Abstract
Pneumocystis jiroveci pneumonia (PCP) is frequently reported in lymphoma patients treated with rituximab-contained regimens. There is a trend toward a difference in PCP risk between bi- and tri-weekly regimens. The aims of this systemic review and meta-analysis were to estimate the risk for PCP in these patients, compare the impact of different regimens on the risk, and evaluate the efficacy of prophylaxis. The cohort studies with incept up to January 2014 were retrieved from the Cochrane Library, Medline, Embase, and Web of Science databases. Studies that compared the incidence of PCP in patients with and without rituximab treatment were conducted. Studies that reported the results of prophylaxis were concentrated to evaluate the efficacy of prophylaxis. Fixed effect Mantel-Haenszel model was chosen as the main analysis method. Funnel plots were examined to estimate the potential selection bias. Egger's test and Begg's test were used for the determination of possible small study bias. Eleven cohort studies that met the inclusion criteria were finally included. Results indicated that rituximab was associated with a significantly increased risk for PCP (28/942 vs 5/977; risk ratio: 3.65; 95% confidence interval 1.65 to 8.07; P=0.001), and no heterogeneity existed between different studies (I2=0%). Little significant difference in PCP risk was found between bi-weekly and tri-weekly regimens (risk ratio: 3.11; 95% confidence interval 0.92 to 10.52, P=0.068). PCP risk was inversely associated with prophylaxis in patients treated with rituximab (0/222 vs 26/986; risk ratio: 0.28; 95% confidence interval 0.09 to 0.94; P=0.039). In conclusion, PCP risk was increased significantly in lymphoma patients subjected to rituximab-contained chemotherapies. Difference in PCP risk between bi-weekly and tri-weekly regimens was not significant. Additionally, prophylaxis was dramatically effective in preventing PCP in rituximab-received lymphoma patients, suggesting that rituximab should be recommended for these patients.