Nonresolving pneumonia is defined as a slow or delayed resolution of clinical symptoms and pulmonary infiltrates despite an adequate course of antibiotic therapy for 10 to 14 days. Patients with nonresolving pneumonia are at increased risk of morbidity and mortality. Several types of infectious as well as noninfectious conditions can be responsible for nonresolving pulmonary infiltrates. This case report highlights noninfectious causes that mimic infectious pneumonia, including adenocarcinoma of the lung.

Case Presentation
A 43-year-old woman presented to the emergency department with a 4-week history of productive cough and dyspnea. One month earlier, her primary care physician had prescribed azithromycin (Zithromax) for presumed acute bronchitis, but her symptoms had not improved after 5 days of therapy. Her physician then ordered a chest radiograph, which revealed a left lower-lobe infiltrate consistent with pneumonia. He prescribed a 7-day course of levofloxacin (Levaquin). Her worsening shortness of breath, despite antibiotic and beta-agonist inhaler therapy, prompted her to come to the emergency department.

She denied having fever, chills, night sweats, contact with sick persons, recent travel, or a history of tuberculosis. Her medical history was significant for hypertension, 15 years or more of cigarette smoking, and occasional alcohol use.

Physical examination showed: temperature, 97.5ºF; pulse, 91 beats/min; respiratory rate, 20 breaths/min; blood pressure, 114/82 mm Hg. The patient appeared to be comfortable, with no respiratory distress. Decreased breath sounds were noted in the lower left lung field. The rest of the examination was unremarkable. Laboratory results were: white blood cell count, 17.1 x 10⁹/L, with 74% neutrophils and 12% eosinophils; D-dimer, 5.56 mg/L (normal, <0.43 mg/L).

Chest radiographs demonstrated a left lower-lobe infiltrate with left effusion (Figure 1). A computed tomography (CT) angiogram of the chest did not reveal pulmonary emboli. Diagnostic thoracentesis was performed, and 250 cc of fluid was drained from the left pleural space. Pleural fluid cytology showed poorly differentiated adenocarcinoma, which is characterized by nests of atypical cells with enlarged nuclei, prominent nucleoli, and eosinophilic cytoplasm (Figure 2). The diagnosis was further confirmed by transbronchial washings of the left lower lobe.

The patient was discharged from the hospital after 2 days. She was scheduled to follow up with an oncologist for consideration of chemotherapy for stage IIIB adenocarcinoma of the lung.

Discussion
Nonresolving pneumonia is defined as a slow or delayed resolution of clinical symptoms and pulmonary infiltrates despite an adequate course of antibiotic therapy for 10 to 14 days. Treatment failure may be the result of inappropriate initial therapy, the presence of unusual organisms, or noninfectious causes of pneumonia that mimic infectious pneumonia.¹

The “typical” resolution of infectious pneumonia is variable and depends on the causative agent and host response to the invading pathogen (Table 1). In as many as half of all cases of pneumonia, the pathogen remains unidentified.²

Reports vary regarding the expected time for complete radiographic resolution. One study of 81 immunocompetent patients with community-acquired pneumonia (CAP) showed that complete clearing of the disease occurred at 2 weeks in half of the patients and at 4 weeks in two thirds of patients.³ A more recent study of 64 elderly patients with CAP showed a radiographic resolution rate of 35% within 3 weeks, 60% within 6 weeks, and 84% within 12 weeks.⁴

Infectious causes
Common bacterial causes of CAP include Streptococcus pneumoniae, enteric gram-negative bacilli, Legionella pneumophila, Hemophilus influenzae, and Staphylococcus aureus. Other infectious etiologies to consider in the differential include tuberculosis, endemic fungal infections, and viral pneumonias. In one large, population-based surveillance study, no etiology could be identified in 60% of patients, despite extensive diagnostic testing.⁵ Similarly, a study of 132 patients evaluated by bronchoscopy within the first 24 hours of establishing a diagnosis could not find etiologic pathogens in 51% of cases using bronchoalveolar lavage fluid culture.² Most patients in this study, however, were already being treated empirically with antibiotics before the procedure, potentially reducing the diagnostic yield.

Difficulty performing sputum Gram’s stains and the insensitivity of noninvasive diagnostic techniques contribute to the difficulty in pinpointing the responsible organism. An inability to identify the causative pathogen greatly hampers the evaluation of nonresolving pneumonia.

Noninfectious causes
One small study showed that noninfectious causes were responsible for 20% of cases of nonresolving pneumonia.⁶ Despite this low incidence, a noninfectious etiology should be considered when patients do not respond to antibiotic therapy. Various neoplastic, inflammatory, vascular, or drug-induced disorders should be included in the differential diagnosis for immunocompetent as well as immunocompromised patients (Table 2). Our discussion is limited to a brief review of several common disorders that can mimic infectious pneumonia.

Lung cancer. An important cause of nonresolving pneumonia is lung malignancy, which can manifest in several ways. Endobronchial tumors may completely or partially block airways, leading to postobstructive pneumonia. In a study of 35 adults with nonresolving pneumonia, 4 (11%) had bronchioloalveolar-cell carcinoma or adenocarcinoma of the lung identified with bronchoscopy.⁷ In most cases of simple pneumonia, there will be no evidence of major atelectasis, because the air that normally fills the alveolar spaces is replaced by inflammatory exudate. Segmental or lobar atelectasis is a common feature of postobstructive pneumonia, resulting from air being reabsorbed from portions of the lung distal to the obstructing lesion. Radiographic signs of volume loss may be subtle or absent in patients with small endobronchial tumors, because the distal areas of the lung remain adequately ventilated.

In recent decades, adenocarcinoma has surpassed squamous-cell carcinoma as the most frequently diagnosed lung cancer in the United States. Adenocarcinoma currently accounts for about 40% of lung cancer cases, whereas the squamous-cell carcinomas commonly linked to smoking are responsible for only about 25% to 30% of cases.⁸

Shifting trends in histologic types may be explained by changes in exposure to tobacco products, diet, environmental or occupational factors, and host characteristics.⁹ Adenocarcinoma is most likely to occur in nonsmokers or former smokers. A recent study demonstrated that from 47% to 70% of patients with adenocarcinoma were either never smokers or former smokers.¹⁰ Adenocarcinoma is the most common form of lung cancer in women¹¹ and in people younger than 45.¹²

Typically, adenocarcinoma presents as a small, peripheral lesion with a high propensity to metastasize to regional lymph nodes and to distant sites. Because the primary tumor tends to develop in peripheral locations, patients are often asymptomatic. In patients with pneumonic-type adenocarcinoma of the lung, cancer can be detected as an incidental finding on routine chest radiography¹³ or by a primary search for distant metastases.¹⁴ Tumors frequently lie just below the pleura, and they can cause pleural retraction and thickening that is evident on chest radiographs.

Adenocarcinomas are subclassified based on the degree of differentiation-identified as well, moderate, or poorly differentiated. Pneumonic-type adenocarcinoma is characterized by segmental, lobar, or lung consolidation that resembles pneumonia on chest radiography and by adenocarcinoma with lepidic growth pattern on histopathologic examination. Bronchorrhea and crepitant rales are common features.¹³

Bronchioloalveolar-cell carcinoma is a subtype of adenocarcinoma that appears as a lung mass, nodule, or pulmonary infiltrates on chest x-ray. This tumor grows in layers within the bronchioles and alveolar spaces. About one third of patients with bronchioloalveolar-cell carcinoma have lobar consolidation; some have air bronchograms on chest films that can mimic pneumonia.¹⁵ Parenchymal lung involvement occurs in approximately 12% of patients with Hodgkin’s lymphoma and in 10% of patients with non-Hodgkin’s lymphoma.¹⁶ The radiographic manifestations of lymphoma in the lung are protean; however, hilar or mediastinal lymphadenopathy is almost always seen in patients with Hodgkin’s lymphoma.¹⁷

Drug-induced pneumonia. Drug-induced pneumonitis may resemble an infectious process, manifesting with high fever and pulmonary infiltrates on chest x-rays. Amiodarone HCl (Cordarone, Pacerone) causes pulmonary disease in as many as 10% of patients; preexisting lung disease and older age may increase risk.¹⁸ Nitrofurantoin (Macrobid, Macrodantin) causes 2 distinct pulmonary syndromes-acute and chronic. Acute reactions are more common, affecting about 1 of 5000 patients after initial exposure to the drug.¹⁹ The acute form can be confused with pneumonia, with symptoms such as fever, dyspnea, and cough generally becoming evident within about 1 month of ingestion. Chest x-ray may show bilateral infiltrates¹⁹; peripheral eosinophilia is a common finding.

Wegener’s disorder. Wegener’s granulomatosis is one of many inflammatory immunologic disorders that can mimic pneumonia. Patients may present with upper airway and sinus disease, as well as with renal involvement. The most common radiographic signs in the lung are parenchymal nodular regions of increased opacity (sometimes with cavitation), increased opacity in the airspaces, and pleural effusions.²⁰

Other etiologies. Another disorder to consider is bronchiolitis obliterans with organizing pneumonia (BOOP), which typically affects individuals in their 60s and 70s.²¹ BOOP has a waxing and waning course and is associated with viral pneumonia, collagen vascular diseases, malignancy, and other conditions. Sarcoidosis may also present with clinical and radiographic findings consistent with infectious pneumonia. Hilar and mediastinal adenopathy are common; interstitial and alveolar infiltrates suggest a more severe case of sarcoidosis.

An initial consideration in our patient was eosinophilic pneumonia, which can present in acute or chronic forms. Peripheral eosinophilia and pulmonary infiltrates are common manifestations of both acute and chronic pneumonia.

Diagnostic evaluation of nonresolving pneumonia
When patients with nonresolving pneumonia deteriorate despite appropriate antibiotic therapy, they should promptly undergo additional testing. Evaluation should be based on whether the culprit is likely to be related to host factors, antibiotic failure, patient noncompliance with therapy, improper dosing, or resistant or unusual pathogens. If there is no explanation for a slow response after 7 to 10 days of therapy, a careful reevaluation to identify treatable causes is indicated.²²

Ventilation-perfusion lung scanning, pulmonary angiography, chest CT, measurement of antineutrophil cytoplasmic antibodies, bronchoscopy, and lung biopsy may be helpful in excluding noninfectious causes. Before performing the invasive studies, however, it is important to review previous chest radiographs to identify preexisting abnormalities from any past surgery or inflammatory disease.

Chest CT, particularly high-resolution CT, is the primary diagnostic tool for assessing treatment failure in presumed pneumonia.²³ It provides superior detection of parenchymal abnormalities, including emphysema, airspace disease, interstitial disease, and nodules. Such findings may narrow the differential diagnosis or suggest new diagnoses. Chest CT also detects sequestered foci of infection, such as lung abscesses and empyema, and helps direct biopsy procedures.²³

The relative ease and low risk of fiberoptic bronchoscopy make it the primary diagnostic procedure for obtaining specimens for pathologic and microbiologic testing. Bronchoscopic evaluation is recommended in patients with unresolved radiographic changes that persist for more than 4 to 8 weeks (again, if the patient’s condition is declining, earlier evaluation and intervention is required).

The purpose of bronchoscopy is to exclude the presence of an unusual infectious or noninfectious process. Studies suggest that this procedure can identify disease in 90% of patients in whom a specific diagnosis can eventually be established.⁷ The diagnostic yield for bronchoscopy is greatest in young nonsmokers with multilobar disease. A previous nondiagnostic bronchoscopy, concern about a patient’s tolerance for bronchoscopy, or the need for larger specimen samples to establish the diagnosis may warrant thoracoscopic or open lung biopsy.¹

Conclusion
Both infectious and noninfectious conditions can be responsible for nonresolving pneumonia. Patients with persistent respiratory symptoms and pulmonary infiltrates, despite 10 to 14 days of antibiotic therapy, require diagnostic reevaluation. CT and bronchoscopy are the most useful diagnostic tools. In our patient, thoracentesis and bronchoscopy confirmed the diagnosis of adenocarcinoma of the lung. Some patients may ultimately require thoracoscopic or open lung biopsy to establish the diagnosis.

References
1. Kyprianou A, Hall CS, Shah R, et al. The challenge of nonresolving pneumonia. Knowing the norms of radiographic resolution is key. Postgrad Med. 2003;113:79-82, 85-88, 91-92.

2. Luna CM, Vujacich P, Niederman MS, et al. Impact of BAL data on the therapy and outcome of ventilator-associated pneumonia. Chest. 1997;111:676-685.

3. Mittl RL Jr, Schwab RJ, Duchin JS, et al. Radiographic resolution of community-acquired pneumonia. Am J Respir Crit Care Med. 1994;149: 630-635.

4. El Solh AA, Aquilina AT, Gunen H, et al. Radiographic resolution of community-acquired bacterial pneumonia in the elderly. J Am Geriatr Soc. 2004;52:224-229.

5. Marston BJ, Plouffe JF, File TM Jr, et al, for the Community-Based Pneumonia Incidence Study Group. Incidence of community-acquired pneumonia requiring hospitalization. Results of a population-based active surveillance study in Ohio. Arch Intern Med. 1997;157:1709-1718.

6. Menendez R, Torres A, Zalacain R, et al, for the Neumofail Group. Risk factors of treatment failure in community acquired pneumonia: implications for disease outcome. Thorax. 2004;59:960-965.

7. Feinsilver SH, Fein AM, Niederman MS, et al. Utility of fiberoptic bronchoscopy in nonresolving pneumonia. Chest. 1990;98:1322-1326.

8. American Cancer Society. All about lung cancer”--”non-small cell. Available at: www.cancer.org/docroot/CRI/CRI_2x.asp?site area=&dt=15. Accessed June 1, 2006.

9. Travis WD, Travis LB, Devesa SS. Lung cancer [published correction appears in Cancer. 1995;75:2979]. Cancer. 1995;75(suppl 1): 191-202.

10. Toh CK, Gao F, Lim WT, et al. Never-smokers with lung cancer: epidemiologic evidence of a distinct disease entity. J Clin Oncol. 2006; 24:2245-2251.

11. Ringer G, Smith JM, Engel AM, et al. Influence of sex on lung cancer histology, stage, and survival in a midwestern United States tumor registry. Clin Lung Cancer. 2005;7:180-182.

12. Yazgan S, Gursoy S, Yaldiz S, et al. Outcome of surgery for lung cancer in young and elderly patients. Surg Today. 2005;35:823-827.

13. Wislez M, Massiani MA, Milleron B, et al. Clinical characteristics of pneumonic-type adenocarcinoma of the lung. Chest. 2003;123:1868-1877.

14. Ishikawa A, Motohashi S, Shibuya K, et al. Small solitary pulmonary metastasis detected before primary sigmoid colon cancer: report of a case. Surg Today. 2003;33:709-711.

15. Albertine KH, Steiner RM, Radack DM, et al. Analysis of cell type and radiographic presentation as predictors of the clinical course of patients with bronchioloalveolar cell carcinoma. Chest. 1998;113:997-1006.

16. Maturen KE, Blane CE, Strouse PJ, et al. Pulmonary involvement in pediatric lymphoma. Pediatr Radiol. 2004;34: 120-124.

17. Romano M, Libshitz HI. Hodkgin disease and non-Hodgkin lymphoma: plain chest radiographs and chest computed tomography of thoracic involvement in previously untreated patients. Radiol Med (Torino). 1998;95:49-53.

18. Ashrafian H, Davey P. Is amiodarone an underrecognized cause of acute respiratory failure in the ICU? Chest. 2001;120: 275-282.

19. Williams EM, Triller DM. Recurrent acute nitrofurantoin-induced pulmonary toxicity. Pharmacotherapy. 2006;26: 713-718.

20. Prince JS, Duhamel DR, Levin DL, et al. Nonneoplastic lesions of the tracheobronchial wall: radiologic findings with bronchoscopic correlation. Radiographics. 2002;22(spec no): S215-S230.

21. Kroegel C, Reibetaig A, Hengst U, et al. Bilateral symmetrical upper-lobe opacities: an unusual presentation of bronchiolitis obliterans organizing pneumonia. Chest. 2000;118:863-865.

22. Halm EA, Teirstein AS. Clinical practice. Management of community-acquired pneumonia. N Engl J Med. 2002;347: 2039-2045.

23. Franquet T. Imaging of pneumonia: trends and algorithms. Eur Respir J. 2001;18:196-208.