| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Division of Pulmonary Sciences and Critical Care Medicine (Drs. Olson and Schwarz), University of Colorado Health Sciences Center, and Interstitial Lung Disease Program (Drs. Cosgrove and Brown), National Jewish Medical and Research Center, Denver, CO.
Correspondence to: Amy L. Olson, MD, Division of Pulmonary Sciences and Critical Care Medicine, University of Colorado Health Sciences Center, 4200 East Ninth Ave, C272, Denver, CO 80262; e-mail: amy.olson{at}uchsc.edu
Key Words: chronic eosinophilic pneumonia • eosinophilic pleural effusion • eosinophilic pneumonia • pleural effusion • rheumatoid arthritis
Case Report
A 63-year-old man was referred for evaluation of a nonresolving pneumonia. Six weeks prior to referral, he was seen by his internist with complaints of a nonproductive cough, shortness of breath, and subjective fevers and chills. He had lost 20 lb since the onset of symptoms, and had increasing stiffness and pain in both hands over this same time period. A chest radiograph revealed a right, peripheral lower lobe infiltrate, and he was treated with a 10-day course of gatifloxacin. His symptoms persisted despite treatment. Ceftriaxone was added, and he was continued on gatifloxacin for an additional 14 days. Two weeks prior to referral, symptoms continued and night sweats and right-sided pleuritic chest pain developed. A repeat chest radiograph revealed progression of the right lower lobe infiltrate and a small right-sided pleural effusion. He was placed on clindamycin and prednisone, 20 mg bid, and referred for additional evaluation.
The patient’s medical history was notable for hypertension and well-controlled diabetes mellitus. Medications included enalapril, hydrochlorothiazide, and glipizide. He had a 40-pack-year smoking history, although he quit 8 years ago.
On presentation, the patient had a nonproductive cough. Temperature was 37.2°C, and respiratory rate was 18 breaths/min with normal oxygen saturation. Inspiratory crackles were noted in the right mid-lung field, with dullness to percussion at the right lung base. His joints were without deformities or synovitis. Treatment with steroids was discontinued.
Notable laboratory results included: peripheral WBC count, 12.9/µL (neutrophils, 76%; lymphocytes, 14%; monocytes, 9%; eosinophils, 1%); hematocrit, 33%; platelets, 699/µL; creatinine, 1.0 mg/dL; total protein, 6.4 g/dL; and albumin, 2.6 g/dL. Urinalysis was normal. A rheumatologic evaluation yielded a Westergren sedimentation rate of 120 mm/h; C-reactive protein of 106 mg/L (normal range, 0 to 8 mg/L); negative anti-neutrophil cytoplasmic antibody panel; positive antinuclear antibody (titer of 1:320 with a speckled pattern); negative anti-double strand DNA; positive serum rheumatoid factor with a titer of 320 IU/mL (normal range, 0 to 39.9 IU/mL); and positive anti-cyclic citrullinated peptide (titer > 200 U). Routine sputum Gram stain and culture results were negative. Several days after corticosteroids were discontinued, significant morning stiffness of the hands, wrists, and elbows developed, and swelling and tenderness of the wrists and metacarpal-phalangeal joints occurred. The peripheral eosinophil count had increased to 1,090/µL (total peripheral WBC count, 12.2/µL; eosinophils, 9%).
Chest radiography revealed consolidation of the right middle and lower lobes, with a wedge-like distribution in the lateral aspect of the middle lobe (Fig 1 ). CT angiogram of the chest revealed no pulmonary emboli, and demonstrated consolidation with air bronchograms in the right middle and lower lobes and a moderate right pleural effusion (Fig 2 ). Bilateral hand and wrist radiographs revealed periarticular osteopenia and subtle, marginal erosions at the distal-interphalangeal joints suggestive of an inflammatory arthritis.
|
|
Discussion of the Case
What Is the Cause of This Patient’s Inflammatory Arthritis?
Although this patient initially presented with pulmonary symptoms, further examination (after corticosteroids were discontinued) revealed an acute, symmetric inflammatory process that involved his hands, wrists, and elbows. After further discussion of these symptoms with the patient, he stated that he had experienced mild morning, stiffness, and pain in both hands several months prior to his presentation. In addition to clinical symptoms, laboratory data revealed a high-titer rheumatoid factor and a high-titer anti-cyclic citrullinated peptide, and hand and wrist radiographs that were consistent with an inflammatory arthritis. This constellation of findings established the diagnosis of rheumatoid arthritis.1
In a Patient With Rheumatoid Arthritis and a Unilateral Middle- And Lower-Lobe Infiltrate With an Associated Pleural Effusion, What Pulmonary Diseases Are Included in the Differential Diagnosis?
Rheumatoid arthritis is a chronic, systemic, inflammatory disease that typically affects the diarthodial joints. Extra-articular disease is common and may be present in as many as 75% of patients with severe rheumatoid arthritis.2 A number of pleuropulmonary manifestations have been associated with this disease, including interstitial lung disease (usual interstitial pneumonia, nonspecific interstitial pneumonia, lymphocytic interstitial pneumonia, organizing pneumonia, and chronic eosinophilic pneumonia), airway disease (bronchiectasis, bronchiolitis obliterans, follicular bronchiolitis, and diffuse panbronchiolitis), rheumatoid nodules, pulmonary vascular disease, and pleural disease (pleuritis), as well as drug-induced lung disease.2 One study3 of patients with rheumatoid arthritis and joint disease of < 2 years in duration reported that 58% of patients had findings consistent with interstitial lung disease, and only male gender was statistically associated with clinically significant disease.
While there are a number of pleuropulmonary manifestations that may occur in rheumatoid arthritis, the consolidation of the right middle and lower lobe seen on CT in this patient narrows the diagnostic possibilities. Organizing pneumonia is one such potential cause and is characterized histologically by the pattern of excessive proliferation of granulation tissue within the alveolar ducts and spaces, and lymphocytic inflammation in the surrounding alveoli and interstitium. The pathologic process may involve the bronchioles, and as a result this disorder was previously referred to as bronchiolitis obliterans organizing pneumonia.4 In order to avoid confusion with other primary disorders of the airway, the term organizing pneumonia is now preferred. Organizing pneumonia may be either idiopathic (in which case the preferred clinical term is cryptogenic organizing pneumonia) or the result of a number of conditions including infections, connective tissue diseases (including rheumatoid arthritis), bone marrow or heart-lung transplantations, malignancies (both solid tumor and hematologic), radiation, or drugs. When organizing pneumonia is associated with an underlying medical condition, the preferred clinical term is secondary organizing pneumonia. CT typically demonstrates bilateral asymmetric alveolar infiltrates with either ground-glass attenuation or consolidation, and with a tendency for involvement of the lower lobes and periphery.4 Although pleural effusions are generally not seen in cryptogenic organizing pneumonia, pleural effusions have been reported to coexist in secondary organizing pneumonia with concurrent connective tissue diseases such as rheumatoid arthritis.5 The unilateral nature of this patient’s pulmonary process, although not unheard of, would argue against a diagnosis of secondary organizing pneumonia.6
Infection remains another diagnostic possibility, and this patient’s radiographic findings are consistent with a community-acquired pneumonia. However, he did not respond to appropriate treatments, making this diagnosis less likely. As he was not receiving chronic immunosuppressive treatment, the diagnosis of an opportunistic infection is likely. An endemic fungal infection, such histoplasmosis, coccidioidomycosis, blastomycosis, and cryptococcosis, can present like a community-acquired pneumonia; however, this patient had not been in an endemic area, and fungal stains and culture results were negative.
One infectious diagnostic possibility in an immunocompetent host with a history of tobacco abuse is actinomycosis. Actinomycosis is an uncommon infection, with an estimated incidence of 1 in 300,000 people per year, and the pulmonary form of disease represents approximately 15% of all infections.7 Infections are reported three times more frequently in male patients than in female patients; in addition to tobacco abuse, alcoholism and poor dentition appear to increase the risk of disease. The pneumonia is often indolent, and symptoms typically include fever, cough with sputum production, weight loss, and chest pain. On chest radiographs and CT, the pulmonary consolidation often is associated pleural involvement, and the pulmonary process may involve the adjacent ribs and extend through the chest wall. Diagnosis of pulmonary actinomycosis is difficult, as the organism is typically not recovered from sputum or bronchoscopy aspirates cultured in aerobic environments. Anaerobic cultures revealing Gram-positive filamentous rods and tissue from biopsies demonstrating characteristic sulfur granules lead to the correct diagnosis. While actinomycosis is a diagnostic possibility in this patient, an association with the onset of rheumatoid arthritis has not been reported.
Tuberculosis is another infectious agent that must be considered in the differential diagnosis, as epidemiologic studies89 have found an elevated risk of tuberculosis in patients with rheumatoid arthritis without a prior history of anti-tumor necrosis factor therapy. Lower lung field tuberculosis has been defined as infection below the level of the hila.10 It may result from primary, reactivated, or endobronchial disease, and is uncommon. The most frequently reported radiographic abnormality is consolidation.11 Although this patient’s radiographic findings could be consistent with lower lung field tuberculosis, BAL fluid culture results were negative for acid-fast bacilli. Lung biopsies are rarely needed to make the diagnosis.11
Primary pulmonary lymphoma complicating rheumatoid arthritis is another diagnostic possibility, as this patient presented with B-symptoms including significant weight loss and a 6-week history of night sweats. Low-grade pulmonary B-cell lymphoma arising from mucosa-associated lymphoid tissue is the most common of the primary pulmonary lymphomas, accounting for as many as 87% of cases.12 Chest radiography typically reveals a solitary, well-delineated soft-tissue mass, although extensive lobar infiltrates with air bronchograms have been described.13 Pleural effusions are rare and occur in < 10% of patients. Almost half of all patients with low-grade pulmonary B-cell lymphoma are asymptomatic; constitutional symptoms may imply extrathoracic involvement. While BAL fluid with a lymphocytic alveolitis may suggest a primary pulmonary lymphoma, a diagnosis is based on histologic examination of lung tissue. In this patient with a pleural effusion, no evidence of extrathoracic involvement, and constitutional symptoms, a diagnosis of primary pulmonary lymphoma is unlikely.
An interesting feature of this case is the increase in the peripheral eosinophil count after empiric steroid therapy was discontinued. Chronic eosinophilic pneumonia, while certainly not a common manifestation of rheumatoid arthritis, has been reported in case series.141516171819 In idiopathic chronic eosinophilic pneumonia, infiltrates are almost always bilateral and asymptomatic small pleural effusions have been reported to occur in < 10% of cases.20 In chronic eosinophilic pneumonia associated with rheumatoid arthritis, pleural effusions have not been reported.
Do the Characteristics of the Pleural Fluid Help To Narrow the Differential Diagnosis?
Symptomatic rheumatoid plural effusions develop in as many as 5% of patients with rheumatoid arthritis, although asymptomatic effusions are likely to be much more common. These effusions occur more often in middle-age men with rheumatoid nodules or with high titer rheumatoid factor, and in most cases arise after the diagnosis of rheumatoid arthritis. Examination of the pleural fluid reveals a sterile exudate (by criteria of Light et al21) with a characteristically low glucose level (often < 60 mg/dL and reportedly as low as 10 mg/dL). Additional findings generally include a pleural fluid acidosis (pH < 7.3), elevated lactate dehydrogenase levels (which may exceed 700 IU/L), high total protein levels, and low complement levels. Rheumatoid factor titers in the pleural fluid may be higher than serum levels.22 Rheumatoid sterile empyematous pleural effusions have been rarely reported. Like rheumatoid plural effusions, empyematous effusions are sterile with low pleural glucose levels. However, empyematous effusions appear purulent with cellular debris, and have a very low pH (< 7.2) and a high WBC count (> 50,000/µL). These effusions may result in significant pleural thickening and lead to symptomatic lung restriction.23 Our patient did not have the typical pleural fluid findings of a rheumatoid plural effusion or rheumatoid empyematous effusion. However, he did have a significant eosinophilia on the pleural fluid cell count differential.
Eosinophilic pleural effusions account for as many as 16% of exudative pleural effusions and result from a variety of causes.24 While the majority eosinophilic effusions are idiopathic, additional secondary causes of eosinophilic effusions include blood or air in the pleural space, infection (this typically occurs late in the course of parapneumonic effusions), malignancy, pulmonary embolism, drug reactions, benign asbestos pleural effusions, uremic pleuritis, hypereosinophilic syndrome, acute and chronic eosinophilic pneumonia, and autoimmune disorders (including rheumatoid arthritis).24 While examination of the pleural fluid did exclude a purulent process either from an infection or a rheumatoid empyematous pleural effusion and the need for immediate drainage,25 the diagnosis of an eosinophilic pleural effusion left a number of possible underlying disorders as the potential cause.
Does the Differential Cell Count on the Bronchoalveolar Fluid Cell Provide Additional Insight Into This Patient’s Diagnosis?
Both rheumatoid arthritis and eosinophilic pleural effusions are rarely associated with chronic eosinophilic pneumonia. In order to clarify his underlying pulmonary process, the patient underwent bronchoscopy with BAL. In patients with chronic eosinophilic pneumonia, the cell differential on the lavage fluid from the affected segment characteristically reveals approximately 10 to 90% eosinophils.20 In this case, the cell differential on the lavage fluid revealed 95% neutrophils and no eosinophils. Neutrophil predominant lavage fluid is seen with a number of pulmonary diseases including interstitial lung diseases (idiopathic pulmonary fibrosis, desquamative interstitial pneumonia, and acute interstitial pneumonia), connective tissue diseases, ARDS, bacterial pneumonia, bronchiolitis, and asbestosis.26 At this point, a unifying diagnosis is still not evident.
