|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Division of Pneumology (Drs. Tamm, Traenkle, Soler, and Bolliger), Department of Internal Medicine, and Institute of Pathology (Mr. Grilli and Drs. Dalquen and Cathomas), University Hospital Basel, Basel, Switzerland.
Correspondence to: Michael Tamm, MD, Division of Pneumology, University Hospital Basel, Petersgraben 4, 4031 Basel, Switzerland; e-mail: mtamm{at}uhbs.ch
| Abstract |
|---|
|
|
|---|
Methods: Five hundred eighty consecutive BAL procedures were analyzed prospectively in 442 immunocompromised and 126 nonimmunocompromised control subjects. CMV culture in BAL fluid was performed by shell vial assay and immunostaining using three monoclonal anti-CMV antibodies.
Results: The incidence of culture results positive for CMV in the BAL fluid varied from 20 to 30% in HIV-positive patients, in patients following stem cell or renal transplantation, and in patients with autoimmune disease or lung fibrosis treated with immunosuppressive agents. CMV was cultured from 4.4% of BALs in patients treated with high-dose chemotherapy and from 2.4% of control subjects. CMV disease developed in 37 patients; in 18 of these patients, CMV pneumonitis was present. The results of CMV immunostaining were positive in a total of 22 BALs, all in patients with CMV disease. The sensitivity, specificity, and positive and negative predictive values of positive CMV immunostaining results for the diagnosis of CMV pneumonitis were 88.9%, 98.6%, 72.7%, and 99.5%, respectively.
Conclusion: The incidence of pulmonary CMV infection is similar in different groups of immunocompromised patients except for patients following high-dose chemotherapy. CMV immunostaining in the BAL fluid is a very helpful method to diagnose CMV pneumonitis in these patients.
Key Words: BAL cytomegalovirus infection cytomegalovirus immunostaining
| Introduction |
|---|
|
|
|---|
The diagnosis of CMV pneumonitis is usually based on symptoms of fever, dyspnea, hypoxemia, and diffuse infiltrates on chest radiograph in combination with the detection of CMV in the BAL fluid. A definitive diagnosis of CMV pneumonitis can be made histologically showing cytomegalic cells associated with inflammatory reactions and tissue destruction in transbronchial or open-lung biopsy specimens. However, transbronchial biopsies are performed rarely because of the risk of bleeding, especially in thrombopenic patients.
Immunostaining with anti-CMV antibodies in BAL fluid has been shown to be helpful for the diagnosis of CMV pneumonitis in selected groups of patients.17 18 19 20 In this study, the incidence of CMV detection in the BAL fluid of different groups of immunocompromised patients was determined and the diagnostic value of CMV immunostaining for detection of CMV pneumonitis was evaluated.
| Materials and Methods |
|---|
|
|
|---|
BAL
Indications for BAL were fever, respiratory symptoms, and/or
infiltrates seen on the chest radiograph. To obtain BAL fluid, 150 to
300 mL of 0.9% NaCl was instilled into the middle lobe or the segment
of the most prominent radiologic infiltrate.21
BAL fluid
was investigated routinely for bacterial, mycobacterial, and fungal
growth. Ten milliliters of BAL fluid was sent to the laboratory for
bacterial culture, and another 10 mL was sent for virus culture.
Conventional cytology was performed with the remaining fluid. Following
centrifugation, the cell pellet was resuspended in cell culture medium
and smears were prepared. Routine staining included Grocott stain for
Pneumocystis carinii and Papanicolaou stain. Enlarged cells
that contained cyanophilic to eosinophilic intranuclear inclusions
surrounded by a halo were considered to be diagnostic for CMV.
