|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From Memorial Sloan-Kettering Cancer Center, New York, NY.
Correspondence to: Diane E. Stover, MD, FCCP, Memorial Sloan-Kettering Cancer Center, 1275 York Ave, New York, NY 10021; e-mail: stoverd{at}mskcc.org
| Abstract |
|---|
|
|
|---|
Key Words: bronchiolitis obliterans organizing pneumonia cryptogenic organizing pneumonia macrolides treatment
| Introduction |
|---|
|
|
|---|
| Case 1 |
|---|
|
|
|---|
|
|
| Case 2 |
|---|
|
|
|---|
Physical examination showed oxygen saturation at rest and on room air of 97% and rhonchorous breath sounds at the right base. A CT scan of the chest showed multifocal areas of alveolar infiltrates with an area of consolidation at the right base. The patient was thought to have a community-acquired pneumonia, and he was placed on clarithromycin (500 mg po bid) for 10 days. He improved symptomatically, but a repeat chest radiograph revealed persistent pulmonary abnormalities. Sputum cultures collected before initiation of antibiotics did not reveal any pathologic organisms. Transbronchial biopsy showed chronic inflammation, interstitial fibrosis, and intra-alveolar organizing fibromyxoid polyps that were compatible with organizing pneumonia. There was no evidence of any pathologic infectious agents or malignancy. After 4 weeks of clarithromycin treatment, he became asymptomatic and remained on the antibiotic for 3 months (dosage was decreased to 250 mg bid and then qd). His chest radiograph showed complete resolution, and he remained without recurrence 3 years after presentation.
| Case 3 |
|---|
|
|
|---|
Physical examination was normal. Oxygen saturation at rest and on room air was 98%. Laboratory data were remarkable for a sedimentation rate of 49 mm/h (normal, 0 to 20 mm/h) and normal pulmonary function tests. Prior to any antibiotic therapy, the patient underwent bronchoscopy with transbronchial biopsy that revealed benign lung tissue and no organisms. The patient completed a 3-week course of clarithromycin with clinical and radiographic improvement. Clarithromycin was discontinued; approximately 3 weeks later, malaise, low-grade fevers, and a dry cough developed. Repeat CT scan showed new consolidation of the right upper lobe and new nodular opacities bilaterally. The patient once again was placed on clarithromycin with improvement clinically and radiographically but relapsed after a 3-week course was discontinued.
Video-assisted thoracoscopic wedge biopsy of the right upper and lower lobes revealed findings consistent with organizing pneumonia. All smears and culture findings were unremarkable without antibiotics for at least 1 month. The patient was placed on oral steroids for 6 months and had complete resolution of symptoms and radiographic findings. She remains well 4 years after the diagnosis of BOOP.
| Cases 4, 5, AND6 |
|---|
|
|
|---|
Case 4
A 56-year-old woman presented with a nonresolving right-sided infiltrate after radiation to right breast cancer. The patient had a diagnosis of organizing pneumonia on open-lung biopsy that initially improved with corticosteroid therapy. She had clinical relapse with cough and radiographic recurrence when the drug was discontinued. Because the patient had severe side effects from corticosteroids, she was successfully treated with clarithromycin (250 mg bid for 2 months and then decreased to 250 mg qd for 1 month). The antibiotic was discontinued after 3 months because of a skin rash, without clinical or radiographic relapse.
Case 5
A 62-year-old woman had a cough, low-grade fever, and bilateral patchy infiltrates with air bronchograms 8 months after receiving radiation to the cervical spine because of impending spinal cord compression due to metastasis from breast cancer. The patient also received paclitaxel. The physical examination was unremarkable, and the oxygen saturation at rest and on room air was 98% and with exercise was 96%. Although radiation-related BOOP was suspected, transthoracic needle biopsy was done to rule out metastatic lung disease and infection. The cell block was consistent with organizing pneumonia. All culture findings for bacteria, mycobacteria, and fungi were unremarkable. Paclitaxel-induced lung injury was also considered but thought to be unlikely due to the pathology findings. Hence, the drug was continued and the patient had a complete resolution of symptoms and radiographic findings after 7 months of corticosteroid therapy. However, after a very slow steroid taper, the patient relapsed with clinical symptoms and radiographic infiltrates similar to the original findings. Because avascular necrosis of the right hip had developed with corticosteroid therapy, and because the symptoms were minimal, the patient was placed on clarithromycin, 250 mg bid. Four weeks after starting the antibiotic, the cough resolved and a repeat chest CT scan showed significant improvement in the infiltrates. She received clarithromycin for a total of 6 months.
Case 6
A 67-year-old man with a diagnosis of large cell lymphoma underwent radiation to the cervical and thoracic spine because of metastatic disease in that area. Approximately 2 weeks after the completion of the radiation, the patient had a very bad cough. At about the same time, an abscess on the right big toe developed and was treated with a 2-week course of levofloxacin. The cough did not respond to the drug, and it lasted for 6 weeks, after it was discontinued but then the cough resolved. Four months after completing radiation, the patient underwent an extent-of-disease evaluation to determine the need for further chemotherapy. His last chemotherapy was 5 months prior, with ifosfamide, carboplatin, and etoposide.
Chest CT scan revealed a new right lower lobe paramediastinal nodular infiltrate with multiple air bronchograms and a right middle lobe ground-glass opacity with irregular borders. The latter was outside the radiation portal. Physical examination at that time was unremarkable, including normal pulmonary function test results, except for a decrease in oxygen saturation from 97% at rest to 93% with exercise.
Radiation-related BOOP was considered likely, but because of the concern for recurrence of lymphoma the patient underwent needle biopsy of the right lower lobe nodular infiltrate. Histopathology showed inflammatory cells with no evidence of malignancy, and all smear and culture findings were negative for organisms. The patient had not received antibiotics for approximately 4 months.
Conventional radiation pneumonitis was considered unlikely because the patient never experienced dyspnea, some of the infiltrates were outside the radiation portal, and pulmonary function tests did not show a decrease in diffusing capacity. Recurrent lymphoma and infection were eliminated based on the pathologic findings and a clinical diagnosis of radiation-related BOOP was made.
The patient was treated with clarithromycin, and he had an excellent response with improvement in the right lower lobe and middle lobe infiltrates. However, due to a misunderstanding, the patient abruptly discontinued the clarithromycin with recurrence of the infiltrates, low-grade fever, and significant pleuritic chest pain. Because of the debilitating symptoms, he was started on a course of prednisone once again with clinical and radiographic improvement.
| Discussion |
|---|
|
|
|---|
In these six patients, the diagnosis of organizing pneumonia is presented. Three cases clinically were idiopathic and considered COP, and three cases were thought to be related to radiation. In all cases, there was no evidence of an infectious or malignant etiology. Although in some, lung injury from chemotherapeutic agents was a possibility, the lack of dyspnea and diffuse interstitial infiltrates mitigated against the diagnosis of conventional chemotherapy -induced lung injury. When BOOP was associated with radiation, conventional radiation pneumonitis was unlikely because dyspnea, the hallmark of radiation pneumonitis, was absent and in all cases infiltrates occurred outside the irradiated portal, which is unusual for radiation pneumonitis.
When tissue documentation was obtained, organizing pneumonia was the predominate pathology. No pathologic organisms were isolated in any of the patients, and all patients responded to macrolide therapy at some point during the course of the illness. Although some patients with organizing pneumonia can be asymptomatic and/or improve without treatment, it would be quite unusual for all of our patients to improve spontaneously. In our experience and that of others,3456 the majority of patients with diffuse infiltrates due to organizing pneumonia will need treatment. Only 4% of patients with this syndrome will undergo spontaneous remission.
The use of macrolides to treat inflammatory diseases of the lungs is not a new concept. One of the first hints that macrolide antibiotics had anti-inflammatory activity associated with lung disease was in 1970, when Itkin et al7 reported their experience in the treatment of asthmatic patients with erythromycin and troleandomycin. They found that adding a macrolide to the drug regimen of steroid-dependent asthmatics enabled steroid taper without deleterious effects. Studies8910 from Japan have shown that short-term therapy with oral erythromycin or roxithromycin was associated with significant improvement of bronchial hyperactivity in asthmatics. More recently, long-term clarithromycin was found to have a steroid sparing effect in elderly patients with prednisone-dependent asthma.11 Initially, macrolides were thought to inhibit theophylline clearance and/or increase the concentration of steroids to explain their beneficial action in asthmatic patients. However, in vivo and in vitro studies89101112131415161718192021222324252627 support the fact that mechanisms other than drug interaction may explain the beneficial effects of macrolides in the treatment of asthma and other inflammatory conditions.
Much of the evidence that macrolides have immunomodulatory effects comes from Japan, where these antibiotics are effective in treating diffuse panbronchiolitis (DPB).121314151617 DPB is a disease of adults that is prevalent in Japan, Korea. and China and is characterized by infiltration of respiratory bronchioles with inflammatory cells. Ever since erythromycin has been used in the treatment of DPB, the 10-year survival rate has increased from 55 to 94%.16 Although the pharmacologic activity of erythromycin is unknown in DPB, data support the hypothesis that its efficacy depends on mechanisms other than bactericidal activity. Polymorphonuclear cells are believed to play a crucial role in the deterioration of patients with DPB, and macrolides not only reduce the influx of polymorphonuclear cells into the lung but also decrease the production of cytokines involved in the inflammatory cascade, such as interleukin (IL)-8 and IL-1ß.17
Macrolides, as anti-inflammatory agents, may also have a role in the treatment of bronchiectasis. A double-blind, placebo-controlled trial18 of low-dose, long-term erythromycin (500 mg bid for 8 weeks) showing improvement in patients with bronchiectasis has been published. In 11 patients who received erythromycin, FEV1 and FVC increased while sputum volume over 24 h decreased significantly compared with those receiving placebo. Because erythromycin is unlikely to be bactericidal with these low doses, the authors18 postulated that its inhibitory effect on stimulated neutrophils and on IL-8 production might be possible mechanisms to explain these findings.
