Risk factors for Pneumocystis carinii pneumonia in kidney transplant recipients: a case-control study

Abstract

Objective: To analyze risk factors for Pneumocystis carinii pneumonia (PCP) in kidney transplant recipients.

Study design: In a case-control study, 17 PCP cases diagnosed between July 1994 and July 2000 were matched with two controls each (previous and subsequent kidney transplant recipients who did not develop PCP during the same follow-up period). Demographics, organ origin, human leukocyte antigen (HLA) mismatches, use of poly- or monoclonal anti-CD3 antibodies (Po/MoAb) for induction or rejection treatment, rejection episodes, cumulative steroid dose for rejection treatment, immunosuppressive regimens, and other infections were analyzed.

Results: No significant differences were seen in gender (male 10 vs. 15), mean age (39.7 vs. 35.4 years), organ origin (cadaver donor 13 vs. 19), HLA mismatches, or Po/MoAb use in induction treatment. Significant differences were observed in PCP cases for rejection history (P=0.02), and median and total number of rejection episodes (P=0.0018). The relative risks for PCP for 1, 2, and > or =3 rejection treatments vs. no such treatment were 1, 1.05, and 6.30, respectively (P=0.021). The relative risk for PCP for steroid-resistant rejection was 4.34 (95% confidence interval [CI], 1.04-18.89) (P=0.019), and that for the use of Po/MoAb for rejection treatment was 7.23 (95% CI, 1.28-49.34) (P=0.006). The relative risk for PCP for 0, 1, and > or =2 previous or concomitant cytomegalovirus (CMV) infection vs. no such infections were 1.0, 2.32, and 13.0, respectively (P=0.012). The relative risks for PCP for tuberculosis (TB) was 18 (95% CI, 1.76-852.03), that for bacterial pneumonia was 14.22 (95% CI, 2.16-150.23), and that for hepatitis C virus infection was 5.25 (95% CI, 1.03-28.91). Immunosuppressive regimens with tacrolimus, mycophenolate mofetil (MMF), steroids (P=0.06), and MMF as a single variable (P=0.05) were more frequently used in cases. Primary trimethoprim-sulfamethoxazole prophylaxis failure was observed in 12 patients in association with heavy immunosuppression and concomitant infections.

Conclusions: The risk of PCP in kidney transplant recipients is related to the number and type of rejection treatments. It is also related to the occurrence of CMV infection, and to other immunomodulating infections such as TB and hepatitis C, and might also be increased with the use of newer and more potent immunosuppressive agents. Primary prophylaxis failure may occur in association with some of these risk factors.