BACKGROUND Extracorporeal photopheresis (ECP) is an immunosuppressive treatment that involves leukocyte apheresis, psoralen and UV light treatment, and subsequent reinfusion. Patients treated with ECP are usually immunosuppressed. Bacterial contamination therefore poses a much unwanted risk, but incidence data are lacking. PATIENTS AND METHODS We screened all 1922 consecutive ECP procedures scheduled within a roughly 3-year period for eligibility. Those with missing data on ECP method (inline or offline) or type of venous access (peripheral or central) were excluded. ECPs with complete aerobic and anaerobic microbial testing of baseline patient blood samples (n = 1637) and of ECP cell concentrates (n = 1814) were included in the analysis. RESULTS A test for microbial contamination was positive for 1.82% of the cell concentrates, with central venous access was the most significant risk factor for the contamination (odds ratio = 19). Patient blood samples were positive in 3.85% of cases, but no patients became septic. Staphylococcus spp. were most abundant, and products with bacterial contamination did not cause side effects after reinfusion. There were no significant differences in contamination rates between inline and offline ECP. CONCLUSION These findings stress the importance of sterile procedures and the benefits of using peripheral over central venous access for reducing the risk of bacterial contamination in ECP.