Immunostaining for CMV was performed using a panel of commercially
available antibodies directed against immediate, early, and late viral
proteins (Biotest- CMV, immediate early antigen [76kD];
Biosoft-Clonatec; Varilhes, France; Dako-CMV, delayed early DNA binding
protein p52; Dako; Glostrop, Denmark; and Clonab-CMV, immediate early
antigen pp--65; Biotest; Dreieich, Germany). Positive
immunostaining of one or several anti-CMV antibodies was considered to
be a positive result. Analyses of the immunostaining results were
performed blindly without knowledge of the clinical findings and the
results of CMV cell cultures. CMV culture was performed by shell vial
assay, spinning the BAL fluid onto human embryonic fibroblasts and
determining the presence of CMV immediate early antigens by
immunofluorescence and the presence of CMV DNA by in situ
hybridization after 1 day and 5 days, respectively.22
Clinical Data
Medical records and chest radiograph findings of all patients
with a culture positive for CMV in their BAL fluid were analyzed
retrospectively for the presence of CMV disease within a period of 3
months before and after BAL was performed. Clinical evaluations for the
presence of CMV disease were performed blindly without knowledge of the
results of cytology and immunostaining in the BAL fluid.
Definitions
Pulmonary CMV infection was defined as a positive CMV culture
result in the BAL fluid, irrespective of symptoms or signs of disease.
CMV disease was defined as CMV infection combined with symptoms and/or
tissue invasion of CMV, including CMV pneumonitis, CMV retinitis, CMV
hepatitis, and CMV syndrome. CMV pneumonitis was defined as the
detection of CMV by culture combined with the presence of fever,
respiratory symptoms (eg, dyspnea, hypoxemia, and reduced
diffusion capacity in pulmonary function tests), and interstitial
infiltrates on chest radiograph. Immunostaining with a monoclonal
antibody against CMV in the BAL fluid and viral inclusion bodies in BAL
cells detected by conventional cytology were not used as diagnostic
markers for the definition of CMV pneumonitis. CMV syndrome was defined
as fever combined with at least one of the following symptoms: elevated
liver enzymes (aspartate aminotransferase and alanine
aminotransferase at least twofold); leukopenia (leukocyte count,
< 3 x 109/L); thrombopenia (platelet count,
< 50 x 109/L) in the absence of other
infections; and graft rejection. CMV hepatitis was present if there
were elevated levels of liver enzymes and if CMV could be detected in a
liver biopsy specimen. CMV retinitis was defined by characteristic
funduscopic lesions. GI CMV disease included GI symptoms combined with
biopsy specimen-proven CMV tissue invasion.
| Results |
|---|
|
|
|---|
|
|
Conventional Cytology and CMV Immunostaining
Inclusion bodies were observed in the BAL fluid of 11 of 442
patients (2.5%) in the immunocompromised patient group, including 4
stem cell transplant recipients, 4 HIV-positive patients, and 3 renal
transplant recipients. Twenty-two of the immunocompromised patients had
positive immunostaining results for CMV in the BAL fluid (Table 2)
. In
contrast, neither inclusion bodies nor positive immunostaining findings
were observed in the BAL fluid of nonimmunocompromised patients.
Immunostaining findings were positive in the BAL fluid of 12 of 227
HIV-positive patients (5.3%), 7 of 69 patients (10.1%) following stem
cell transplantation, and 3 of 49 solid-organ transplant recipients
(6.1%). There were no positive immunostaining findings in patients
with autoimmune disease/lung fibrosis and none in patients following
high-dose chemotherapy.
The sensitivity of CMV inclusion bodies in regard to positive CMV immunostaining findings was 36.3%, the specificity was 99.3%, the positive predictive value was 72.7%, and the negative predictive value was 96.8%.
CMV Disease (Including CMV Pneumonitis)
A diagnosis of CMV disease, including CMV pneumonitis, was made in
37 patients, as shown in Table 2
. There was no CMV disease in patients
undergoing high-dose chemotherapy or in patients with autoimmune
disease treated with immunosuppressive agents. Seven patients (14.3%)
developed CMV disease following renal transplantation and 8 patients
after stem cell transplantation (11.6%), and all of them showed a
positive p65 antigenemia in the peripheral blood. Seven of the eight
patients in the stem cell transplantation group experienced
graft-vs-host disease. Three of the seven patients who developed CMV
disease after renal transplantation were CMV seronegative and had
received a graft from a seropositive donor. In the HIV-positive group,
CMV disease was present in 22 patients (9.7%). Retinitis was found in
six patients, esophagitis was found in five patients, and CMV hepatitis
was found in one patient from the HIV-positive group. In five of these
cases, patients also had Kaposis sarcoma.
CMV pneumonitis was diagnosed in a total of 18 patients. In four HIV-positive patients, CMV pneumonitis was combined with CMV esophagitis or retinitis. CMV syndrome occurred in seven patients; four were stem cell transplant recipients and three were renal transplant recipients. All except one patient with CMV disease following allogenic stem cell transplantation experienced graft-vs-host disease. In three of these patients, bronchiolitis obliterans developed in the later course. Two of four renal transplant recipients with CMV pneumonitis were CMV seronegative before transplantation but had received a graft from a seropositive donor.
Diagnostic Value of CMV Immunostaining
CMV immunostaining was positive in a total of 22 patients, all of
them suffering from CMV disease. None of the patients without CMV
disease showed a positive CMV immunostaining finding in the BAL fluid,
but CMV disease was present in 15 patients without positive
immunostaining findings. The sensitivity and specificity of positive
immunostaining findings for the diagnosis of CMV disease, including CMV
pneumonitis, were 59.5% and 100%, respectively, and the positive and
negative predictive values were 100% and 96.4%, respectively (Table 3
). Sixteen of 18 patients with CMV pneumonitis revealed positive
immunostaining findings in BAL fluid (Table 4
). There was no evidence of CMV pneumonitis in six patients with
positive immunostaining findings in the BAL. However, all of these six
patients with positive immunostaining findings, but without clinical
evidence of CMV pneumonitis, had extrapulmonary CMV disease.
Sensitivity, specificity, and positive and negative predictive values
for the diagnosis of CMV pneumonitis by immunostaining were 88.9%,
98.6%, 72.7%, and 99.5%, respectively.
|
|
| Discussion |
|---|
|
|
|---|
The incidence of CMV retrieval from the BAL fluid of HIV-infected patients varies between 20% and 70% in the literature.7 8 9 10 Except for patients undergoing lung transplantation, only a small amount of data is available concerning the incidence of pulmonary CMV infection in solid-organ transplant recipients and in patients treated with immunosuppressive agents for interstitial lung disease or autoimmune disease. Surprisingly, except in patients following high-dose chemotherapy, there was no significant difference in the incidence of pulmonary CMV infection within the different study groups of immunocompromised patients (Table 2) . CMV disease as defined by internationally used criteria1 2 was found in patients following solid-organ and stem cell transplantation as well as in patients with AIDS. None of the patients who had undergone immunosuppression for lung fibrosis or autoimmune disease developed CMV disease, despite an incidence of pulmonary CMV infection that was similar to that in other immunocompromised patient groups (Table 2) . The risk for pulmonary CMV infection, therefore, might be influenced primarily by the degree of immunosuppression rather than by the underlying disease. In contrast, the risk for developing CMV disease, including CMV pneumonitis, depends on the degree of immunosuppression and the underlying disease. At least in transplant recipients, allogenic stimulation and associated CMV reactivation may be partly responsible for this difference.25
The diagnostic and prognostic value of pulmonary CMV infection in HIV-positive patients is controversially discussed in the literature.7 8 9 10 11 12 13 14 15 16 Applying criteria that are identical to those for transplant recipients, the incidence of CMV disease and CMV pneumonitis in HIV-positive patients was 9.7% and 4.4%, respectively, in our study. This rate is lower than in the groups following transplantation, but this difference is not statistically significant. In accordance with most authors, we believe that pulmonary CMV infection, as defined by a positive CMV culture finding in BAL fluid, is of limited clinical importance. However, our results show that positive immunostaining findings for CMV also indicate invasive CMV disease in the HIV patient group. Therefore, HIV-positive patients with positive CMV immunostaining findings in BAL may benefit from antiviral therapy.
The diagnostic value of CMV immunostaining for the diagnosis of CMV pneumonitis is not only good in HIV-positive patients but also in other immunosuppressed patients, as shown by a high sensitivity and specificity. All patients showing a positive immunostaining finding in BAL fluid who did not fulfill the clinical criteria of CMV pneumonitis had extrapulmonary CMV disease. Cases of CMV pneumonitis not accompanied by a positive immunostaining finding were rare. Therefore, positive CMV immunostaining finding in BAL fluid is a useful marker for the presence of invasive pulmonary CMV infection (ie, CMV pneumonitis) and helps to select patients who may be treated with antiviral agents. The differentiation between pulmonary CMV infection and CMV pneumonitis is especially important in the nontransplantation patient groups. In allogenic stem cell transplant recipients, the risk for developing CMV pneumonitis is high in patients with pulmonary CMV infection.26 Based on the considerable mortality rate in this patient group, antiviral treatment is usually initiated in patients with positive CMV culture findings in BAL fluid, irrespective of the results of immunostaining or cytology.
