American Journal of Transplantation 2015; 15: 190–199 Wiley Periodicals Inc.
C
Copyright 2014 The American Society of Transplantation and the American Society of Transplant Surgeons doi: 10.1111/ajt.12947
Risk Factors of Pneumocystis Pneumonia in Solid Organ Recipients in the Era of the Common Use of Posttransplantation Prophylaxis y
y
X. Iriart1,2,3,4,*, , T. Challan Belval5, , J. Fillaux1, L. Esposito5, R.-A. Lavergne1,2,3,4, I. Cardeau-Desangles5, O. Roques5, A. Del Bello5, O. Cointault5, L. Lavayssie`re5, P. Chauvin1, S. Menard2,3,4, J.-F. Magnaval1, S. Cassaing1, L. Rostaing2,4,5, N. Kamar2,4,5,z and A. Berry1,2,3,4,z 1
Department of Parasitology-Mycology, CHU Toulouse, Toulouse, France 2 INSERM U1043, Toulouse, France 3 CNRS UMR5282, Toulouse, France 4 Centre de Physiopathiologie de Toulouse Purpan (CPTP), UPS, Universite´ de Toulouse, Toulouse, France 5 Department of Nephrology and Organ Transplantation, CHU Toulouse, Toulouse, France Corresponding author: Antoine Berry,
[email protected] or Xavier Iriart,
[email protected] y Both authors contributed equally. z Co-senior authors. Pneumocystis pneumonia (PCP) in solid organ transplant (SOT) recipients becomes rare in the immediate posttransplantation period thanks to generalized prophylaxis. We aimed to identify the predictive factors for PCP in the era of universal prophylaxis and to propose a strategy for preventing PCP beyond the first year after transplantation. In a retrospective case–control study, 33 SOT cases with PCP diagnosed between 2004 and 2010 were matched with two controls each to identify risk factors for PCP by uniand multivariate analysis. All the patients benefited from 6 months of posttransplantation trimethoprim– sulfamethoxazole prophylaxis. Most PCP in SOT patients occurred during the second year posttransplantation (33%). By univariate analysis, age, nonuse of tacrolimus, total and CD4 lymphocyte counts, gammaglobulin concentration and cytomegalovirus (CMV) infection appeared to be PCP risk factors. In the final multivariate analysis, age (adjusted odds ratio [OR] 3.7, 95% confidence interval [CI]: 1.3–10.4), CMV infection (OR: 5.2, 95% CI: 1.8–14.7) and total lymphocyte count (OR: 3.9, 95% CI: 1.4–10.7) were found to be independently associated with PCP. The second year posttransplantation appeared to be the new period of highest risk of PCP. Age, CMV viremia and lymphocytes were the most pertinent predictive criteria to evaluate the risk of PCP in clinical practice. 190
Abbreviations: AZA, azathioprine; BAL, bronchoalveolar lavage; CMV, cytomegalovirus; CsA, cyclosporine; Ct, cycle threshold; DNA, deoxyribonucleic acid; IA, immunofluorescence assay; MPA, mycophenolic acid; mTORi, mammalian target of rapamycin inhibitors; OR, odds ratios; PCP, Pneumocystis pneumonia; PCR, polymerase chain reaction; S, steroids; SOT, solid organ transplant; Tac, tacrolimus; TMP–SMX, trimethoprim–sulfamethoxazole Received 08 November 2013, revised 21 July 2014 and accepted for publication 22 July 2014
Introduction Pneumocystis jirovecii is an opportunistic fungus localized in the alveoli of the lungs that causes pneumonia. Pneumocystis pneumonia (PCP) remains a frequent cause of infection among immunocompromised patients (1,2). In solid organ transplant (SOT) recipients, PCP is associated with a high mortality rate, despite an adapted treatment (2–5). Before the institution of routine prophylaxis, the highest risk period for the development of PCP was the first 6 months after transplantation (6–12) with a PCP incidence of about 10% (5–15%) (3,13,14). PCP prophylaxis with trimethoprim– sulfamethoxazole (TMP–SMX) has proven highly effective in preventing PCP in SOT recipients (15,16) as it reduces the risk by 91% (17). Even if no consensus exists about the optimal duration of PCP prophylaxis, the generalized use of primary TMP–SMX prophylaxis in SOT recipients 6–12 months after transplantation has drastically decreased the incidence of this infection during this period (3,18). However, after an initial universal prophylaxis, there are no clear recommendations for a reinitiation of anti-PCP prophylaxis. Therefore, it is mandatory to identify risk factors for PCP that may occur after the first year posttransplantation. A threshold of risk (peripheral CD4þ T cell counts 22) (25) were considered as
American Journal of Transplantation 2015; 15: 190–199
colonized if an alternative diagnosis to PCP had been selected by the clinicians and if the patient presented a favorable evolution in the absence of a specific treatment for P. jirovecii. The colonized patients have been excluded from the analysis. The reference day (D0) for the case patients was defined as the day when microbiological evidence of Pneumocystis infection was obtained in BAL and the follow-up time was the time between transplantation and D0. Each positive case of PCP was matched with two control patients who were transplanted immediately after the index case. To avoid bias due to asymptomatic Pneumocystis infection or colonization, control patients must have undergone BAL with a negative diagnosis of P. jirovecii by specific PCR. The date of this BAL serves as the date of reference for the comparison with the case, to be sure that the data were collected when the controls were not infected or colonized with Pneumocystis. As the patients were hospitalized for this BAL to be carried out, they have generally benefited from a complete biologic and clinical evaluation. The BAL must have been performed at least 6 months after the transplantation when the control patients were no longer covered by PCP prophylaxis. Moreover, controls must have an approximately equal follow-up time without PCP (time between the transplantation and the negative BAL for Pneumocystis) as the matched case.
Analyzed data Demographics (sex, age and ethnic group), transplantation, Pneumocystis prophylaxis, immunosuppressive regimens (immediately after the transplantation and at D0) and CMV data were collected for both cases and controls. CMV infection was defined by the presence of a positive CMV viremia detected by PCR (26,27). Acute rejection episodes were also assessed within the study period, that is between the transplantation and D0. In addition, the following biological parameters were collected at and before D0: leukocyte count, lymphocyte count, CD4þ T cell count, serum creatinine level, estimated GFR using the MDRD equation, total gamma globulin concentration and CMV viremia.
Analysis and statistical methods Data were analyzed with SIGMA Stat1 (2.03; Jandel Corporation, San Jose, CA) software using the Mann–Whitney rank sum test for a two-group comparison. To compare patients’ characteristics in each group, we used the chi-square test or Fisher’s exact test as appropriate. Values were reported as the median and interquartile range (IQR) [25%; 75%]. Binary variables with p-value
