Pneumocystis jirovecii Pneumonia Outbreaks in Renal Transplant Populations in Quebec, Canada : Clinical and Molecular Epidemiology Sébastien Poulin, MD, MSc 1* , Simon-Frédéric Dufresne, MD 2 , Philippe J. Dufresne, Ph.D., RMCCM 3 and Alex Carignan, MD, MSc 1 1 Département de microbiologie et d’infectiologie, Université de Sherbrooke, Sherbrooke, QC, Canada; 2 Microbiologie et Infectiologie, Université de Montréal, Montréal, QC, Canada 3 Institut national de santé publique du Québec, Laboratoire de santé publique du Québec, Sainte-Anne-de-Bellevue, QC, Canada; Results Introduction : Is Pneumocystis jirovecii a contagious fungus ? Acknowledgements Methodology Highlights of results Background: In 2013 and 2014, four Pneumocystis jirovecii pneumonia (PJP) outbreaks occurred among renal transplant recipient cohorts in three academic hospitals in Quebec, Canada. We investigated the outbreaks with an aim to: 1) determine the risk factors associated with development of PJP and 2) elucidate the likely mode of transmission. Methods: A case-control study was conducted. Cases were chosen from among patients who had undergone kidney transplantation in one of the three centers and had a confirmed diagnosis of PJP during the outbreak periods. Controls were randomly selected from among all the patients who were followed up at outpatient transplant clinics during the outbreak periods. Bronchoalveolar lavage samples from 15 cases were available and used for Pneumocystis jirovecii (PJ) genotyping by multilocus sequence typing. Air samples were collected for PJ DNA (mitochondrial large subunit RNA gene) detection using real-time polymerase chain reaction (PCR). Results: A total of 22 cases and 88 controls were identified. Genotyping confirmed isolate clonality within each outbreak. PJP cases occurred at a median time of 112 months after transplantation. None of the patients were receiving targeted PJP prophylaxis at the time of diagnosis. Independent risk factors for PJP development included administration of anti-thymocyte antibodies after the transplant (adjusted odds ratio [aOR] 4.73, 95% confidence interval [95% CI] 1.12-19.89), glomerular filtration rate <30 mL/min (aOR 4.65, 95% CI 1.22-17.78), and absolute lymphocyte counts below 1,000 (aOR 5.53, 95% CI 1.40-21.94). Evidence from contacts in the outpatient clinic showed an increased tendency for PJP patients to have attended the outpatient clinic on the same day as an individual who developed PJP in the future (within 2 months; OR 1.56, 95% CI 0.59-4.13). Air samples from the outpatient transplant clinic of one center also showed the presence of PJ DNA. Conclusion: The development of PJP is associated with host risk factors that translate to increased immunosuppression. Collectively, the molecular and epidemiological data support person-to-person transmission of the disease, possibly via airborne particles. This study underlines the need for prophylaxis in certain contexts and emphasizes the importance of avoiding close contact between patients with PJP and other susceptible hosts. Conclusion and future direction Abstract Objectives of the study Poster # 343 To determine the risk factors associated with the development of PJP in outbreaks among renal transplant recipient cohorts To elucidate the likely mode of transmission We identified 22 cases and 88 controls. Genotyping confirmed isolate clonality within each outbreak. Independent risk factors for PJP: anti-thymocyte antibodies after the transplant, GFR <30 mL/min, and absolute lymphocyte count <1,000 (dose-response effect). Increased tendency for patients to have attended the outpatient clinic on the same day as an individual who developed PJP in the future (within 2 months) (not statistically significant). None of the patients were receiving targeted PJP prophylaxis at the time of