Objectives: Immunosuppressive therapies have improved survival in solid-organ transplant recipients at the expense of increased prevalence of opportunistic infections. We investigated the prevalence, risk factors, and prognosis of Pneumocystis jirovecii pneumonia in solid-organ transplant recipients who were followed by our transplant unit.
Materials and Methods: We conducted a retrospective observational study using medical record reviews to identify all adult solid-organ transplant recipients who underwent bronchoscopy and bronchoalveolar lavage between January 2011 and 2018. We collected clinical characteristics, including risk factors and prognosis. Pneumocystis jirovecii pneumonia symptoms com-patible with microscopy and/or positive nucleic acid amplification assays were defined as proven infection by P. jirovecii pneumonia.
Results: We identified 312 adult solid-organ transplants (114 renal, 1 kidney and pancreas, 197 liver) in this period. Overall, 113 (36.2%) pulmonary disease consultations were performed in the posttransplant stage, and 46 (40.7%) patients underwent bronchoalveolar lavage with P. jirovecii screening. We identified 18 patients who tested positive for P. jirovecii infection according to nucleic acid amplification assay; 3 were not proven, and 7 had a transplant date before 2011. The prevalence was 8/312 (2.6%); of these 8 patients, 5 had the same genotype cluster. Median P. jirovecii pneumonia development time was longer in renal transplant recipients (P = .016). Only renal transplant recipients were offered Pneumocystis prophylaxis for 6 months. Concomitant viral infection including cytomegalovirus was the only significant factor for P. jirovecii pneumonia development (P = .028). Intensive care admission was 40% (n = 6), and disease-related mortality was 33% (n = 5).
Conclusions: The overall prevalence of P. jirovecii pneumonia in solid-organ transplant recipients was similar to other single-center reports. Prophylaxis prevented early posttransplant P. jirovecii pneumonia. However, P. jirovecii pneumonia may develop at any posttransplant stage, and viral infections other than cytomegalovirus should also be considered as a predictor.
Key words : Bronchoalveolar lavage, Bronchoscopy, Kidney transplant, Liver transplant, Opportunistic infection
Solid-organ transplant (SOT) recipients are prone to infections with common bacterial and viral pathogens as well as opportunistic organisms. The interaction between the patient’s exposures and the intensity of immune suppression determines the infection risk in SOT recipients.1 Despite reduced rejection rates and improved survival with immunosuppressive therapies and new antimicrobial agents, pulmonary infections are one of the most life-threatening forms of invasive infections observed in these patients.2,3
An etiological unclassified pulmonary infection incidence of 13.4% has been reported for liver transplants in our transplant unit,4 which subsequently increased after multidisciplinary teamwork and diagnostic procedures such as bronchoscopy. This yielded more accurate diagnosis of opportunistic infections such as Pneumocystis jirovecii pneumonia (PJP). Diagnosis of PJP depends on the appropriate sample selection such as induced sputum or bronchoalveolar lavage (BAL). The sensitivity of microscopic methods is often too low to diagnose PJP in non-HIV immune-compromised patients, and the development of highly sensitive molecular diagnostic tools such as polymerase chain reaction (PCR) using various gene targets has led to effective diagnosis of PJP in both HIV-positive and HIV-negative patients.5-7 However, although nucleic acid amplification techniques have a high sensitivity, the specificity in discrimination between colonization and infection is limited.8 Therefore, SOT patients having compatible symptoms and positive validated nucleic acid tests can be accepted as proven PJP infection.9
Approximately 5% to 15% of patients who undergo SOT develop PJP in the absence of PJP prophylaxis.10,11 Pneumocystis jirovecii can easily cause outbreaks in SOT recipients who are under regular posttransplant immunosuppressive therapy and treated in a single center. Several clusters or outbreaks of PJP in SOT patients, especially in renal transplant recipients, have been described.12-14 Currently, we experienced a progressive increase in the PJP-diagnosed SOT recipients followed by our transplant unit as well as a PJP cluster. To our knowledge, there are no data on the prevalence of P. jirovecii in SOT recipients for Turkey. Therefore, we intended to investigate the prevalence, risk factors, and prognosis of PJP in SOT recipients of a single transplant unit in a given period.
Materials and Methods
The study was organized as a retrospective cohort. We screened adult (< 18-year-old) SOT recipients who were evaluated in the posttransplant period by a pulmonary disease consultant in our tertiary care university hospital between 2011 and 2018. Those who underwent fiberoptic bronchoscopy (FOB) and BAL were included in the study. The data, including demographic, clinical, laboratory, and radiological characteristics, were analyzed from the electronic medical records of the hospital database after necessary permissions and local Ethical Committee approval were obtained.
Immunosuppressive protocol and antimicrobial prophylaxis
The immunosuppressive protocol included corticosteroids, mycophenolate mofetil, and cyclosporine or tacrolimus. In liver transplants, ceftriaxone was administrated after induction, 30 minutes before the surgical incision for intraoperative prophylaxis, and was continued for 1 to 3 days after transplant. In renal transplants, cefotaxime was given for 3 days. Antiviral and antifungal prophylaxes were not routinely used; however, kidney transplant recipients and kidney-pancreas recipients were provided with trimethoprim-sulfamethoxazole prophylaxis for 6 months posttransplant.
Bronchoalveolar lavage fluid collection and processing
The process of sampling by inserting the tip of the bronchoscope into the appropriate segment is a semi-invasive diagnostic method for detecting infection. Bronchoalveolar lavage was performed with 20-mL portions of sterile normal saline at room temperature. A total of up to 300 mL was instilled, and more than 60% of the fluid was aspirated. Specimens were sent to microbiology and mycobacteria laboratories for isolate identification. Bronchoalveolar lavage fluid specimens were aliquoted in 2 parts for P. jirovecii detection, ie, 1 for microscopic examination and 1 for PCR processing.
Microscopic detection of Pneumocystis jirovecii
The first BAL aliquot was centrifuged at 1500 rpm for 10 minutes. Cytospin-concentrated smears were prepared from the pellets and stained with Giemsa and methenamine silver for detection of the P. jirovecii cysts and trophozoites.
DNA extraction, polymerase chain reaction detection, and sequencing
DNA extraction was performed in accordance with the manufacturer’s guidelines (Macherey-Nagel). Extracted DNA samples were stored at -20 °C until the amplification procedure. Ultrapure distilled water was used as a negative control, and P. jirovecii isolate that was previously obtained from a patient diagnosed with PJP infection was used as positive control.15
Nested PCR was performed with the P. jirovecii mtLSU-rRNA gene region as the target. At the first PCR round, we used primers pAZ102-E (5′-GATGGCTGTTTCCAAGCCCA-3′) and pAZ102-H (5′-GTGTACGTTGCAAAGTACTC-3′). In the second PCR round, we used pAZ102-X (5′-GTGAAAT ACAAATCGGACTAGG-3′) and pAZ102-Y (5′-TCACTTAATATTAATTGGGGAGC-3′). DNA amplification was performed using protocols described in the literature.16 The detection of an amplicon 267 bp in length showed that a specimen was positive for P. jirovecii.16 Polymerase chain reaction products of the mitochondrial large subunit ribosomal RNA (mtLSUrRNA) loci were sequenced by the Microsynth company. The sequencing was performed using the Sanger dideoxy sequencing method with an ABI 3730XL automatic sequencing apparatus and the Big Dye Terminator cycle sequencing reaction kit (Applied Biosystems) in both forward and reverse directions. Nucleotide sequences were aligned and edited using the DNADynamo 2004-2019 program (Blue Tractor software). Genotypes of P. jirovecii were distinguished by identifying the polymorphisms at positions 85 and 248 of the mtLSUrRNA and were numbered using the method described by de Armas and colleagues.17
Definition of proven Pneumocystis jirovecii pneumonia infection
Pneumocystis jirovecii diagnosis was based on consensus guidelines, which require positive identification by direct microscopy with immunofluorescence on induced sputum or BAL and/or a positive PCR assay on a BAL specimen.8 A definitive or proven PJP infection is defined by demonstration of organisms in induced sputum or BAL by direct microscopy and/or PCR assays in SOT recipients with compatible syndromes. Symptoms compatible with PJP were defined as fever, dyspnea, and cough.9 Detection of P. jirovecii by PCR without specific symptoms and negative microscopy is regarded as colonization.18
Data were analyzed with SPSS software (version 22). We used mean and standard deviation or median and minimum-maximum values according to the distribution of the data. For parametric evaluations, we used a t test; for nonparametric values, we used the Mann-Whitney U test. Categorical data were evaluated by chi-square or the Fisher exact test. P < 0.05 was considered significant.
Since 1992, there have been 1042 SOTs (mainly renal and liver) performed in the transplant unit of tertiary care at our university hospital (Dokuz Eylul University Faculty of Medicine). These patients, as well as some others who had undergone SOT in another transplant center, have been followed by our transplant team from that time. After 2011, a multidisciplinary organization was established that facilitated a scheduled evaluation by specialized consultants including pulmonary specialists; also, since then, P. jirovecii detection has been performed with molecular diagnostic tests.