At This Point, What Would You Do Next To Diagnosis This Pulmonary Process?
As lavage culture results were negative and the cell differential was nondiagnostic and the eosinophilic pleural effusion did not narrow the diagnostic possibilities, the patient was referred for a video-assisted thorascopic lung biopsy to definitively diagnosis his pulmonary process. Lung tissue revealed numerous eosinophils within the alveolar spaces and interstitium, and small foci of organizing pneumonia (Fig 3
), findings consistent with chronic eosinophilic pneumonia.27
|
Idiopathic chronic eosinophilic pneumonia is a disease in which the pulmonary parenchyma is infiltrated by eosinophils, and patients have a dramatic response to corticosteroid therapy.28 The onset of this syndrome is typically insidious, and common symptoms include cough, fever, dyspnea, and weight loss.29 Peripheral blood eosinophilia occurs in most patients, and chest radiographs frequently reveal peripheral infiltrates. These infiltrates may be extensive and bilateral and resemble the photographic negative image of pulmonary edema.2 As many as 80% of patients have bilateral infiltrates by chest radiography, whereas high-resolution CT identifies bilateral infiltrates in almost all patients.28 In this case, both the chest radiograph and CT demonstrated the peripheral nature of the pulmonary process although it was unilateral. Pleural effusions associated with idiopathic chronic eosinophilic pneumonia are unusual. In one multicenter study20 of patients with chronic eosinophilic pneumonia, pleural effusions were present in only 6.5% of cases and were characteristically small, unilateral, and asymptomatic. However, clinically relevant effusions have been reported.30
Marchand and colleagues20 reported on 41 patients with chronic eosinophilic pneumonia who underwent BAL and found BAL eosinophilia in all cases (range, 12 to 95%). Two possible explanations exist for the neutrophil-predominant alveolar lavage seen in our patient: treatment with corticosteroids (as occurred in our patient prior to referral) has been shown to markedly reduce lavage eosinophilia in patients with chronic eosinophilic pneumonia,31 and it has been our personal experience that degranulated eosinophils may appear to resemble neutrophils on Wright stain. Although not done in this case, Giemsa stain may be more sensitive to the identification of eosinophils.28
In 1980, Payne and Connellan14 first described a relationship between rheumatoid arthritis and chronic eosinophilic pneumonia. Their patient, who had long-standing rheumatoid arthritis, had chronic eosinophilic pneumonia in association with an exacerbation of rheumatoid arthritis. Subsequent similar cases were reported,1516 but some thought this was likely coincidental.17 More recent case reports1819 have described the concurrent development of rheumatoid arthritis and chronic eosinophilic pneumonia, and continue to raise the possibility that chronic eosinophilic pneumonia may be a consequence of a rheumatoid arthritis-induced systemic, inflammatory response. Activated eosinophils may be the pathophysiologic mechanism whereby the two diseases are related. Studies of patients with active rheumatoid arthritis have demonstrated that eosinophils are activated (or at least primed) and are involved in the inflammatory process,32 and activated eosinophils have been demonstrated in the BAL fluid of patients with chronic eosinophilic pneumonia.2 With the growing evidence in the literature that rheumatoid arthritis and chronic eosinophilic pneumonia are associated, we believe that the term secondary chronic eosinophilic pneumonia should be used to describe this pulmonary process when it is associated with rheumatoid arthritis.
This may be the first case presenting simultaneously with rheumatoid arthritis, chronic eosinophilic pneumonia, and an eosinophilic pleural effusion. The mechanism of pleural effusion remains unknown but may be the result of eosinophil-induced tissue damage. Although a causal relationship remains undefined, secondary chronic eosinophilic pneumonia should be considered in the differential diagnosis of patients with symptoms of rheumatoid arthritis and an infiltrative pulmonary process with or without a pleural effusion.
Clinical Follow-up
After the diagnosis of chronic eosinophilic pneumonia and rheumatoid arthritis was made, the patient was initiated on prednisone (60 mg/d po). After 5 days, there was notable clearing of the chest radiograph (Fig 4 ). After 2 months of steroid therapy, the patient had no pulmonary symptoms and his rheumatoid arthritis was well controlled. The plan for his ongoing treatment is to continue 6 months of steroid therapy. If this is successful, he will then be transitioned to a more traditional form of rheumatoid arthritis therapy.
|
Footnotes
The authors have no conflicts of interest to disclose.
Received for publication July 27, 2006. Accepted for publication August 12, 2006.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
