Baum’s Textbook of Pulmonary Disease
7th Edition

The differential diagnosis of an immunocompromised patient with abnormal findings on the chest roentgenogram may include one or more of the following entities: (a) extension of the basic disease process, (b) opportunistic infection, (c) drug-induced lung disease, (d) alveolar hemorrhage, (e) nonspecific interstitial pneumonitis, (f) lymphocytic interstitial pneumonitis, (g) B-cell neoplasia and other cancers associated with chronic immunosuppression, (h) a new pulmonary problem unrelated to the other conditions. A combination of two or more of the possibilities listed is relatively common. Of these, the first four can cause acute or life-threatening respiratory distress, whereas the others can be subclinical and cause subacute or slowly progressive chronic illness.
Pulmonary Extension of Basic Disease Process
The basic disease processes listed in Table 55.1 are often associated with various degrees of suppression of the immune defense mechanisms. In the present context, patients with such disorders may be considered immunocompromised hosts. Clinicians who encounter pulmonary problems in such patients must ascertain whether the pulmonary manifestations represent an extension of the basic disease into the lungs or other complications. Intrathoracic spread of primary pulmonary and extrapulmonary malignancies is the most common form of pulmonary extension of the basic disease process. Lymphangitic metastases from pulmonary and nonpulmonary malignancies should be included in the differential diagnosis of lung opacities on chest imaging studies in this group of patients.
Hematologic malignancies, especially leukemias and lymphomas, also can cause chest roentgenographic abnormalities (Fig. 55.2). Leukemias as a group are associated with mediastinal and hilar lymphadenopathy in 50% of cases at autopsy and with pulmonary parenchymal involvement in 25%. Occasionally, patients may present with acute respiratory distress resulting from extensive leukemic infiltrates in the lungs. Hodgkin lymphoma causes roentgenographically detectable mediastinal lymphadenopathy in 50% of patients and pulmonary parenchymal lesions in 30%. Non-Hodgkin lymphoma causes mediastinal lymphadenopathy in 35% of cases. A primary pulmonary lymphoma may present as alveolar opacities or as a homogeneous mass. In such cases,

elaborate diagnostic investigations may be necessary to rule out an extrapulmonary focus of the lymphoma. Lung involvement is more common in untreated lymphoproliferative diseases than in myeloproliferative diseases. Opportunistic infections are the most common cause of morbidity and mortality in these patients. However, opportunistic infections of the lungs are very uncommon in patients with hematologic malignancies who have no history of chemotherapy, radiation therapy, bone marrow transplantation, or malnutrition.
FIGURE 55.2. Pulmonary extension of non-Hodgkin lymphoma refractory to chemotherapy. Diffuse alveolar opacities are seen in both lungs.
Connective tissue diseases and the vasculitides are often associated with pulmonary manifestations. Untreated rheumatologic entities, such as rheumatoid arthritis, systemic lupus erythematosus, progressive systemic sclerosis or scleroderma, polymyositis/dermatomyositis, Sjögren syndrome, mixed connective tissue disease, Wegener granulomatosis, temporal arteritis, and Churg-Strauss vasculitis, are well-known causes of pulmonary complications. The pulmonary manifestations of these disorders are discussed at length in Chapter 26 and Chapter 30. As in patients with malignancies, pulmonary opportunistic infections are uncommon in these patients unless they have received prolonged immunosuppressive therapy for their disease. Because most patients with symptomatic rheumatologic and vasculitic disorders are treated with immunosuppressive therapy, they become prime targets for opportunistic infections, drug-induced lung disease, and other pulmonary complications.
Opportunistic Pulmonary Infections
Overwhelming pneumonias remain an important cause of morbidity and mortality in immunocompromised patients. Many of these infections progress rapidly and are fatal. An immunocompromised patient is prone to the development of pulmonary infections caused by viruses, bacteria, mycobacteria, fungi, protozoa, and parasites (see Chapter 18).