Between January 2011 and January 2018, there were 312 adult patients who underwent SOT (115 renal transplants, 197 liver transplants), and 113 patients (36.2%), including the ones with earlier transplant dates (the earliest in 1992), had at least 1 pulmonary disease consultation for pulmonary infections in the posttransplant stage.
Of these 113 patients, 46 patients (40.7%), which included 7 patients who had their transplants in a different transplant unit, underwent FOB and BAL with P. jirovecii screening in our clinic. The mean age of these patients was 50.6 ± 10.9 years. Seventeen transplant recipients (37%) received organs from deceased donors, 28 recipients (61%) received organs from living donors, and 1 recipient (2%) received organs from both living and deceased donors (this was a kidney-pancreas transplant). Most of the living donors were first degree relatives (n = 18, 62%), followed by 6 spouses (21%), 3 second degree relatives (10%), and 2 third degree relatives (7%). Twenty-five patients (54%) had kidney transplants, 20 patients (44%) had liver transplants, and 1 patient (2%) had a dual pancreas and kidney transplant (Figure 1).
Pneumocystis jirovecii was detected by PCR in 18 of 46 patients (39.1%) for whom FOB and BAL were performed. Microscopy for PJP was also positive in 3 patients. Incidence of PJP infection was more prevalent in renal transplant recipients (n = 10; 63%) than in liver transplant recipients. Of these 18 PCR-positive patients, 3 patients (2 liver transplant and 1 renal transplant) did not have clinical symptoms and/or radiological findings indicative for PJP; therefore, P. jirovecii positivity for these patients was accepted as proof of colonization. The median posttransplant PJP development time was 365 days (minimum 9 days, maximum 7480 days) for PJP-proven patients. Seven patients had a transplant date before 2011. For the remaining 8 patients, the prevalence was 8/312 (2.6%) at a median of 174 days posttransplant (minimum, 9 days; maximum, 510 days). For renal transplant recipients, PJP prevalence was 3/115 (2.6%), and for liver transplant recipients it was 5/197 (2.5%). All PJP-positive patients with PCR are shown in Table 1.
Among PJP-proven patients, 5 were diagnosed consecutively in 6 months and shown to have the same genotype; 2 of these patients had liver transplants, 2 had renal transplants, and 1 had both a renal transplant and a pancreas transplant (Table 1).
When the BAL samples for SOT recipients (n = 46) were compared, there was no statistical difference between those samples that were positive versus negative for P. jirovecii with regard to demographic factors such as age, sex, donation type, and transplanted organ, as well as prognosis. Concomitant viral infections (7 cytomegalovirus, 1 respiratory syncytial virus, 1 adenovirus, and 1 bocavirus infection) were the only risk factors found to be associated with PJP in our cohort. Fever, dyspnea, and cough were present in 10 of the PJP-proven patients (67%). The most commonly observed radiological finding (n = 10; 67%) was diffuse interstitial pattern. Focal infiltration, pleural effusion, lymphadenopathy, and cavity were other findings that were observed, but rarely. Admission to intensive care unit (6 patients; 40%) and mortality (5 patients; 33%) were fairly high in PJP-proven patients (Table 2).
When renal and liver transplant recipients were compared, demographic characteristics such as age, sex, and donation type were similar. Intensive care unit admission and mortality rates were higher and PJP treatment duration was shorter in liver transplant recipients, but these differences were not statistically significant. Median posttransplant PJP development time was significantly (> 6 months) longer in renal transplant recipients (P = .016) (Table 3).
Among SOT recipients who underwent FOB and BAL, for 23 patients (50%) trimethoprim-sulfamethoxazole treatment was started upon clinical suspicion and before laboratory samples were obtained. On the other hand, among 15 PJP-proven patients, 5 did not have PJP treatment, 2 of them died rapidly after clinical presentation, and 1 was discharged in response to individual factors; the remaining 2 patients survived with other antimicrobial agents. Unfortunately, the samples of untreated patients had been studied fairly late after the clinical presentation. Ten patients had trimethoprim-sulfamethoxazole treatment for a median of 14 days (minimum 2 days, maximum 26 days).
The risk of opportunistic pulmonary infections such as PJP is increased in SOT recipients as a result of the nature of the underlying disease and/or immune-modulatory therapies. Soon after we determined a small PJP cluster in our transplant unit, we decided to investigate the prevalence, characteristics, and prognosis of PJP in SOT recipients who were followed by our transplant unit for a given period. To our knowledge, PJP prevalence in SOT recipients has not been previously studied in Turkey. We found that the overall PJP prevalence was 2.6% in this single-center study; for renal and liver transplant recipients, the prevalence was 2.6% and 2.5%, respectively, similar to the results from previous studies. There are recommendations for the prophylaxis and treatment of PJP in SOT recipients; however, routine practice may change in different centers.19 In our transplant unit, only renal transplant recipients were offered PJP prophylaxis; this was not offered to liver transplant recipients because of the risk of hepatotoxicity. Viral infections were the only predictors of PJP, and the prognosis was fairly poor, with 5 deaths (33%) out of 15 patients with PJP-proven infections in which 3 of these patients did not receive prophylaxis in the era of prophylaxis.
Between 2011 and 2018, one-third (36%) of the SOT recipients followed by our transplant unit were evaluated by a pulmonary consultant, and 41% were offered FOB and BAL. Fiberoptic bronchoscopy with BAL or transbronchial lung biopsy has been reported to provide a diagnostic challenge, and early diagnostic bronchoscopy is recommended in acute, subacute, or chronic pulmonary processes, especially when an infectious etiology is suspected.20,21 In their report of 998 patients, Eyuboglu and colleagues reported that, of 90 patients (9%) who had FOB, causative organisms were identified in > 30% of patients. However, BAL was obtained in only 10 patients.22 In our study, we found that FOB and BAL were frequently used procedures in our transplant center for the diagnostic algorithm of pulmonary problems as recommended.
Pneumocystis jirovecii is a ubiquitous fungus with trophic, cystic, and sporoid forms.23 Pneumocystis jirovecii pneumonia develops in 3 ways: interindividual airborne transmission from a PJP patient or asymptomatic carriers, reactivation of previously untreated infection, and environmental exposure.11 Studies have suggested that asymptomatic carriers and interindividual transmissions may have contributed to the spread of infection in isolated PJP or outbreaks among renal, liver, and heart transplant recipients.24-26 In our study, we determined a period (from April to August 2017) during which we diagnosed PJP patients consecutively, and genotype sequencing demonstrated that all of these patients had the G2 genotype, which meant there was a cluster due to an asymptomatic reservoir or an environmental exposure in the transplant unit. It has been reported that there is a tendency of PJP development in warm seasons like we observed, suggesting an environmental exposure.27 The sequencing could be performed soon after the determination of infections; therefore, although we took an action about a cluster, we could not have proved it simultaneously.
Pneumocystis jirovecii pneumonia is most frequent seen in renal transplant recipients, as it is the most common transplant type. However, heart, lung, and combined heart-lung transplant recipients have the greatest risk regardless of prior prophylaxis.12,28 Pneumocystis jirovecii pneumonia infection prevalence was reported to be 5% to 15% in SOT recipients before the era for which PJP prophylaxis became standard.11 In the era of prophylaxis, the prevalence of PJP has decreased to 0.4% to 3.7%. In our study, we found the overall PJP prevalence of 2.6% in a group of patients to whom prophylaxis was given and not given. Although previous studies have demonstrated that renal transplant recipients who receive prophylaxis have a PJP prevalence of 0.4% to 2.2%,6,14 we found a PJP prevalence rate of 2.6%, which was a bit higher than those reports. On the other hand, among liver transplant recipients who did not have prophylaxis, a PJP infection rate of 3% to 11% has been reported,12,29 and, for this group, our PJP prevalence rate was lower (2.5%). In our cohort, most of the PJP-proven patients had renal transplants (n = 10), although the number of liver transplant procedures was nearly twice the number of renal transplants. This result might be due to higher number of renal transplant recipients who underwent FOB procedure.
Before the routine use of prophylaxis, the risk of PJP has been reported to be highest between the second and sixth month after transplant.30,31 In the era of posttransplant prophylaxis, infection is observed after the first year, especially after the continuation of trimethoprim-sulfamethoxazole prophylaxis.11 In our study, PJP was diagnosed after posttransplant month 6 in renal transplant recipients who were provided PJP prophylaxis within the first year. However, in liver transplant recipients, PJP development was earlier during the posttransplant period as a result of the lack of prophylaxis. Among PJP-proven patients the only difference was PJP development time between renal end-stage liver transplant recipients; this difference was attributed to previous prophylaxis given to renal transplant recipients (P = .016) in our cohort (Table 3).