diagnosis. Air samples obtained from the outpatient transplant clinic in one center showed the presence of PJ DNA in samples. The development of PJP is associated with host risk factors that result in increased immunosuppression. Collectively, the molecular and epidemiological data support person-to-person transmission of the disease, possibly via airborne particles. This study underlines the need for prophylaxis in certain contexts and emphasizes the importance of avoiding close contact between patients with PJP and other susceptible hosts. Universal and targeted prophylaxes are acceptable strategies for outbreak management. * Corresponding author: [email protected] Case-control study Genotyping Air sampling Case-control study Approved by all three ins2tu2onal ethics boards: Hôpital MaisonneuveRosemont (H1), Hôpital NotreDame de Montréal – CHUM (H2), Centre Hospitalier Universitaire de Sherbrooke CHUS (H3) Popula2on: Renal transplant recipients who visited the outpaGent transplant clinic at any of the three centers (H1, H2, and H3) at least once during the outbreak periods in 2013/2014 Study period: Two months aOer the last PJP outbreak Cases: Transplant paGents with a confirmed diagnosis of PJP during the outbreak periods Controls: Randomly selected (unmatched) transplant paGents without PJP during the outbreak periods; four controls per case Sta2s2cs: Univariate logisGc regression analysis followed by a mulGvariate model built in a stepwise fashion that incorporates variables with P<0.10 in the univariate analysis and/ or biological plausibility. All analyses performed with Prism®6 (GraphPad SoOware, Inc.) and Stata®13 (StataCorp LP). . H1 : 6/6 and 3/3 BAL H2 : 2/3 BAL H3 : 3/10 BAL 3 MLST targets - mt26S – MtLSU - ITS1 - CYB Coriolis® Air Sampler Real-Time PCR : mtLSU rRNA 100 99 98 97 96 95 94 93 92 91 MY080095 MY080097 47210 MY078051 MY079216 MY079217 MY079218 MY078052 MY079175 MY079179 MY079181 MY079183 MY079187 MY079177 REFERENCE MAITTE MY081075 MY081076 MY081077 MY078050 MY078053 100 99 98 97 96 95 94 93 92 91 100 99 98 97 96 95 94 93 92 91 1 mL BAL 5 min, 10 000g NucliSens DNA extraction (Biomérieux) Real-Time PCR SYBr Green DNA purification Sequencing Assembling Assignation of genotypes Machine disposition: 1 m above floor at different locations (see results) Parameters: 300 L/min x 5min (1.5m 3 ) Protocol : 15 mL sterile PBS with 0.0002% Tween 80 Vortex Centrifugation 2500 g x 10 min Resuspend in 1 mL PBS with 0.0002% Tween 80 Some recommend strict hospital segregation of immunocompromised PJP patients and the use of facemasks to prevent transmission among infected individuals. -However, formal recommendations regarding infection control in hospitals or healthcare clinics cannot be made yet. -Prophylaxis in susceptible patients is effective at preventing infection. -Pneumocystis jirovecii (PJ) is an opportunistic fungus without any known environmental reservoir. -Its precise mode of transmission is still unknown, and the previously held theory of latency and reactivation is now uncertain. -A theory of intermittent colonization (or infection/re-infection) and transmission has been proposed. -More than 30 outbreaks of PJP have been described in the literature since the 1980s. -The population of renal transplant recipients is overrepresented for yet unknown reasons. -Most studies are small and observational, not always supported by molecular evidence, and rarely involve a case- control or cohort study. RESULTS (continued) - No difference in median doses of immunosuppressive drugs for cases and controls - None of the PJP cases were on anti-PJ prophylaxis - No anti-PJ prophylaxis was prescribed with corticosteroids and/or with acute rejection of graft - All four outbreaks rapidly stopped with various interventions: universal prophylaxis of the cohort (outbreak B and D), contact prophylaxis (outbreak C) and prolongation of prophylaxis in new transplant recipients and those with acute rejection (outbreak A). In 2013 and 2014, four outbreaks involving a total of 22 cases occurred among renal transplant recipient cohorts in three academic hospitals in Quebec (Canada). A case-control and molecular epidemiology study was pursued for further investigations. * Dissociation curves showed specific amplifications for all specimens We thank Lyne D’Amour (LSPQ), Dr Marie Gourdeau (CHUQ), Dr Claude Lemieux (CHUM), Dr Jacques Pépin (CHUS) and Dre Cécile Tremblay (CHUM) Targets with high discriminatory power according to Maitte et al, JCM, 2013 Protocol of PJ DNA detection according to Choukri et al, CID, 2010 Figure 2 : Phylogenetic tree of cases Figure 1 : Epidemic curves Table 1 : Basic characteristics (univariate analysis) Controls Cases OR 95% CI p n (%) n (%) Age (years, median) 57 56 0.22 Sex Women 42 (48) 15 (68) 1 Men 46 (52) 7 (32) 0.43 0.16-1.17 0.10 Time since transplantation (months, median) 112 112 0.38 Type of graft Cadaveric 68 (79) 18 (82) 1 Living 18 (21) 4 (18) 0.84 0.25-2.81 0.78 Anti-lymphocytes antibodies at time of transplantation 7 (10) 7 (39) 5.45 1.48-20.05 0.009* Duration of hemodialysis before transplantation (months, median) 21 23 0.54 Cellcept 52 (60) 11 (50) 0.67 0.26-1.73 0.41 Tacrolimus 56 (64) 17 (77) 1.88 0.63-5.66 0.25 Cyclosporine 19 (22) 3 (14) 0.57 0.15-2.17 0.4 Rapamycine 9 (10) 2 (9) 0.88 0.17-4.42 0.87 Prednisone 83 (94) 22 (100) 0.25 COPD 3 (3) 2 (9) 2.8 0.43-18.24 0.26 Active smoking 8 (9) 1 (5) 0.48 0.06-4.08 0.49 Pulmonary neoplasia 0 (0) 2 (9) 0.04* HIV No 87 (100) 22 (100) 1 Rheumatologic/ immunologic diseases 13 (15) 3 (14) 0.91 0.23-3.55 0.89 Table 2 : Additional characteristics (univariate analysis) Controls Cases OR 95% CI p n (%) n (%) Induction / steroids >20 mg/d x 20 days < 6 months 3 (3) 3 (14) 4.42 0.80-24.48 0.10 CMV disease < 6 months No 87 (100) 22 (100) 1 Graft reject < 12 months 3 (3) 6 (27) 10.5 2.14-51.56 0.02* Steroids induction for graft reject (<12 months) 3 (3) 5 (23) 8.24 1.66-40.89 0.008* Rituximab for graft reject (<12 months) 0 (0) 2 (9) 0,04* Absolute lymphocytes values <1000 9(11) 9 (41) 0.001* Glomerular filtration rate <30 ml/min 12 (14) 9 (41) 0.005* Table 3 : Exposition of controls and cases in the outpatient clinic (univariate analysis) Controls Cases OR C.I. 95% p n (%) n (%) Number of visits during study (median) 2 3.5 0.004* Same day, future case 3 months 37 (42) 11 (50) 1.38 0.54-3.54 0.5 Same day, future case 2 months 27 (31) 9 (41) 1.56 0.59-4.13 0.36 +/- 24h, future case 3 months 44 (50) 12 (55) 1.2 0.47-3.08 0.7 +/- 24h, future case 2 months 33 (38) 10 (45) 1.39 0.54-3.59 0.49 Table 4: Independent risk factors for development of PJP (multivariate regression logistic model) OR 95% CI p Glomerular filtration rate <30 mL/min 4.65 1.22-17.78 0.03* Absolute lymphocytes count <1000 5.53 1.40-21.94 0.02* Anti-lymphocyte antibodies at time of presentation 4.73 1.12-19.89 0.03* Table 5: Air detection (H3) of Pneumocystis jirovecii DNA (Real-Time PCR mtLSUrRNA) Ct Negative control #1 (outside of the hospital) - Negative control #2 (outside ofthe hospital) - Possible PJP case (at bedside) + (36.3) Possible PJP case (at door) + (38.9) Nephrology ward room #1 (not PJP) at bedside + (39.7) Nephrology ward room #2 (not PJP) at bedside + (42.7) Outpatient clinic waiting room (AM) - Outpatient clinic waiting room (PM) + (45.9) Collection center (AM) - Collection center (PM) + (46.2) ICU room #1 (not PJP) at bedside - ICU room #2 (not PJP) at bedside - GIII GIII GI GI GI GIV GII GII GII GII GII GII GII GII GII . . . . . . . . . . . . . . . H1 H1 H2 H2 H2 H1 H1 H1 H1 H1 H1 H1 H3 H3 H3 C C A A A Isolated case B B B B B B D D D Genotype UPGMA / concatenated ITS-mt26S-CYB sequences Hospital Outbreak