Patterns of Pulmonary Injury and Common Causes
Alveolar Hemorrhage
Pulmonary alveolar hemorrhage is an important cause of respiratory symptoms and respiratory distress syndrome in immunocompromised patients (14) (see Chapter 31), and it accounts for 11% to 64% of pneumonic infiltrates in this group. However, alveolar hemorrhage is rarely the sole cause of pulmonary infiltrates, with fewer than 5% of patients exhibiting pulmonary hemorrhage as the only respiratory manifestation. Even when it is clinically considered an isolated phenomenon in immunocompromised patients, it is important to exclude occult invasive aspergillosis infection; close to 50% of patients with severe pulmonary hemorrhage may have documented aspergillosis. The association between thrombocytopenia and Aspergillus infection and pulmonary hemorrhage is significant. Alveolar hemorrhage is also associated with other complications, such as mucormycosis, pulmonary veno-occlusive disease, graft versus host disease, mitomycin therapy, and other processes. Alveolar hemorrhage is seen frequently in recipients of heart transplants. In one study, 75% of BAL samples deemed positive for alveolar hemorrhage were in recipients of heart transplants. In another study, alveolar hemorrhage was detected in 21% of 141 consecutive recipients of autologous bone marrow transplants. Alveolar hemorrhage is significantly associated with thrombocytopenia (platelet count <50,000/mm3), other coagulopathies, renal failure (serum creatinine = 2.5 mg/dL), a history of heavy smoking (>10 pack-years), leukopenia, thoracic radiation, and chemotherapy.
The difficulty of establishing the diagnosis of alveolar hemorrhage can be ascribed to the following factors: highly nonspecific clinical and roentgenographic features, absence of hemoptysis in most patients, and lack of specificity of the imaging procedures, including chest roentgenography, HRCT, and radionuclide scans. The diffusing capacity of the lung for carbon monoxide, when measured serially, has been reported to increase during alveolar bleeding as a result of an increased uptake of carbon monoxide by red blood cells in the alveoli. The need to perform this test serially renders it impractical in sick patients. Furthermore, its reliability in the diagnosis of alveolar hemorrhage has not been established.
Although thoracoscopy and open lung biopsy can document the diagnosis, these are high-risk procedures in immunocompromised patients. BAL has been used to diagnose alveolar hemorrhage in immunocompromised patients. The mere presence of hemosiderin-laden macrophages in the BAL effluent without quantification is not diagnostic. Therefore, estimation of the number of hemosiderin-laden macrophages is used to diagnose this complication. In a study of 240 BAL fluid samples in 194 immunocompromised hosts, a proportion of siderophages of at least 20% was considered to be diagnostic of alveolar hemorrhage (14). By this definition, alveolar hemorrhage was present in 87 (36%) of the samples; a proportion of siderophages of 20% to 65% was correlated with moderate hemorrhage (Golde score between 20 and 100), and a proportion higher than 67% was correlated with severe hemorrhage (Golde score >100). Even when a diagnosis of alveolar hemorrhage is established, it is essential to exclude the coexistence of the basic disease process and infections in the lungs.
Obstructive Airways Disease
Progressive airways disease leading to life-threatening respiratory distress is one of the most serious pulmonary complications encountered in immunocompromised patients. Obstructive airways disease can take the form of bronchospastic disease, lymphocytic bronchitis, or bronchiolitis obliterans.

A lymphocyte-mediated pathologic process is most likely responsible for these complications.
Obstructive airways disease secondary to immunocompromise is almost exclusively limited to recipients of organ transplants (Fig. 55.3). Patients with graft versus host disease are at risk for the development of this complication. Bronchiolitis obliterans secondary to rheumatoid arthritis and other collagen diseases is not considered here. However, it is important to recognize that in a patient with rheumatoid arthritis or another collagen disorder who is being treated with specific nonsteroidal antiinflammatory agents (e.g., penicillamine, gold preparations), the development of features of obstructive airways disease may be caused by the drugs themselves.
FIGURE 55.3. Bronchiolitis obliterans–associated progressive obstruction of airflow following bone marrow transplantation. Note fibrous thickening of the submucosa and distortion of the muscularis, with resultant compromise of the bronchiolar lumen.