Although inclusion bodies are the hallmark of CMV disease in conventional cytology, our data show that immunostaining is clearly superior for the diagnosis of CMV pneumonitis independent of the underlying disease. This is likely due to the fact that full virus replication with a cytopathic effect is needed to show typical inclusion bodies, which corresponds to an advanced stage of CMV infection. Therefore, the sensitivity of positive inclusion bodies in regard to the diagnosis of CMV disease is low. However, an already dormant/latent virus can be cultured with the shell vial technique, explaining the low specificity of this diagnostic approach for the presence of CMV disease. The diagnostic value of quantitative polymerase chain reaction in BAL fluid for the diagnosis of CMV pneumonitis is not yet clear, despite its ability to determine the viral load. However, there is a considerable variability in fluid and cell recovery from BAL fluid, which might influence the quantitative polymerase chain reaction approach.
In conclusion, pulmonary CMV infection is frequent in all groups of immunocompromised patients except among patients following high-dose chemotherapy. Immunostaining with a monoclonal antibody against CMV in BAL fluid is a very helpful method for identifying patients with CMV pneumonitis.
| Footnotes |
|---|
Received for publication April 11, 2000. Accepted for publication September 27, 2000.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
D. Stolz, A. Stulz, B. Muller, A. Gratwohl, and M. Tamm BAL Neutrophils, Serum Procalcitonin, and C-Reactive Protein To Predict Bacterial Infection in the Immunocompromised Host Chest, August 1, 2007; 132(2): 504 - 514. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Franquet, S. Rodriguez, R. Martino, A. Gimenez, T. Salinas, and A. Hidalgo Thin-Section CT Findings in Hematopoietic Stem Cell Transplantation Recipients with Respiratory Virus Pneumonia Am. J. Roentgenol., October 1, 2006; 187(4): 1085 - 1090. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Joos and M. Tamm Breakdown of Pulmonary Host Defense in the Immunocompromised Host: Cancer Chemotherapy Proceedings of the ATS, December 1, 2005; 2(5): 445 - 448. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Tueller, P. N. Chhajed, C. Buitrago-Tellez, R. Frei, M. Frey, and M. Tamm Value of smear and PCR in bronchoalveolar lavage fluid in culture positive pulmonary tuberculosis Eur. Respir. J., November 1, 2005; 26(5): 767 - 772. [Abstract] [Full Text] [PDF] |
||||
![]() |
M W Beresford, A G Cleary, J A Sills, J Couriel, and J E Davidson Cardio-pulmonary involvement in juvenile systemic lupus erythematosus Lupus, February 1, 2005; 14(2): 152 - 158. [Abstract] [PDF] |
||||
![]() |
R. F. Chemaly, B. Yen-Lieberman, E. A. Castilla, A. Reilly, S. Arrigain, C. Farver, R. K. Avery, S. M. Gordon, and G. W. Procop Correlation between Viral Loads of Cytomegalovirus in Blood and Bronchoalveolar Lavage Specimens from Lung Transplant Recipients Determined by Histology and Immunohistochemistry J. Clin. Microbiol., May 1, 2004; 42(5): 2168 - 2172. [Abstract] [Full Text] [PDF] |
||||
![]() |
G.-C. Chang, C.-L. Wu, S.-H. Pan, T.-Y. Yang, C.-S. Chin, Y.-C. Yang, and C.-D. Chiang The Diagnosis of Pneumonia in Renal Transplant Recipients Using Invasive and Noninvasive Procedures Chest, February 1, 2004; 125(2): 541 - 547. [Abstract] [Full Text] [PDF] |
||||
![]() |
A T Jones, R M Du Bois, and A U Wells The pulmonary physician in critical care * Illustrative case 2: Interstitial lung disease Thorax, April 1, 2003; 58(4): 361 - 364. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |