BACKGROUND Debate is increasing on whether mycophenolic acid (MPA) provides survival benefits comparable to azathioprine (AZA) after renal transplantation. METHODS This retrospective cohort study compared safety and efficacy of AZA (n = 662) vs. MPA (n = 267) in low-immunologic-risk kidney transplant recipients (KTR) receiving tacrolimus (TAC) and steroids between 1998 and 2007. Primary outcomes were treatment discontinuation and infection. Secondary endpoints included survival free from biopsy-proven acute rejection, graft loss, death, and renal function. RESULTS The 5-year survival free of treatment discontinuation was higher in the MPA compared to the AZA group (74.1% vs. 60.3%, P < 0.001). MPA was discontinued exclusively because of adverse events (16.4%), while AZA was discontinued primarily for lack of efficacy (21.2%). In univariable analysis, MPA was associated with higher incidence of total (561.5 vs. 667.5 episodes/1000 person-year, P < 0.001), bacterial (167 vs. 158 episodes/1000 person-years, P = 0.001), and viral infections (83.2 vs. 100.4 episodes/1000 person-years, P = 0.001), but this association was not confirmed in multivariable analysis. Over 29% of viral infections in the AZA group occurred after conversion to MPA. A high incidence of tuberculosis was observed (2.9 episodes/1000 person-years) with a higher incidence (but not a statistically significant difference) in the AZA group. No significant differences were found in patient survival (90% vs. 89%, P = 0.78) or graft survival (81% vs. 77.7%, P = 0.08), but infection accounted for >50% of all deaths. CONCLUSIONS The type of antimetabolite, AZA or MPA, was not independently associated with any safety or efficacy outcome 5 years after transplantation, suggesting that AZA is still a viable option for low-risk KTR receiving TAC and steroids.