Nonspecific Interstitial Pneumonitis
The histopathologic findings in the lung tissue of immunocompromised patients with a diffuse pulmonary process may not fit any specific pattern and are therefore described as nonspecific. The nonspecificity of a biopsy-based diagnosis is helpful in excluding other causes of an interstitial process (Fig. 55.4). For example, in a study of 70 immunosuppressed patients with diffuse lung disease who underwent open lung biopsy, even though the procedure was diagnostically accurate in 97%, 45% of the diagnoses were nonspecific (fibrosis); no significant difference in mortality was found between the patients with a specific diagnosis and those without, or between the patients whose biopsy diagnosis resulted in an alteration of therapy and those whose biopsy diagnosis did not. In contrast, another study noted a recovery rate of only 25% in patients without a specific diagnosis following lung biopsy, whereas in patients in whom a treatable problem had been diagnosed, the overall recovery rate was 70%.
FIGURE 55.4. Nonspecific interstitial pneumonitis in an immunocompromised patient in whom dyspnea and cough developed after thoracic radiation and multiple courses of chemotherapy for Hodgkin lymphoma. Open lung biopsy revealed nonspecific inflammation and fibrosis.
High-dose whole-body irradiation is commonly included in the conditioning regimens that precede bone marrow transplantation in patients with hematologic malignancies. Interstitial pneumonitis is a major complication after bone marrow transplantation, and nearly one fourth of all patients with bone marrow transplants die of this complication. In approximately half of these patients, an infectious agent, particularly cytomegalovirus, is involved. Additional factors, such as remission-induction chemotherapy, cyclophosphamide, methotrexate, cyclosporine, and graft versus host disease, combine to cause interstitial lung disease in these patients.

Lymphocytic Interstitial Pneumonitis
The differential diagnosis of lymphocytic interstitial pneumonitis is discussed in Chapter 24. In immunocompromised patients, the causes of lymphocytic interstitial pneumonitis include HIV infection (see Chapter 18), other viral infections, graft versus host disease, and agammaglobulinemia. Because many of the diseases associated with lymphocytic interstitial pneumonitis are forms of lymphoma, immunocompromised patients with lymphocytic interstitial pneumonitis should be closely observed to detect possible lymphoproliferative disease.
Pulmonary “Immune” Neoplasia
The frequency of non-Hodgkin lymphomas is increased in immunocompromised patients as a result of iatrogenic immunosuppressive therapy. Such lymphomas are more common in organ transplant recipients than in other immunocompromised patients. Lymphomas may be oligoclonal or polyclonal in origin (Fig. 55.5), and they may be related to the use of cyclosporine. In contrast, the non-Hodgkin lymphomas seen in patients with AIDS are Burkitt-like lymphomas, B-cell lymphomas, or B-cell immunoblastic sarcomas, with or without plasmacytoid features. They tend to be extranodal.
FIGURE 55.5. B-cell lymphoma developing in the right upper lobe of a patient on long-term immunosuppressive therapy after renal transplantation.
As a result of successful treatment with radiation, chemotherapy, or both, patients with lymphoma are living longer. The long-term follow-up of these patients has revealed an increased incidence of lung carcinoma. The relative risk for the development of new malignancies in patients treated for Hodgkin or non-Hodgkin lymphoma is two to three times that in the normal population. Previous radiation therapy also may increase the risk for the development of pulmonary nonlymphomatous malignancies.
Small cell carcinoma is the predominant histologic type of lung cancer in both smokers and nonsmokers treated with irradiation. However, patients with Hodgkin lymphoma who receive supradiaphragmatic irradiation or combined-modality therapy may be at increased risk for the development of non–small cell carcinoma of the lung. In a study of such patients, the risk ratio for the development of lung cancer among patients with Hodgkin lymphoma was 5.6 times that expected in the general population. The median ages of patients in whom Hodgkin lymphoma and lung carcinoma were diagnosed were 39 and 45 years, respectively. The interval between the diagnosis of Hodgkin lymphoma and metachronous lung cancer averaged 7 years but appeared to vary inversely with age at diagnosis, a finding that underscores the need for long-term close observation.
Pulmonary Problems Unrelated to Immunodeficiency
Immunocompromised patients are more susceptible to the disease processes that affect nonimmunocompromised subjects. Abnormal findings on a chest roentgenogram in an immunocompromised patient may represent cardiac or noncardiac pulmonary edema, pulmonary embolism, a community-acquired pulmonary infection, aspiration pneumonitis, or delayed effects of thoracic irradiation. More than one third of immunocompromised patients have a combination of two or more of these complications.