Risk factors for PJP infection have been researched in many studies. The main risk factors in SOT recipients have been defined as specific immunosuppressive regimens. Maintenance immunosuppression regimens containing the combination of tacrolimus plus sirolimus, cyclosporine plus mycophenolate mofetil, and sirolimus plus mycophenolate mofetil were found to be significant risk factors for contracting a Pneumocystis infection.29 In our study, all PJP patients were receiving mycophenolate mofetil, and 3 of them were being treated with cyclosporine plus mycophenolate mofetil. Age, underlying pulmonary diseases, and transplant dysfunction are other risk factors that have been investigated, and cytomegalovirus infection was identified as a clear risk factor for PJP in SOT patients.8,24,32,33 In our study, none of the PJP-proven patients was older than 65 years and none had underlying pulmonary diseases or other previously defined risk factors, although the number of viral infections was significantly greater in PJP-proven patients (P = .028). These findings demonstrate that cytomegalovirus, as well as other viruses, might be a risk for SOT recipients. Pneumocystis jirovecii pneumonia should be considered in the differential diagnosis of patients with constitutional symptoms and viral infections.
The clinical course of PJP is more acute.11 Common pneumonia symptoms such as fever, dyspnea, and a dry cough often develop over a few days and may progress to acute respiratory failure more frequently. In our study, we found all these 3 symptoms together in 10 of the PJP patients (63%). The most commonly reported radiological finding related to PJP has been diffuse interstitial pattern,11,30 and 9 of our PJP-proven patients (56%) had a diffuse interstitial pattern in radiological evaluations.
The diagnosis of PJP is still an important concern. In vitro Pneumocystis culture is not possible, even with recent technological advancements. A semi-invasive diagnostic test is an effective method; testing BAL fluid by PCR excludes PJP, if negative.34 Lower levels of PJP, detected by PCR, may reflect colonization, which in turn is subject to further validation by evidence of compatible symptoms. All of our patients with PJP-compatible symptoms were diagnosed by nucleic acid test. Three patients with PCR positivity were regarded to be associated with colonization as diagnosed by the absence of compatible symptoms, and 2 patients were also diagnosed as positive by microscopy. These results show the importance of nucleic acid testing in the diagnosis of PJP in patients with clinically compatible symptoms.
In renal transplant recipients with PJP, the number of patients who were admitted into intensive care units and in need of mechanical ventilation was reported to be between 8% and 71%, and the mortality rate was about 14%.24,35 In liver transplant recipients, a mortality rate of 7% to 50% was observed.26 Pneumocystis jirovecii pneumonia-related mortality was observed to be lower for SOT recipients than for patients with other types of immunosuppression (odds ratio, 0.08; 95% CI, 0.02-0.31).28 In our study, the intensive care unit admission rate was 38%, and the mortality rate was 31%, which is within the range of the recent literature.
Limitations of the study
The most important limitation of this study is that the BAL specimens were studied with nucleic acid testing together and only when an adequate number of samples were collected. This resulted in delayed serological diagnosis, and therefore one-third of the PJP-proven patients did not receive PJP treatment, whereas one-half of the patients who underwent FOB received PJP treatment in response to compatible symptoms without definitive serological evidence. Another limitation is that we did not have the results of CD4 lymphocytes, which may be an important risk factor especially in the course of viral infections. We also were not able to describe the source of the cluster, and FOB was not performed in all patients compatible with respiratory symptoms; however, FOB procedure rates at our center were high.
Pneumocystis jirovecii pneumonia is an opportunistic infection that should be considered with respect to the concomitant increase in immunosuppressive conditions that have accompanied the increase in the number of SOT procedures. Strategies identifying SOT recipients who have a higher risk for PJP with regard to previously defined risk factors such as viral infections may help to identity patients who would benefit from PJP prophylaxis beyond the first year posttransplant. Pneumocystis jirovecii pneumonia may develop at any stage of the posttransplant period. Solid-organ transplant recipients with a clinical syndrome compatible with PJP may benefit from the information provided by a nucleic acid test.
DOI : 10.6002/ect.2020.0080
Table 1. Characteristics of Patients Who Tested Positive for Pneumocystis jirovecii Pneumonia, as Tested With Polymerase Chain Reaction
Table 2. Features of Solid-Organ Transplant Recipients Who Underwent Fiberoptic Bronchoscopy for Opportunistic Pulmonary Infection According to P. jirovecii Positivity by Polymerase Chain Reaction
Table 3. Characteristics of Patients According to Transplanted Organ
Figure 1. Flow Chart of the Study Inclusion Protocol