Other in vivo and in vitro anti-inflammatory mechanisms for the macrolides include the following: their ability to inhibit neutrophil oxidation bursts, to decrease elastase activity, to suppress granulocyte macrophage-colony simulating factor, and to reduce or block production of both tumor necrosis factor-
and certain adhesion molecules such as soluble intercellular adhesion molecule-1.1920212223242526 The last two are necessary for neutrophil migration.
Clinical and pathologic similarities between DPB and cystic fibrosis (CF) have lead to the investigation of macrolides in the treatment of CF. Wolter et al27 conducted a 3-month, prospective, placebo-controlled, double-blind trial of azithromycin, 250 mg/d, in 60 adults with CF. The treatment group had statistically significant improvements in FEV1 percentage of predicted and FVC compared to the placebo-treated group. Furthermore, the azithromycin group required fewer days of therapy with IV antibiotics and had greater improvements in dyspnea, fatigue, emotional, and mastery domains on the chronic respiratory disease questionnaire compared with those in the placebo study arm.
In vitro studies show that short-term administration of macrolides result in enhancement of the immune response benefiting the treatment of infections, while long-term administration results in immunosuppression possibly helping patients with inflammatory conditions like DPB, CF, and organizing pneumonia.28 This may explain why many of our patients (cases 1,2, 3, and 6) worsened or relapsed when macrolide therapy was discontinued after 10 to 21 days of initial improvement and why the CF patients of Wolter et al27 improved with prolonged treatment.
The anti-inflammatory effects of macrolides appear to be dependent not only on duration of therapy but on their structure. The 14- and 15-membered ring structures, such as erythromycin, clarithromycin, and azithromycin, are believed to be the most effective anti-inflammatory macrolides.19
There are some BAL data29 on patients with organizing pneumonia that show a mixed WBC pattern with increased lymphocytes (20 to 40%), neutrophils (up to 10%), and eosinophils (up to 5%). The lymphocyte pattern shows an increase in total lymphocyte count and a decrease in CD4/CD8 ratio due to an increase in cytotoxic T-cells.30 Aoki and Kao26 demonstrated that erythromycin may exert anti-inflammatory effects on T-cells by inhibiting cytokine gene expression at the level of transcription activation. Keeping in mind the results of the BAL findings, the beneficial effects of macrolides in patients with COP or BOOP related to other disorders may not only be due to their immunosuppressive effect on polymorphonuclear cells and their products but also to their influence on T-cells.
Since BAL findings suggest that organizing pneumonia is an inflammatory disease within the distal airways and the alveoli, it is not surprising that macrolides as anti-inflammatory agents may be therapeutic. There is some scanty clinical evidence that suggests this is the case. Epler et al,2 Arbetter et al,4 and Ichikawa et al31 reported small numbers of patients with this syndrome who improved with macrolide therapy, but details of these cases are not given. We also have several presumptive cases of radiation-related BOOP (not reported here) that have responded to clarithromycin. Because of the waxing-and-waning nature of symptoms, signs, and radiograph findings, the improvement with treatments with macrolides of their patients and some of our patients may have been coincidental. This seems unlikely in most of our cases, since there was a direct correlation between administration of the drug and improvement and, in some cases, relapse when it was stopped. As with other studies345 involving treatment with low dose macrolides, we encountered no major complications. A rash developed in one patient (case 4), which subsided after clarithromycin was discontinued.5
Based on our experience, circumstances under which macrolide therapy for organizing pneumonia can be considered include the following: (1) in patients with minimal symptoms and/or minimal physiologic impairment; (2) as adjuvant therapy in patients receiving steroids; steroids can be tapered while initiating and continuing macrolides; and (3) in patients who cannot tolerate steroids. Just as with corticosteroids, macrolides should be continued for a prolonged period of time (3 to 6 months or longer) and tapered to prevent relapse.
Limitations
A limitation of this study is that all patients did not have a definitive diagnosis of organizing pneumonia. When feasible, tissue documentation is optimal, since other more serious disorders can present similar to COP/BOOP in the immunocompromised host.
| Conclusion |
|---|
|
|
|---|
| Acknowledgements |
|---|
| Footnotes |
|---|
Received for publication September 8, 2003. Accepted for publication June 27, 2005.
| References |
|---|
|
|
|---|
, and interleukin 8 in bronchoalveolar lavage fluid of patients with diffuse panbronchiolitis: a potential mechanism of macrolide therapy. Respiration 1996;63,42-48[ISI][Medline]
B, but not NFAT, through calcineurin-independent signaling in T-cells. Antimicrob Agents Chemother 1999;43,2678-2684This article has been cited by other articles:
![]() |
A. Knyazhitskiy, R. G. Masson, R. Corkey, and J. Joiner Beneficial Response to Macrolide Antibiotic in a Patient With Desquamative Interstitial Pneumonia Refractory to Corticosteroid Therapy Chest, July 1, 2008; 134(1): 185 - 187. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |