > 1 (OR 2.62, 95% CI 1.59-4.33), performance status (PS) > 1 (OR 1.67, 95% CI 1.23-2.27), elevated lactate dehydrogenase (LDH) (OR 2.23, 95% CI 1.54-3.22), bone marrow involvement (OR 2.85, 95% CI 1.99-4.07), more than one extranodal involvement (OR 2.61, 95% CI 1.93-3.54), presence of B symptoms (OR 1.87, 95% CI 1.37-2.56) and testicular involvement (OR 3.83, 95% CI 1.84-7.97) were associated with increased risks of CNS relapse. This meta-analysis demonstrated a lower incidence of CNS relapse of DLBCL in the rituximab era. The risk of CNS relapse can be assessed by stage, IPI, PS, LDH, presence of B symptoms, number of extranodal sites, bone marrow and testicular involvement.