August 16, 2007

Pulmonary Infections in Transplantation Pathology

By Stewart, Susan

Context.-Pulmonary infections are common and often life- threatening in solid organ and stem cell transplant recipients. Understanding their pathology is critical to making improvements in care and survival as well as in surgical techniques, immunosuppression management, prophylaxis, and treatment. Pulmonary infections are particularly common and serious in the susceptible population of lung transplant recipients. Objective.-To summarize recent updates in the field for opportunistic infections and some common pathogens, and to consider the role of the diagnostic pulmonary histopathologist as well as advances in molecular diagnosis.

Data Sources.-This work is based on a selected review of the relevant medical and scientific literature, with emphasis on lung transplantation experience gained during 2 decades of practice.

Conclusions.-Pulmonary infections in transplant recipients present a diagnostic challenge and are a continuing source of mortality and morbidity despite improvement in prophylaxis and treatment. Accurate diagnosis requires multidisciplinary input from clinicians, radiologists, and pathology disciplines as well as complementary molecular methods.

(Arch Pathol Lab Med. 2007;131:1219-1231)

Lung transplant recipients have the highest incidence of infections and present the greatest diagnostic challenges among organ transplantation patients. Histology is pivotal in early and accurate diagnosis and exclusion of differential conditions. It is also important as a gold standard for evaluation of new molecular techniques.

The incidence of infectious complications after organ transplantation has decreased with the introduction of prophylactic strategies and improvement in immunosuppressive treatment.1 The lungs represent particularly vulnerable organs and remain the leading site of infections in both lung and heart transplant recipients and the second most common site in liver transplant recipients.2-5 Kidney transplants have the lowest incidence of pulmonary infection, which is partly because of the lower level of immunosuppression required by these patients. The range of microorganisms responsible for posttransplantation pulmonary infections is similar among the various solid organ recipients, and there is a characteristic timeline of these infectious complications.6 The first postoperative month is influenced by the infectious risks associated with surgery and intensive care and also with the initiation of immunosuppressive agents. Pneumonia in this immediate postoperative period and throughout the first month is most commonly caused by bacteria, particularly nosocomial gram- negative and staphylococcal species. Recipients are exposed to these pathogens during their postoperative recovery, but in the case of lung transplant recipients there is the unique route of infection through donor organ transmission.6

Donor lungs are inevitably retrieved from patients who have sustained brain death and, through ventilation, have bypassed normal airway defenses. The donors have often been exposed to nosocomial pathogens during their hospitalization. 7 Lung transplant recipients with chronic sepsis such as bronchiectasis, including cystic fibrosis, are also at risk from bacterial colonization of their sinuses and large airways, which can recrudesce after surgery.8 In single-lung recipients there is also the hazard of a potentially infectious sump remaining in the form of the native lung. In addition, the transplanted lung is denervated and no longer retains a normal cough reflex. Mucociliary clearance is impaired and lymphatics are disrupted, and these necessary consequences of harvesting and implantation are compounded by airway ischemia in which mucosa can slough and accumulate with secretions. Bacterial pneumonia is therefore one of the most common infections in the perioperative period in lung transplant recipients and contributes significantly to early morbidity and mortality.8 The incidence of pneumonia in the first 2 weeks after surgery in lung transplantation has been reported to be as high as 35%, but has fortunately improved with the introduction of routine antibiotic prophylaxis.6

The second stage of infectious risk following transplantation extends from 1 to 6 months, which is the period of maximum sustained immunosuppression in order to minimize acute allograft rejection.6 The patients are often discharged into their normal environment and the risk of bacterial pneumonia subsides. The effects of induction therapy may still be realized beyond the first month following transplantation, and this second stage is characterized by the emergence of opportunistic pathogens. The most problematical of these include cytomegalovirus (CMV) and Aspergillus, which will be dealt with in detail later.

Beyond 6 months, a third stage in the timeline of infectious complications is reached. The level of immunosuppression is often reduced in recipients in this period and infections are less commonly due to opportunistic organisms, although these may be seen following periods of augmented immunosuppression for the treatment of acute or chronic rejection. The infections in the third stage are largely the result of common community-acquired pathogens and again, the lung transplant population appears particularly prone to these infections involving the lower as well as the upper respiratory tract and resulting in longer-term sequelae such as loss of lung function.4,6

The risk of pulmonary infection in transplant recipients includes the immunosuppressed status, which should also take into account any induction therapy. The latter can include antilymphocyte antibodies that rapidly deplete circulating lymphocytes and interleukin 2 receptor antibodies that competitively bind to CD25 on T-lymphocyte surfaces. Both these regimes affect lymphocyte function and the overall immune response against pathogens requiring cell-mediated immunity.1 If antilymphocyte antibodies are used outside the induction period for the treatment of recurrent or refractory acute rejection or rapidly progressive chronic rejection, the susceptibility of opportunistic infection continues beyond the characteristic period. 6 Surgical technique is a risk of infection, which is obviously greatest in intrathoracic transplantation, but extensive surgical manipulation of the upper abdomen also contributes to risk (eg, in liver transplant recipients). In the case of stem cell transplantation, chemoradiation conditioning greatly increases the risk of infection, and some aspects of this will be considered in the sections pertaining to specific opportunistic organisms.1

The use of prophylaxis has been effective in decreasing the incidence of both common and opportunistic infections following transplantation, but does not reduce the need for vigilance in identifying infections in this rather modified setting. Prophylaxis may have the effect of timeshifting the infection to a later period (eg, CMV infection when prophylaxis is tailed off) and it must also be remembered that compliance issues are also important in prophylactic regimens.9,10 The infection risk following transplantation varies geographically, and with the proliferation of transplantation centers, including those in lessdeveloped countries, the incidence of infections such as tuberculosis can be up to 15% of recipients.11 The proliferation of transplant centers also increases the need for knowledge and expertise in the diagnosis and treatment of pulmonary infections in transplantation, and pathologists are key to the early recognition and diagnosis of these conditions. The emphasis, as always, is on multidisciplinary team discussion so that clinical, radiologic, histopathologic, microbiological, and serologic evidence can be assimilated into a working diagnosis.


Invasive filamentous fungal infection is an important cause of mortality in allogeneic hematopoietic stem cell transplant recipients.12 Although prolonged and persistent neutropenia is a critical factor for invasive aspergillosis, most cases of invasive filamentous fungal infection occur after neutrophil recovery in hematopoietic stem cell transplant recipients. This is in the setting of potent immunosuppressive therapy for graft-versus-host disease. Both animal models of invasive aspergillosis and clinical experience suggest that Aspergillus infection occurring during neutropenia is pathologically and immunologically distinct from infection in the absence of neutropenia.12 Shaukat et al12 analyzed 22 consecutive cases of invasive filamentous fungal infection in allogeneic hematopoietic stem cell transplant recipients and all but one case were diagnosed following resolution of their neutropenia. They found the predominant histopathologic finding was coagulative necrosis, which is usually not associated with invasive filamentous fungal infection in nonneutropenic hematopoietic stem cell transplant recipients. In contrast, in most solid organ transplant recipients, pulmonary lesions of invasive Aspergillus infection consist of neutrophilic and monocytic infiltrates with inflammatory necrosis and scant intra-alveolar hemorrhage13 (Figures 1 and 2). This underlines the fact that patients with chemotherapy-induced neutropenia and solid organ transplant recipients treated with cyclosporine and corticosteroids are 2 immunologically distinct populations. Shaukat et al12 speculated that the predominance of coagulative necrosis in their series could reflect high doses of corticosteroids used to treat graft-versus-host disease that may have disabled leukocyte trafficking and hyphal killing. The important message for histopathologists is to be aware of the clinical details of the stem cell transplant recipient in whom Aspergillus infection is suspected as the appearances can vary depending on whether the patient is in the postengraftment rather than neutropenic period. Within the invasive filamentous fungi, Aspergillus species usually predominate both in pulmonary and more widely systemic infections, but other fungi, including Rhizopus, Acremonium, Mucor, and Penicillium species, are encountered from time to time.12-14 However, there is little information on the histopathologic features of non-Aspergillus mold infections in the literature.

Another important take-home message in the histopathologic diagnosis of lung infections in transplantation is the need for intense scrutiny of areas of coagulative necrosis.14 Correlation of histopathology with molecular data is now possible, and Lass-Florl et al15 have shown that positive results became negative shortly after the commencement of antifungal therapy, but did not correlate with the clinical response to treatment. Polymerase chain reaction (PCR) methods detect circulating fungal DNA of several Aspergillus species, including A fumigatus, A flavus, A tereus, and A niger, with genus-specific oligonucleotide probes.16 It is important to define invasive Aspergillus infection accurately in order to assess the specificity and sensitivity of the PCR methods. The European Organization for Research and Treatment of Cancer/Mycoses Study Group criteria define proven, probable, and possible infections with tight host factor, microbiological, and clinical criteria.17,18 Polymerase chain reaction can be performed on whole blood and also on tissue. The half-life of fungal DNA in blood is about 5 minutes, and PCR can be positive with a large fungal burden.

It is important to recognize that there is a high rate of transient Aspergillus fungemia in patients without evidence of invasive aspergillosis, emphasizing the need for molecular diagnostic methods to be interpreted in full clinical context. The sensitivity in lung and blood for proven infections according to the European Organization for Research and Treatment of Cancer/Mycoses Study Group criteria are 100% and 40%, respectively.15 Sensitivity for probable infection is between 66% and 44%. The negative predictive value for a patient with invasive Aspergillus infection on treatment is 44%. The benefit of PCR is therefore limited when treatment has already been started. Antimycotic treatment may be partly responsible for clearance of fungi from the blood to nondetectable levels, whereas clearance from tissue does not occur. A positive signal is observed only when the fungal burden is large enough,16 whereas screening for and diagnosis of aspergillosis prior to treatment gives better sensitivity. Polymerase chain reaction is not yet developed as a means of monitoring responding and nonresponding patients.19,20 The way forward is likely to combine microscopy, culture, and PCR to improve the diagnostic outcome. Further limitations in the use of PCR in clinical applications include contamination and the distinction between colonization and invasive disease.16


The lung transplant population is unique in its predisposition for, and clinical manifestations of, Aspergillus infection. 21-24 These infections are documented in 6% to 8% of lung transplant recipients and are therefore among the most significant opportunistic infectious organisms following lung transplantation. The transplanted lung is in direct communication with the environment and also suffers from impairment of local host defenses in terms of mucociliary clearance and the cough reflex, together with disruption of lymphatic drainage. The airways may suffer ischemic injury because of implantation, and the additional requirement of immunosuppression in this population, compared with other solid organ transplants, to control allograft rejection further increases susceptibility to Aspergillus disease. This can range from simple Aspergillus airway colonization to invasive aspergillosis, as described previously. The isolated tracheobronchitis and bronchial anastomotic infections due to Aspergillus are entities entirely distinct from Aspergillus pneumonia. Endobronchial aspergillosis is uniquely encountered in the lung transplant population, with an observed frequency of around 5%.25

In most cases of endobronchial aspergillosis, the infection is localized to the bronchial anastomosis where a combination of devitalized cartilage and suture material can create a nurturing environment for fungus.1,25 Infection can less commonly involve a more diffuse ulcerative bronchitis or tracheobronchitis with the formation of pseudomembranes, which again typically occur following ischemic injury to the mucosa. These infections are usually seen within the first 6 months posttransplantation, but can be asymptomatic in the majority of patients. The detection of endobronchial aspergillosis therefore requires surveillance bronchoscopy (Figures 3 and 4). It may also be detected at clinically directed bronchoscopies for other infectious and immune complications following lung transplantation. There is a small but worrying risk of progression from endobronchial disease to invasive pneumonia or erosion of adjacent structures, including the pulmonary artery.25-27 In association with underlying ischemic injury to the airways, endobronchial aspergillosis is associated with an increased risk of subsequent bronchial stenosis or bronchomalacia, which may require stenting. There are clinical differences in Aspergillus infections in the type of transplant performed and therefore implications for management. A review of published reports for invasive aspergillosis by Singh and Hussain22 revealed a mean incidence of Aspergillus infections in lung transplant recipients of 6.2%. Fifty-eight percent of the Aspergillus infections were tracheobronchial or bronchial anastomotic infections, 32% were invasive pulmonary infections, and 22% were disseminated infections. They identified that single-lung transplant recipients with Aspergillus infections were slightly older than other recipients and more likely to have chronic obstructive pulmonary disease as the indication for transplantation. Single-lung recipients were more likely to develop Aspergillus infections later following transplantation and also had a higher incidence of invasive aspergillosis than other lung transplant recipients. The overall mortality in lung transplant recipients with Aspergillus infections in this study was 52%. In this review, A fumigatus was the most common species cultured followed by A niger, A versicolor, A flavus, A nidulans, A glaucus, and A tereus. In the majority of single-lung transplant recipients, invasive aspergillosis was documented in the native lung, suggesting that the remaining lung may harbor a nidus of Aspergillus and serve as an ongoing source of infection in these patients.

The high incidence of chronic obstructive pulmonary disease as the underlying disease necessitating transplantation is known to predispose to airway colonization with Aspergillus. In a molecular epidemiologic study in lung transplant recipients that used DNA primers for strain typing, the clinical strain of a single-lung transplant recipient was identical to the one collected at home, whereas the isolates in other transplant recipients were thought to be more likely nosocomial in origin.28 Bilateral lung and right lung transplant recipients were documented to have a high incidence of bronchial and anastomotic infection. Bronchopleural fistulae were documented in 4.4% of lung transplant recipients with tracheobronchial Aspergillus infections. Critically, fever was documented in only 15% of patients with Aspergillus infections, and patients with invasive pulmonary or extrapulmonary aspergillosis were more likely to be febrile than those with tracheobronchitis or bronchial anastomotic infections. Aspergillus lung infections in lung transplant recipients frequently lacked a characteristic radiologic appearance and presented most often with focal areas of patchy consolidation or infiltration. Nodular lesions were documented in 30% of the patients with pulmonary aspergillosis in the study by Singh and Hussain22 and 27% in the report by Shreeniwas et al.29

A further retrospective study of more than 250 lung transplant patients was reported by Sole et al30 to determine the prevalence, clinical presentation, and mortality of Aspergillus infection in order to devise specific risk factors and to compare survival in patients with and without infection. Aspergillus was isolated from 33% of cases that involved colonization, tracheobronchial lesions, and invasive aspergillosis. There was a significant impact of Aspergillus infection on survival, with 5-year mortality rates higher in single-lung transplant recipients with bronchial and anastomotic infection and in those with later-onset infections and chronic rejection. A significant association was also found between acute rejection and the time at which fungal infection was diagnosed. Typically, Aspergillus was detected in respiratory samples 1.8 +- 4 months preceding an acute rejection episode. There was a significant association between the time that Aspergillus species were first isolated and the time of diagnosis of acute rejection, the administration of tacrolimus, and the administration of mycophenolate. However, there was no significant association with CMV infection or corticosteroid doses. The incidence of acute rejection was similar between the 3 different forms of Aspergillus infection, that is, colonization, airway infections, and invasive infections. Bronchiolitis obliterans syndrome was diagnosed in 87% of cases of invasive Aspergillus infection compared with 44% of colonized cases and 57% of those with isolated airway infections. The mortality of Aspergillus infection was 14% in cases of tracheobronchitis, 28% in colonized cases, and 78% in cases with invasive infections.30 This study reiterates the need for close monitoring, consideration of preemptive antifungal therapy, and meticulous diagnosis of Aspergillus disease in this vulnerable group of patients. In terms of cystic fibrosis patients colonized previously with Aspergillus, this study showed a significant increase of bronchial and anastomotic infections. Helmi et al23 also reported on Aspergillus infection in lung transplant recipients with cystic fibrosis, and concluded that the increased risk of tracheobronchial aspergillosis warrants early surveillance bronchoscopy for its detection, particularly in recipients with pretransplant colonization. One preoperatively colonized cystic fibrosis recipient developed dehiscence of the involved anastomosis, but none of the cystic fibrosis recipients developed disseminated aspergillosis or pneumonia. Overall, 53% of cystic fibrosis recipients had A fumigatus isolated from their respiratory secretions prior to undergoing transplantation and 59% of these recipients had A fumigatus persistently present in respiratory secretions posttransplant.

In summary, many factors may predispose to a fungal infection in lung transplant recipients, including preoperative chronic lung diseases, immunosuppression, intraoperative complications such as bronchial anastomotic problems or lung injury, and postoperative complications such as augmented immunosuppression for rejection, graft dysfunction, viral and bacterial infections, and bronchiolitis obliterans syndrome. Kubak21 reported that the risk factors and time course for fungal infection in lung transplant recipients parallel the observations in other solid organ transplant patients. Early fungal infections are related to surgical complications and the period of 1 to 6 months posttransplant reflects opportunistic, relapsed, or residual infection. Fungal infections occurring 6 months or later are usually associated with treatment for chronic rejection or persistent bronchial airway mechanical abnormalities. Most fungal infections in lung transplant recipients involve Aspergillus species followed by Candida, Pneumocystis, Cryptococcus, geographically restricted endemic fungi, and some newly emerging fungal pathogens.13,21

Other forms of Aspergillus infection are described in lung transplantation and include obstructing tracheobronchial aspergillosis, which is noninvasive and can be treated by bronchial toilet (Figure 5). Bronchocentric granulomatous mycosis is also described with characteristic histologic features, obviously requiring a tissue sample for a firm diagnosis (Figure 6). Likewise, allergic bronchopulmonary aspergillosis together with eosinophilic bronchitis and bronchiolitis due to Aspergillus infection can be seen (Figure 7) and requires biopsy material for confirmation of diagnosis.31 In the setting of immunosuppression, particularly if it is augmented for the treatment of acute rejection, these forms of Aspergillus infection may progress to invasive disease and the diagnostic threshold should always be low. As there is evidence of reactivation rather than reinfection in some cases of Aspergillus infection in single-lung recipients, it is important to examine the explanted lung thoroughly to exclude clinically unsuspected Aspergillus colonization or infection as the remaining in situ native lung in a single-lung recipient may represent a continuing source of infection.26,32,33


Cytomegalovirus is the most common viral pathogen encountered in all solid organ recipient populations. Infection can occur by transmission of the virus within the allograft or by reactivation of latent virus in the recipient. The greatest risk of developing infection is in seronegative recipients acquiring seropositive donor organs. The severity of infection is also enhanced by the use of antilymphocyte antibody therapy for induction and immunosuppression. Cytomegalovirus infection usually occurs 1 to 3 months after transplantation, although the clinical onset is often delayed in patients receiving prophylaxis. In the lung transplant population, this time shift of CMV infection prevents this opportunistic infection from occurring at a time of many other postoperative complications including acute rejection. Cytomegalovirus infection can be subclinical or manifest as clinical disease with fever, malaise, leukopenia, and organ-specific involvement of lungs, liver, myocardium, gastrointestinal tract, and central nervous system. Cytomegalovirus infection also appears to act as a host immunosuppressive with other opportunistic infections emerging in its wake. Tumor necrosis factor released during acute rejection episodes can trigger viral replication, leading to reactivation in solid organ graft recipients. Antiviral prophylactic strategies are well developed, and the incidence of CMV pneumonitis ranges from 0% to 9.2% among liver transplant recipients, 0.8% to 6.6% among heart transplant recipients, and less than 1% in renal transplant recipients.1 After lung transplantation, the incidence is reported as between 15% and 55%, consistent with the lung being a major site of CMV latency with significant viral load transmitted in the donated organ.34

Subclinical disease is detected in lung transplant patients with the use of surveillance bronchoscopy in this population.35,36 Up to 15% of cases of CMV pneumonitis are asymptomatic in lung transplant recipients. The viral prodrome of nonproductive cough and dyspnea and associated laboratory findings of leukopenia and thrombocytopenia can be rather nonspecific. The radiologic findings are similarly nonspecific with ground-glass opacities, air space consolidation, and nodules.37,38 A firm histologic diagnosis of CMV pneumonitis can be established with the presence of characteristic viral inclusions in tissue specimens associated with an inflammatory response (Figure 8).31 The yield of transbronchial biopsies can be low and affected by prior use of prophylaxis, which modifies the appearance of the viral inclusions (Figures 9 and 10). Bronchoalveolar lavage can show viral inclusions, but without evidence of the tissue reaction that is usually seen in biopsy material (Figure 11). The routine use of immunohistochemistry to demonstrate CMV in lung transplant biopsies has been advocated and may be helpful where the inclusions have been modified by prior prophylaxis.39 The rapid shell viral culture of bronchoalveolar lavage (BAL) fluid is efficient in detecting virus, but a positive result must be interpreted in the appropriate clinical setting as viral shedding may occur without pulmonary disease.

Precise quantification of viral load in peripheral blood and lavage fluid can be achieved by the pp65 antigenemia assay and PCR techniques. However, data on the role of these assays as surrogate diagnostic markers of invasive disease are limited.34,40 Sanchez et al34 showed that the blood viral load measured by PCR was significantly higher in lung transplant recipients showing biopsy- proven pneumonitis compared with those without histologic evidence. The positive predictive value of this technique in diagnosing CMV pneumonitis was only 40%, although the negative predictive value was 89% using a threshold optimized by receiver-operating curve analysis. Studies have also shown that a high or rapidly increasing peripheral blood viral load can be a sensitive but nonspecific marker of imminent progression to symptomatic disease with a positive predictive value of 49% to 64% and a negative predictive value in excess of 95%.40,41 Viral load quantification may be most appropriately employed in targeting patients with preemptive prophylaxis. Its role in diagnosing both established and incipient CMV disease, including CMV pneumonitis, is limited at the current time.

Other tests used for the diagnosis of CMV disease include the hybrid capture assay, which is a non-PCR molecular quantitative and highly sensitive assay. This has been shown to be effective in lung transplant recipients, whereas the pp65 antigenemia assay is only semiquantitative. The quantitative PCR and hybrid capture assays are fully quantitative molecular techniques. Clinicians are increasingly using these quantitative assessments for predicting both the severity of CMV illness and impending CMV illness.42,43 The tests are commercially available, but it is recommended that the chosen technique should be validated in each individual transplant center.

Westall et al44 compared the CMV load in paired BAL and plasma samples in a prospective cohort of lung transplant recipients. As CMV selectively targets the lung allograft, they thought that measurement of CMV dynamics in the local environment of the lung compartment may offer advantages over the routine measurement in the peripheral blood. The threshold they used was arbitrarily set as 2 SDs above the mean BAL virus load in lung transplant recipients without histologically proven CMV infection and was calculated to be 46 000 copies per milliliter. The applicability of this virus load threshold in the diagnosis of histologically proven CMV infection was confirmed by the receiver-operating characteristic curve for the test. The authors compared 182 paired samples from 41 lung transplant recipients and found they were concordant for 79.7% of the paired samples; that is, virus detected or absent in 8.1% and 71.5% of paired samples, respectively.44 In the remaining 20.3% of samples, CMV was detected in the lavage fluid but not the plasma sample, but the converse was not found in any pairing. Their use of a threshold virus load in BAL samples to predict histologically proven CMV infection was associated with improved diagnostic sensitivity and specificity. They speculate that a high BAL CMV load not only predicts the presence of CMV inclusions in the lung allograft, but it is also more likely to be associated with specific symptomatic disease such as CMV pneumonitis. They noted that CMV DNA could be detected in lavage samples regardless of whether CMV inclusions were seen in the lung allograft early after transplantation, suggesting that ganciclovir prophylaxis was not completely suppressing CMV replication. Ideally, a threshold value could be established for CMV DNA in the blood that could be used to guide preemptive antiviral therapy. The study of lavage CMV load may also elucidate indirect long-term effects of subclinical CMV disease on the outcome of lung allografts. The diagnosis of CMV disease and its distinction from subclinical infection is yet another area where multidisciplinary correlation between histopathologists, cytopathologists, virologists, and immunologists is essential. Plasma PCR for CMV is inadequate as a diagnostic tool, but BAL is obviously an invasive procedure. Zedtwitz-Leibenstein et al45 examined the association of CMV DNA concentration in epithelial lining fluid and symptomatic CMV infection in lung transplant recipients. Whereas the quantitation of CMV DNA from BAL allowed no discrimination between symptomatic and asymptomatic infection in individual cases, when the urea dilution method was used to recalculate the CMV DNA concentration for the epithelial lining fluid diluted in the BAL, a CMV DNA level of more than 104 copies per milliliter was clearly associated with symptomatic disease. There is increasing literature on the comparison of PCR antigenemia assays, rapid blood culture detection, and non-PCR methods in the lung transplant community.46,47 Chemaly et al39 have correlated viral loads of CMV in blood and BAL specimens from lung transplant recipients determined by histology and immunohistochemistry.

In the setting of stem cell transplantation, CMV infection differs from solid organ allografts. Human stem cell transplant recipients are at high risk for developing CMV pneumonia because of the delayed reconstitution of cytotoxic T-cell responsiveness and the need to administer immunosuppressive drugs for the prevention of graft-versus-host disease. Without antiviral prophylaxis, the incidence of CMV pneumonia after allogeneic procedures is 20% to 35% compared with only 1% to 6% after autologous transplantation.48-50 The majority of episodes of CMV disease are due to reactivation of latent virus in seropositive recipients. Seronegative patients receiving stem cells from a seropositive donor have a lower risk of posttransplantation CMV disease than do seropositive recipients. The situation clearly contrasts with that seen after solid organ transplantation, where the greatest risk of disease is in the seronegative recipient of a seropositive donor. The incidence of CMV disease can be reduced by prophylaxis and preemptive treatment after the detection of subclinical viremia by pp65 antigenemia or PCR assay.

Prophylaxis is complicated by a high rate of neutropenia. Walter et al51 have shown that donor-derived, CMVspecific, cytotoxic T lymphocytes expanded in vitro and infused into the recipient resulted in effective reconstitution of serum immunity against CMV. Human leukocyte antigen tetramers can also be used to isolate antigen-specific cells for in vitro expansion and transfer to patients for antiviral immunotherapy. Indeed, initial studies with CMV-specific human leukocyte antigen class 1 tetramers have helped to define the nature of anti-CMV T-cell response in stem cell transplant patients and to determine a threshold cytotoxic T- lymphocyte level required for controlling CMV reactivation.52,53 Cytomegalovirus disease in candidates for stem cell transplantation is increasingly observed and it is a recognized risk for complications following transplantation. Patients with CMV disease prior to transplantation showed higher incidence if they suffered severe underlying immune deficiency syndromes compared with patients suffering from hematologic malignancy.54


Infections due to community respiratory viruses include influenza, parainfluenza, adenovirus, and respiratory syncytial virus, which are common in the general population where they typically present as mild self-limiting upper respiratory tract illnesses. It is not clear whether solid organ transplant recipients are at greater risk of acquiring these viral infections, but there is a greater propensity for these pathogens to involve the lower respiratory tract and therefore to cause more severe illness.55,56 The highest rate of infection among solid organ recipients is also in lung transplant recipients, up to 21% of whom develop respiratory viral infections.57 However, these patients are often more intensively monitored both clinically and functionally with spirometry as well as with clinically directed and surveillance bronchoscopies, which may partly account for the higher rates of infection described. Mortality rates from community respiratory viral infections vary from 0% to 20% in the various solid organ transplant populations, but the increased risk of long-term sequelae such as chronic rejection is obviously limited to lung transplant recipients.

There are rather limited treatment options currently and there is a significant risk of bacterial superinfection. In the lung transplant population, community respiratory viruses can cause bronchiolitis with or without pneumonitis.55,56 The respiratory symptoms are often nonspecific with fever, dyspnea, cough, and wheezing. The chest radiograph can be normal or show only subtle interstitial changes, whereas computed tomography is more sensitive, often showing ground-glass air space consolidation, nodules, and tree-in-bud opacities. The definitive diagnosis of community respiratory viral infection depends on the demonstration of virus in respiratory secretions obtained through nasopharyngeal swabs, nasal wash, or BAL. Although viral culture represents the gold standard, the delay in diagnosis of up to 2 weeks is not practical. However, the more rapid diagnostic tests including enzyme-linked immunosorbent assays and immunofluorescent techniques to identify viral antigens should be corroborated by standard viral culture. At the present time, reverse transcription PCR is limited by the large numbers of viral types and subtypes. Garbino et al58 assessed the frequency and potential role of respiratory viruses on disease outcome in hospitalized patients, which included lung transplant recipients who underwent BAL for an acute respiratory infection using real-time PCR for 11 different viruses as well as Mycoplasma pneumoniae, Chlamydia pneumoniae, and Legionella pneumophila. They found that only 30% of cases that were virus positive by molecular methods were also positive by cell culture analysis. Rhinovirus was the most frequently identified virus, found in 56% of cases, followed by respiratory syncytial virus in 27% of cases. In the lung transplant recipient population, the rate of viral infections was 55% in cases with respiratory symptoms compared with only 4% in control subjects.

In these lung transplant recipients, respiratory viral infections were associated with significant lung function abnormalities. Pulmonary function decline persisted for more than 3 months in a substantial number of the lung transplant recipients after the acute episode of infection. Garbino et al58 also found that pre-BAL and post-BAL specimens tested negative for the virus identified during the period of respiratory symptoms, strongly suggesting that it is unusual to detect respiratory viruses without the presence of respiratory symptoms. They concluded that respiratory viral infections are frequent in lung transplant recipients, particularly during the autumn/winter season when community outbreaks occur, and that these viral infections played an important role in the onset of respiratory symptoms. It appeared that asymptomatic viral shedding was uncommon.

A study by Khalifah et al59 has shown that communityacquired respiratory viral infections are a distinct risk for bronchiolitis obliterans syndrome and death. These trends were more pronounced in patients with evidence of lower respiratory tract community- acquired respiratory viral infections compared with upper tract involvement alone. The potential mechanisms responsible for the association between community-acquired respiratory viral infections and bronchiolitis obliterans syndrome have been reported at length. There have been studies linking respiratory viral infections in lung transplant recipients rather than with acute rejection, and in some cases these were diagnosed concomitantly.60,61 However, it is extremely difficult to make a firm diagnosis of acute cellular rejection in the presence of viral infection, which can mimic rejection histologically with interstitial and often perivascular mononuclear cell infiltrates.62 The possibility is raised that respiratory viral infections may masquerade as acute cellular rejection histologically rather than there being a true association. It is likely that respiratory viral infection in the setting of lung transplantation can produce chronic changes that either initiate or accelerate bronchiolitis obliterans syndrome.59 However, the cause of the variability in onset of postviral changes still needs to elucidated.

It is noted in the nontransplant literature that respiratory viral infection with respiratory syncytial virus can result in a chronic asthma phenotype that may persist for many years.63 There may be direct viral-dependent injury of airway epithelium with epithelial expression of injury response genes and cascades of acute and chronic inflammatory mediators. Both alloimmune and innate mechanisms of airway fibrosis need to be considered. Kotsimbos et al64 have described C pneumoniae serology in donors and recipients and the risk of bronchiolitis obliterans syndrome following lung transplantation. High donor titers were associated with bronchiolitis obliterans syndrome in the recipient, whereas high recipient titers were inversely associated with bronchiolitis obliterans syndrome. The risk of developing bronchiolitis obliterans syndrome was 75% in the case of primary sero-mismatch for C pneumoniae (donor+/recipient), whereas a reverse mismatch had a risk of 4.6%. Chlamydia pneumoniae is a common but difficult to diagnose respiratory pathogen that has a recognized propensity to latency. It is known to have the potential to cause chronic persistent and latent infection in airways; therefore, the potential transfer from donor to recipient can be expected to contribute to the development of bronchiolitis obliterans syndrome.64,65 These phenomena are also of interest in the relationship between C pneumoniae infection and chronic inflammatory airway diseases, including asthma and chronic obstructive pulmonary disease outwith the transplant population. Gerna et al66 have looked at the impact of human metapneumovirus (hMPV) and human CMV with other respiratory viruses on the lower respiratory tract infections of lung transplant recipients. They examined 49 symptomatic and 26 asymptomatic lung transplant recipients during 3 consecutive winter/spring seasons. In the symptomatic patients, hMPV predominated over the other respiratory viruses in BAL samples, being responsible for 60% of positive specimens, whereas other viruses were present in nasopharyngeal aspirates at a comparable rate. They found real-time PCR to be superior to monoclonal antibody for detecting infection. In addition they found CMV in association with a respiratory virus in 4 of 18 CMVpositive patients and a high CMV viral load was detected in CMV- positive BAL samples and pneumonia. They also observed that coinfections and sequential infections by CMV and respiratory viruses were significantly more frequent in patients with acute rejection and steroid treatment. There is, of course, the caveat of the difficulty of diagnosing acute cellular rejection histologically in the presence of confirmed viral infections (Figure 12).62 They concluded that about 50% of respiratory tract infections in lung transplant recipients were associated with one or more respiratory viruses, and that hMPV predominates in the lavage of symptomatic lung transplant recipients, thereby suggesting a causative role in lower respiratory tract infection in this patient group.

Human metapneumovirus infection of the lung transplant population has only been described relatively recently and there are some early data on clinical presentation and epidemiology. Larcher et al67 conducted a prospective study using nasopharyngeal aspirates and BALs from immunocompromised adults, lung transplant recipients, and a bone marrow recipient. They also examined hospitalized children who were not immunocompromised. They found 25% of lung transplant recipients were positive for hMPV together with 14% of nonimmunocompromised children. Most of the cases clustered within 2 outbreaks in late autumn and early spring. In immunocompromised patients, hMPV was isolated throughout the entire observation period, and the same viral strains were noted to circulate in hospitalized children and in lung transplant recipients. A different hMPV strain was isolated during the interepidemic period, suggesting that hMPV infections may be transmitted to lung transplant recipients independently from community outbreaks. The clinical signs and symptoms varied from nil to severe pneumonia or acute graft rejection. The researchers also noted that the identification of replicating hMPV significantly correlated with acute rejection symptoms at the time of sample collection and suggested that hMPV be added to the list of pathogens that are possibly associated with episodes of allograft rejection.67

Human metapneumovirus was initially isolated in The Netherlands in 2001 and analysis of stored nasopharyngeal swab samples has indicated that it may account for a significant proportion of respiratory tract infections with previously unknown etiology.68-70 This infection can cause acute upper respiratory tract infection or severe bronchiolitis and pneumonitis with a pattern similar to that seen in respiratory syncytial virus infections. Sumino et al71 evaluated consecutive BAL and bronchial wash samples from 688 patients, of whom 72% were immunocompromised and were predominantly lung transplant recipients. They used quantitative real-time PCR for detection of hMPV and correlated positive results with clinical outcome, viral cultures, in situ hybridization, and histopathologic assessment of the lung. They identified 6 cases of hMPV that had a similar frequency and age range as other paramyxoviral infections. Four of the 6 infections occurred in immunocompromised patients and this infection was confirmed by in situ hybridization for the viral nucleocapsid gene. Histopathologic assessment of the lung tissue samples showed acute and organizing injury with smudge cell formation distinct from findings in other paramyxoviral infections. The smudged hyperchromatic nuclei were in enlarged type 2 pneumocytes and resembled the smudge cells found in adenovirus infection. However, immunostaining for adenovirus, CMV, and herpes simplex virus were all negative. The 2 lung transplant recipients in whom tissue was examined showed no evidence of acute rejection. Sumino et al71 were unable to detect hMPV mRNA in the smudge cells although it was seen in the alveolar and airway epithelial cells. They postulate that the smudge cells could be reactive in hMPV infection rather than representing infected cells with viral inclusion bodies.

In contrast, in nonhuman primates, hMPV replication occurs mainly in ciliated epithelial cells and rarely in type 1 pneumocytes. In this setting, the histopathologic features include erosive and inflammatory changes confined to the conducting airways. It is clear that these preliminary observations of a pattern of acute lung injury and smudge cell formation in hMPV infection will need to be followed up with additional histologic studies before it can be concluded that this is a distinctive reaction pattern. Although there is no established treatment for hMPV infection, the diagnosis and detection in immunocompromised patients may at least prevent these high-risk patients from undergoing additional diagnostic procedures with attendant risks of morbidity and even mortality.


Tuberculosis has been reported in between 0.5% and 2% of organ transplant recipients in the United States and Europe,72-74 but in up to 15% of recipients in endemic areas. 11 Although it is relatively uncommon posttransplantation in developed countries, it should still be recognized that the annualized rate of infection is between 30- and 100-fold higher than that of the general population.72-74 The predominant mechanism for the development of active tuberculosis after solid organ transplantation is thought to be reactivation, although this is difficult to verify. Less commonly, posttransplant tuberculosis can be acquired through nosocomial outbreaks and even donor transmission. The onset of infection is usually within the first year following transplantation, and about 50% of patients have tuberculosis restricted to the lungs. However, 33% have been reported to have disseminated disease.72 The overall rate of lung involvement in solid organ transplant tuberculosis was about 70%. Although fever is the most common presenting symptom in patients with disseminated disease, it is only found in about two thirds of patients with pulmonary tuberculosis.75,76 The chest radiograph abnormalities are highly variable, encompassing focal infiltrates, miliary nodules, pleural effusions, diffuse interstitial lung infiltrates, and, less commonly, cavitary lung disease. The mortality remains high, in the range of 25% to 40%. This is partly because of the infection itself, but also the impact of enhanced rejection and graft loss among treated but often suboptimally immunosuppressed patients. Many patients will have comorbid conditions that contributed to the development of tuberculosis in the first place and which further contribute to mortality.

Among lung transplant recipients, nontuberculous mycobacteria are more common than Mycobacterium tuberculosis. 75,76 Within this group, the majority are infections by Mycobacterium avium complex. Pulmonary nontuberculous mycobacteria infection tends to occur late in the posttransplantation period and is associated with chronic rejection (ie, bronchiolitis obliterans syndrome) in more than 50% of cases. However, the mycobacterial infection is not usually a primary cause of death in these patients. Nontuberculous mycobacterial infection is less common in other solid organ recipients, with pulmonary involvement in fewer than 2% of heart transplant recipients.77 In stem cell transplant recipients, tuberculosis is also uncommon in nonendemic areas, but with a rising frequency where tuberculosis is more prevalent. Infection with nontuberculous mycobacteria has been uncommon in stem cell transplant recipients, although a higher infection rate has been reported from the Memorial Sloan-Kettering Cancer Center. 78

The specialist histopathologist has a significant role to play in the diagnosis of mycobacterial infections in all transplant populations, both tuberculous and atypical infections (Figure 13). The differential diagnosis of granulomatous inflammation in the lung is wide, and mycobacterial infection must always be considered, with mandatory use of special stains to identify causative organisms. However, the yield is often low and correlation should always be made with culture and, if available, PCR results in these settings. The diagnosis may be made on routine bronchoscopy because patients may be asymptomatic. In the lung transplant population, the importance of examination of the explanted lungs may be significant in the identification of previously unsuspected mycobacterial infection. 32 In the study reported by Bravo et al,76 tuberculosis was diagnosed in 12 patients from series of 187 lung transplant patients during a 12-year period. Six of the 12 patients had the diagnosis determined from the explanted lungs and the remainder were diagnosed during followup, with fever and dyspnea being the most common symptoms. In their study, samples of explanted lungs were routinely obtained for culture as well as histopathologic studies. After formalin inflation fixation, the lungs were cut sagittally into 1-cm-thick slices and, along with hilar samples, 5 parenchymal samples per lobe were taken, with additional sampling of any further localized lesions. Bronchial brush smears of the implanted lungs were routinely obtained for histopathologic analysis and mycobacterial culture. If the Ziehl-Neelsen stain proved negative in cases with caseating granulomatous lesions, the diagnosis of tuberculosis was not established as definite, but rather considered highly suggestive. In 2 of the 6 patients diagnosed on explant lung pathology, the diagnosis was conclusive; the other 4 diagnoses were considered highly suggestive.

Kesten and Chaparro79 reported 5 of 8 lung transplant recipients with mycobacterial infections as having infection in their native lungs. They concluded in their retrospective review of 219 transplant procedures, of which 16 were single-lung and 159 were double-lung transplants, that mycobacterial disease was a rare occurrence following lung transplantation and that the greatest evidence of infection was in native lungs. They also thought that cultures for mycobacteria in surveillance BAL specimens were likely to be unnecessary in the absence of symptoms.


Pneumocystis jiroveci (formerly carinii) pneumonia (PJP) is a far less common infection following the widespread introduction of chemoprophylaxis. Prior to initiation of prophylaxis, the incidence had been described as up to 33% for heart lung recipients, 11% for liver recipients, and 4% for kidney and heart recipients.80 The risk of PJP falls beyond the second to sixth month posttransplantation period which, as noted previously, is the period of greatest risk for opportunistic infections. However, in lung transplant recipients the risk does not decline beyond the first year and prophylaxis may need to be continued. Lung recipients with chronic allograft rejection appear to be at increased risk for the late development of PJP, and in some centers indefinite prophylaxis is advocated.80 In solid organ transplant recipients, PJP often presents subacutely with dyspnea, fever, and cough, and as radiographic abnormalities, which are best demonstrated on computed tomography scan, where they are typically bilateral and may appear as interstitial, alveolar, or ground-glass opacities. The diagnosis can be established by BAL alone in about 90% of cases, but the performance of transbronchial biopsies will modestly enhance the yield (Figure 14). In the lung transplant population, transbronchial biopsies are advocated for the further information required about the state of the graft, including acute cellular rejection, which PJP can mimic.62 In stem cell transplant recipients, prophylaxis also reduces the incidence of PJP, but trimethoprine/sulphamethoxazole is often poorly tolerated by this population because of the associated marrow suppression, and adverse reactions requiring discontinuation of the drug have been reported in up to 60% of patients.81

Prophylaxis is generally given from the time of engraftment until 6 months posttransplantation for allogeneic recipients, but will need to be extended beyond this point for those with ongoing immunosuppressive therapy and chronic graft-versus-host disease. The course of PJP in stem cell transplant recipients displays a more fulminant onset and progression than that seen in solid organ allograft recipients. Despite effective therapy, mortality rates are as high as 90% for infections within the first 6 months and 40% for later infections.82 In the lung transplant population, it is important to remember that PJP can present with atypical histologic patterns including granulomatous inflammation. The diagnosis of PJP can be enhanced by immunofluorescent staining, simple PCR, or nested PCR.83-87 Polymerase chain reaction is emerging as an important tool for the epidemiologic study of Pneumocystis infection in high-risk patients and produces high-sensitivity negative predictive values for BAL specimens.88 Pneumocystis carinii f sp hominis has been renamed Pneumocystis jiroveci and, as the organism cannot be cultured, diagnosis relies on microscopic identification of the organism by the histopathologist and/or cytopathologist using stains or antibodies. These tests are simple but insensitive, and obviously require expertise for accurate interpretation.

Therefore, there has been great enthusiasm for the development of molecular methods with greater specificity and sensitivity. Recently, Alvarez-Martinez et al89 have described a real-time PCR assay for the detection of Pneumocystis in BAL fluid using primers and fluorescent resonance energy transfer probes that target the CDC2 gene of P jiroveci. They found no cross-reactivity with other pathogens and a 21% increase in clinical sensitivity. Brancart et al86 developed a real-time PCR for the detection and quantification of P jiroveci in lavage specimens based on primers and probes targeting the gene encoding beta tubulin. This assay was able to detect 50 DNA copies per milliliter of a standard plasmid containing the target sequence. Of 53 lavage samples sent for diagnosis of pneumocystosis, the PCR-negative samples were negative by microscopy, and of the 24 PCR-positive samples, 8 were positive by lavage microscopy. The copy numbers of target genes were between 4.4 x 109 and 2.8 x 106/mL for the microscopic-positive samples and between 8 and 9.2 x 103/mL for the microscopic-negative samples. However, there have also been reports of stem cell recipients who were thought to have Pneumocystis infection who have not had this confirmed at autopsy. Clearly, further work is needed to establish the true clinical utility of molecular diagnostic methods for PCR in both this setting and solid organ transplantation.


There are many exciting developments in the study of infections of the immunosuppressed transplant population, including both solid organ and stem cell recipients. The accurate and timely diagnosis of infection needs the input of many disciplines within pathology, imaging, clinical, immunologic, and infectious disease specialities. As better immunosuppression reduces the frequency and severity of acute rejection, it appears that it also reduces the incidence of chronic rejection. Infection by both common and opportunistic organisms remains a significant cause of morbidity and mortality, and therefore a huge challenge for improving patient outcomes yet further. This article has aimed to highlight areas of interest and development that are pertinent to the diagnostic histopathologist, particularly the pulmonary specialist, in maintaining their critical diagnostic role as part of the multidisciplinary transplant team.


1. Kotloff RH, Vivek N, Crawford SW. Pulmonary complications of solid organ and hematopoietic stem cell transplantation. Am J Respir Crit Care Med. 2004; 170:22-48.

2. Maurer JR, Tullis E, Grossman RF, Velland H, Winton TL, Patterson GA. Infectious complications following isolated Lung transplantation. Chest. 1992; 101:1056-1059.

3. Cisneros JM, Munoz P, Torre-Cisneros J, et al, Spanish Transplantation Infection Study Group. Pneumonia after heart transplantation: a multi-institutional study. Clin Infect Dis. 1998;27:324-331.

4. Kramer MR, Marshall SE, Starnes VA, Gamberg P, Amita Z, Theodore J. Infectious complications in heart-lung transplantation: analysis of 200 episodes. Arch Intern Med. 1993;153:2010-2016.

5. Kusne S, Dummer JS, Singh N, et al. Infections after liver transplantation: an analysis of 101 consecutive cases. Medicine (Baltimore). 1988;67:132-143.

6. Fishman JA, Rubin RH. Infection in organ transplant recipients. N Engl J Med. 1998;338:1741-1751.

7. Weill D, Dey GC, Hicks RA, et al. A positive donor gram stain does not predict outcome following lung transplantation. J Heart Lung Transplant. 2002; 21:555-558.

8. Chan KM, Allen SA. Infectious pulmonary complications in lung transplant recipients. Semin Respir Infect. 2002;17:291-302.

9. Kocher AA, Bonaros N, Dunkler D, et al. Long-term results of CMV hyperimmune globulin prophylaxis in 377 heart transplant recipients. J Heart Lung Transplant. 2003;22:250-257.

10. Akalin E, Sehgal V, Ames S, et al. Cytomegalovirus disease in high-risk transplant recipients despite ganciclovir or valganciclovir prophylaxis. Am JTransplant. 2003;3:731-735.

11. John GT, Shankar V, Abraham AM, Mukundan U, Thomas PP, Jacob CK. Risk factors for post-transplant tuberculosis. Kidney Int. 2001;60:1148-1153.

12. Shaukat A, Faris Bakri, Young P, et al. Invasive filamentous fungal infections in allogeneic hematopoietic stem cell transplant recipients after recovery from neutropenia: clinical, radiologic, and pathologic characteristics. Mycopathologia. 2005;159:181-188.

13. Bag R. Fungal pneumonias in transplant recipients. Curr Opin Med. 2003; 9:193-198.

14. Zihlif M, Khanchandani G, Ahmed HP, Soubani AO. Surgical lung biopsy in patients with haematological malignancy or hematopoietic stem cell transplantation and unexplained pulmonary infiltrates: improved outcome with specific diagnosis. Am J Hematol. 2005;78:94- 99. 15. Lass-Florl C, Gunsilus E, Gastl G, et al. Diagnosing invasive aspergillosis during antifungal therapy by PCR analysis of blood samples. J Clin Microbiol. 2004;42:4154-4157.

16. Van Burik J-A, Myerson D, Schreckhise RW, Bowden RA. Panfungal PCR assay for detection of fungal infection in human blood specimens. J Clin Microbiol. 1998;36:1169-1175.

17. Subira M, Martino R, Rovira M, Vazquez L, Serrano D, De La Camara R. Clinical applicability of the new EORTC/MSG classification for invasive pulmonary aspergillosis in patients with hematological malignancies and autopsy-confirmed invasive Aspergillosis. Ann Hematol. 2003;82:80-82.

18. Ascioglu S, Rex JH, de Pauw B, et al. Defining opportunistic invasive fungal infections in immunocompromised patients with cancer and hematopoietic stem cell transplants: an international consensus. Clin Infect Dis. 2002;34:7-14.

19. Kawazu M, Kanda Y, Goyama S, et al. Rapid diagnosis of invasive pulmonary aspergillosis by quantitative polymerase chain reaction using bronchial lavage fluid. Am J Hematol. 2003;72:27-30.

20. Zaas AK, Alexander BD. Galactomonannan and advances in fungal diagnostics. Curr Opin Organ Transplant. 2005;10:307-311.

21. Kubak BM. Fungal infection in lung transplantation. Transpl Infect Dis. 2002;4(suppl 3):24-31.

22. Singh N, Husain S. Aspergillus infections after lung transplantation: clinical differences in type of transplant and implications for management. J Heart Lung Transplant. 2003;22:258- 266.

23. Helmi M, Love RB, Welter D, Cornwell RD, Meyer KC. Aspergillus infection in lung transplant recipients with cystic fibrosis: risk factors and outcomes comparison to other types of transplant recipients. Chest. 2003;123:800-808.

24. Cahill BC, Hibbs JR, Savik K, et al. Aspergillus airway colonization and invasive disease after lung transplantation. Chest. 1997;112:1160-1164.

25. Mehrad B, Paciocco G, Martinez FJ, Ojo TC, Iannettoni MD, Lynch JP III. Spectrum of Aspergillus infection in lung transplant recipients: case series and review of the literature. Chest. 2001;119:169-175.

26. Sharma OP, Chwogule R. Many faces of pulmonary Aspergillosis. Eur Respir J. 1998;12:705-715.

27. Kessler R, Massard G, Warter A, Wihlm JM, Weitzenblum E. Bronchialpulmonary artery fistula after unilateral lung transplantation: a case report. J Heart Lung Transplant. 1997;16:674- 677.

28. Brenier-Pinchart MP, Lebeau B, Devouassoux G, et al. Aspergillus and lung transplant recipients: a mycologic and molecular epidemiologic study. J Heart Lung Transplant. 1998;17:972- 979.

29. Shreeniwas R, Schulman LL, Berkman YM, McGregor CC, Austin JHM. Opportunistic bronchopulmonary infections after lung transplantation: clinical and radiographic findings. Radiology. 1996;200:349-356.

30. Sole A, Morant P, Salavert M, Peman J, Morales, P, Valencia Lung Transplant Group. Aspergillus infections in lung transplant recipients: risk factors and outcome. Clin Microbiol Infect. 2005;11:359-365.

31. Tazelaar HD, Baird AM, Mill M, Grimes MM, Schulman LL, Smith CR. Bronchocentric mycosis occurring in transplant recipients. Chest. 1989;96:92-95.

32. Stewart S, McNeil K, Nashef SA, Wells FC, Higenbottam TW, Wallwork J. Audit of referral and explant diagnosis in lung transplantation: a pathologic study of lungs removed for parenchymal disease. Heart Lung Transplant. 1995;14: 1173-1186.

33. Ayman O, Soubani MD, Pranatharthi H, Chandrasekar MD. The clinical spectrum of pulmonary aspergillosis. Chest. 2002;121:1988- 1999.

34. Sanchez JL, Kruger RM, Paranjothi S, et al. Relationship of cytomegalovirus viral load in blood to pneumonitis in lung transplant recipients. Transplantation. 2001;72:733-735.

35. Hopkins PM, Aboyoun CL, Chhajed PN, et al. Prospective analysis of 1,235 transbronchial lung biopsies in lung transplant recipients. J Heart Lung Transplant. 2002;21:1062-1067.

36. Trulock EP, Ettinger NA, Brunt EM, Pasque MK, Kaiser LR, Cooper JD. The role of transbronchial lung biopsy in the treatment of lung transplant recipients. Chest. 1992;102:1049-1054.

37. Collins J, Muller NL, Kazerooni EA, Paciocco G. CT findings of pneumonia after lung transplantation. AJR Am J Roentgenol. 2000;175:811-818.

38. Franquet T, Lee KS, Muller NL. Thin-section CT findings in 32 immunocompromised patients with cytomegalovirus pneumonia who do not have AIDS. AJR Am J Roentgenol. 2003;181:1059-1063.

39. Chemaly RF, Yen-Lieberman B, Castilla EA, et al. Correlation between viral loads of cytomegalovirus in blood and bronchoalveolar lavage specimens from lung transplant recipient determined by histology and immunohistochemistry. J Clin Microbiol. 2004;42:2168- 2172.

40. Humar A, Gregson D, Caliendo AM, et al. Clinical utility of quantitative cytomegalovirus viral load determination for predicting cytomegalovirus disease in liver transplant recipients. Transplantation. 1999;68:1305-1311.

41. Rayes N, Seehofer D, Schmidt CA, et al. Prospective randomised trial to assess the value of preemptive oral therapy for CMV infection following liver transplantation. Transplantation. 2001;72:881-885.

42. Senechal M, Dorent R, du Montcel ST, et al. Monitoring of human cytomegalovirus infections in heart transplant recipients by pp65 antigenemia. Clin Transplant. 2003;17:423-427.

43. Zamora MR. Cytomegalovirus and lung transplantation. Am J Transplant. 2004;4:1219-1226.

44. Westall GP, Michaelides A, Williams TJ, Snell GI, Kotsimbos TC. Human cytomegalovirus load in plasma and bronchoalveolar lavage fluid: a longitudinal study of lung transplant recipients. J Infect Dis. 2004;190:1076-1083.

45. Zedtwitz-Liebenstein K, Jaksch P, Bauer C, et al. Association of cytomegalovirus DNA concentration in epithelial lining fluid and symptomatic cytomegalovirus infection in lung transplan