Unfashionably Late
http://www.100md.com
《新英格兰医药杂志》
In this Journal feature, information about a real patient is presented in stages (boldface type) to an expert clinician, who responds to the information, sharing his or her reasoning with the reader (regular type). The authors' commentary follows.
An 18-year-old man presented with shortness of breath, a cough that was productive of clear sputum, and a two-week history of pleuritic chest pain. He also reported night sweats, fever, and fatigue, but no hemoptysis, weight loss, recent travel, or new environmental exposures.
The combination of progressive shortness of breath and pleuritic chest pain suggests either of two pathogenetic sequences. In one scenario, the primary process could have originated at the pleural surface and produced either a pleural effusion or a pneumothorax large enough to cause dyspnea. Alternatively, the process could have originated in the pulmonary parenchyma (causing dyspnea) and extended to the pleural surface (causing pleuritic chest pain). In either case, the presence of night sweats, fever, and fatigue suggests an underlying infection or a noninfectious inflammatory process, although a neoplasm (such as lymphoma) or even recurrent pulmonary embolism is also possible.
The patient had had two episodes of community-acquired pneumonia as a teenager; the last episode occurred two years before the current illness. Other than having a cyst removed from his neck as a child, he reported no serious illnesses or operations. He was taking no medications. There was no family history of cystic fibrosis, lung disease, or immune deficiencies. His immunizations were current. He was a recent high-school graduate who lived with his parents in suburban Michigan. He said that he did not use alcohol, tobacco, or illicit drugs, and he had no pets.
The patient's history of previous episodes of community-acquired pneumonia raises the question of an underlying susceptibility to bacterial or other infections, either because of a systemic problem associated with impaired pulmonary defense mechanisms (such as an immune deficiency, ciliary dysfunction, or cystic fibrosis) or because of focal endobronchial disease predisposing him to postobstructive infection (especially if the earlier bouts of pneumonia were in the same anatomic location). Although what he described as a neck cyst was probably unrelated to the current symptoms, the possibility of an underlying anatomic problem or complication following excision or that the cyst had been an infected lymph node (as in cases of tuberculous lymphadenitis), rather than a sterile cyst, should also be considered.
The patient was a thin white man in no acute distress. His temperature was 36.6°C, the heart rate 88 beats per minute, the respiratory rate 18 breaths per minute, the blood pressure 106/56 mm Hg, and the oxygen saturation 98 percent while he was breathing ambient air. His height was 163 cm, and his weight 55 kg. He had no cervical lymphadenopathy. He had bronchial breath sounds at both lung bases and egophony in his right mid-chest area. His back and both thighs were noted to have small acneiform lesions; he had no clubbing. The remainder of his physical examination was normal.
The white-cell count was 7.0 per cubic millimeter, with 74 percent neutrophils, 16 percent lymphocytes, and 10 percent monocytes. The hematocrit was 40 percent, and the platelet count 344,000 per cubic millimeter. The levels of aspartate and alanine aminotransferase were normal at 30 and 22 units per liter, respectively. The serum level of lactate dehydrogenase was 325 units per liter (normal range, 60 to 200). The results of an electrolyte panel and urinalysis were normal, as was the level of creatinine. A chest radiograph showed air-space opacities in the right middle lobe, lingula, and both lower lobes (Figure 1).
Figure 1. Chest Radiographs.
A posteroanterior view (Panel A) and a lateral view (Panel B) show bilateral air-space opacities in the right middle lobe, lingula, and both lower lobes.
The extensive air-space opacities and the consolidation suggest that the process causing pleuritic chest pain started in the pulmonary parenchyma and extended to the pleural surface. In addition, the multifocal nature of the opacities on the chest radiographs is inconsistent with a focal endobronchial obstruction leading to recurrent postobstructive pneumonia. My primary diagnostic considerations remain infection and noninfectious inflammatory disease, specifically bronchiolitis obliterans with organizing pneumonia or hypersensitivity pneumonitis, since both can have patchy airspace opacities that mimic bacterial pneumonia. The subacute clinical presentation, which took place over a number of weeks, combined with constitutional symptoms is particularly suggestive of bronchiolitis obliterans with organizing pneumonia, whereas the history of what was reported as two previous episodes of community-acquired pneumonia raises the possibility of recurrent episodes of hypersensitivity pneumonitis misdiagnosed as pneumonia.
The patient was admitted to the hospital and was given ceftriaxone and azithromycin. A test for the human immunodeficiency virus was negative, as were the results of tuberculin skin testing. Quantitative levels of immunoglobulins and of complement were normal. Tests for the presence of antinuclear antibody and antineutrophil cytoplasmic antibody were both negative; the Westergren erythrocyte sedimentation rate was 57 mm per hour. Several blood and sputum cultures obtained during febrile periods revealed no growth. Serologic tests for the presence of cytomegalovirus, Epstein–Barr virus, herpes simplex virus, and parvovirus B19 were negative. Immunodiffusion studies for histoplasmosis, coccidioidomycosis, and blastomycosis were nonreactive. A transthoracic echocardiogram showed no abnormalities.
Although patients with a clinical picture and a chest radiograph such as this patient had are typically given a diagnosis of bacterial pneumonia and started empirically on antibiotic treatment, the normal leukocyte count and the subacute presentation argue against this diagnosis. Consequently, my preference would have been to withhold empirical antibiotic therapy, unless the patient appeared particularly ill. One nonbacterial infection I would consider is blastomycosis, and the negative results on the immunodiffusion study do not dissuade me from this possibility, given the relatively poor diagnostic sensitivity of the test. The chest radiograph in a patient with pulmonary blastomycosis often shows one or more areas of consolidation that may mimic the appearance of a bacterial pneumonia — similar to the abnormalities seen in this patient.
Furthermore, if the acneiform lesions noted on the physical examination appear to be quite new, they could represent an early cutaneous manifestation of Blastomyces dermatitidis, whose name reminds us of the potential for involvement of the skin. However, I remain concerned about bronchiolitis obliterans with organizing pneumonia. As with blastomycosis, the areas of consolidation on the chest radiograph mimic the signs of pneumonia, but the clinical onset of bronchiolitis obliterans with organizing pneumonia is typically more gradual than the onset of bacterial pneumonia. The absence of a response to antibiotic treatment is often a clue to the diagnosis of bronchiolitis obliterans with organizing pneumonia.
High-resolution computed tomographic (CT) scans of the chest (Figure 2) revealed extensive air-space opacification involving the right lower lobe, accompanied by additional patchy and confluent areas of air-space opacification in the right middle lobe, lingula, and left lower lobe. There was also mild bronchiectasis. No lymphadenopathy was present. Examination by bronchoscopy showed normal airways; bronchoalveolar-lavage fluid was negative for acid-fast bacilli, as determined by staining, and for Pneumocystis carinii, as determined by polymerase-chain-reaction (PCR) assay. A transbronchial biopsy revealed nonspecific plasma-cell and lymphocytic infiltrates. The patient's fever abated and his dyspnea improved while he was receiving ceftriaxone and azithromycin. He was discharged to his home without antibiotics, having completed a 10-day course of treatment as an inpatient.
Figure 2. High-Resolution CT Scan of the Chest.
A representative CT image shows extensive air-space opacification involving the right lower lobe, accompanied by additional patchy and confluent areas of opacification in the lingula and the left lower lobe and by mild bronchiectasis. An area of opacification was also present in the right middle lobe (not shown).
Although the dense areas of consolidation in both lungs, as seen on the CT scans, are consistent with a diagnosis of bacterial pneumonia, they are also consistent with either blastomycosis or bronchiolitis obliterans with organizing pneumonia. In particular, the peripheral nature of some of the areas of consolidation may be indicative of bronchiolitis obliterans with organizing pneumonia. The presence of plasma-cell and lymphocytic infiltrates in the absence of neutrophils argues against a diagnosis of bacterial pneumonia, and the absence of eosinophils makes a diagnosis of either acute or chronic eosinophilic pneumonia improbable.
Although the patient is reported to have improved while receiving antibiotic agents, it is unclear whether the improvement was spontaneous and coincidental with institution of this treatment or was actually due to antibiotic action. At the moment, my leading diagnosis would be bronchiolitis obliterans with organizing pneumonia, but respiratory infections can be the precipitant for an inflammatory response with the pattern of bronchiolitis obliterans with organizing pneumonia, so that the diagnoses of infection and bronchiolitis obliterans with organizing pneumonia are not mutually exclusive.
During the two weeks after discharge, the patient became febrile, with a temperature as high as 40°C. He was readmitted to the hospital and given treatment with cefepime, clindamycin, and azithromycin. He showed little improvement after five days of hospitalization. Since the previous transbronchial biopsy had been nondiagnostic, open-lung biopsy with mediastinoscopy was performed and revealed necrotizing granulomas with atypical lymphocytic infiltration. Immunophenotyping showed no evidence of lymphoma. Staining for organisms was also negative. Initial cultures of the tissue specimen obtained during the lung biopsy showed no growth.
Despite the presumed improvement associated with the first regimen of antibiotic agents, it is clear that the clinical problem was far from resolved, and the biopsy to obtain tissue was appropriate. The finding of necrotizing granulomas with atypical lymphocytic infiltration on histopathological examination is not diagnostic, but it makes some earlier diagnostic considerations less probable and at the same time raises new possibilities. Mycobacterial or fungal infections, especially blastomycosis, remain a concern, despite the negative tuberculin skin test and the negative serologic studies for fungus. The histopathological features were not suggestive of bronchiolitis obliterans with organizing pneumonia, so among the noninfectious inflammatory (or even neoplastic) disorders, I would now be more concerned about those entities with prominent atypical lymphocytic or necrotizing granulomatous inflammation, such as lymphomatoid granulomatosis. The patient's history of persistent cough, dyspnea, and fever is consistent with this disorder, although imaging studies typically show nodules or masses, rather than air-space opacification. Nevertheless, at this point, my leading diagnosis would be either lymphomatoid granulomatosis or an infectious process, either fungal (especially blastomycosis) or mycobacterial.
In light of the pathological findings on lung biopsy, the patient was given a preliminary diagnosis of either bronchiolitis obliterans with organizing pneumonia or a lymphoproliferative disorder of unclear cause; he was started on oral prednisone (60 mg daily). He noted moderate improvement in his dyspnea and cough. He was discharged to his home, still taking corticosteroids, and with close outpatient follow-up included in his treatment plan.
I am concerned about the empirical use of corticosteroids in this setting, without a definite diagnosis. Through the nonspecific suppression of the inflammatory response, corticosteroids can result in initial symptomatic improvement in many clinical conditions, but the ultimate control or cure of the underlying problem requires additional or alternative treatment. The early improvement in the patient's dyspnea and cough after the institution of corticosteroids does not assure me that the presumed diagnosis of either bronchiolitis obliterans with organizing pneumonia or a lymphoproliferative disorder is correct. The possibility of an unrecognized infection remains, and the use of corticosteroids in the absence of appropriate antimicrobial coverage carries a risk, since these drugs will suppress his cellular immune response.
The patient continued with the oral prednisone at 60 mg daily. He was seen in the clinic three weeks after his discharge, and while he was there, the return of his dyspnea, fevers as high as 39.4°C, and cough with pleuritic chest pain were noted. He was admitted again to the hospital, and a review of the culture from the lung-biopsy specimen obtained three weeks earlier revealed growth of Burkholderia cepacia. He was started on intravenous ceftazidime and levofloxacin. A repeated high-resolution CT scan of the chest showed consolidation in both lower lobes and the right middle lobe, findings that were unchanged from the previous examination.
We now have an organism growing in culture from a lung-biopsy specimen, and the results are surprising. B. cepacia can be an environmental contaminant, a colonizer, or a true cause of disease; in this particular case, I would worry that it is a real pathogen. However, it typically occurs in the setting of certain disorders characterized by impaired pulmonary defense mechanisms, specifically in patients with cystic fibrosis or chronic granulomatous disease. Although both of these diseases are usually diagnosed during childhood, they can sometimes be found in late adolescence or early adulthood. Recurrent pneumonia and mild bronchiectasis could be sequelae of cystic fibrosis, and the necrotizing granulomas found on biopsy could indicate coexistent infection with nontuberculous mycobacteria, a well-recognized complication of cystic fibrosis. Alternatively, recurrent B. cepacia pneumonia associated with chronic granulomatous disease is also possible, in which case the granulomatous inflammation found on biopsy is a characteristic of the underlying disease.
Careful review of this patient's family history again revealed no relatives with cystic fibrosis or immune deficiencies. The sweat chloride test was normal, with a sweat weight of 22 g and a sweat chloride level of 24 mEq per liter (normal range, 1 to 39). A PCR assay for an abnormal cystic fibrosis transmembrane regulator (CFTR) gene was negative for a standard panel of 25 genes. The patient remained febrile while receiving intravenous antibiotics; the results of a repeated bronchoscopy showed normal airways. Bronchoalveolar lavage was performed, and the fluid was negative for bacteria and viruses on culture, for acid-fast bacilli on staining, and for Pneumocystis carinii on PCR assay.
Although a normal result from a chloride sweat test and negative test results for the standard panel of abnormal CFTR genes, taken alone, can each miss a small percentage of cases of cystic fibrosis, the fact that both of these tests were negative makes a diagnosis of cystic fibrosis extremely unlikely. Therefore, I believe that the unexpected growth of B. cepacia, in the setting of a negative workup for cystic fibrosis, leads us to chronic granulomatous disease as the most likely disorder responsible for this patient's clinical presentation, course, and histologic findings. Whereas we often consider an unusual presentation of a common disease to be more likely than a typical presentation of an uncommon disease, I think we may be dealing here with an even less likely scenario: an unusual presentation of an uncommon disease.
A neutrophil oxidative burst assay was performed. The patient's neutrophils showed no evidence of an oxidative burst in response to stimulation, whereas an assay performed concurrently in a normal person as a control showed a normal oxidative burst pattern (Figure 3). The results in the patient were considered diagnostic of chronic granulomatous disease. Another review of his history showed no evidence of recurrent infections or lymphadenopathy as a child, stomatitis, or enteritis. Genetic screening was offered to the patient's family.
Figure 3. Neutrophil Oxidative Burst Assay.
In the assay, neutrophils from the patient and a normal control are exposed to dichlorodihydrofluorescein diacetate (DCFH-DA), and a respiratory burst is stimulated with phorbol myristate acetate. As a result of the respiratory burst, DCFH-DA is oxidized to the green fluorescent compound dichlorodihydrofluorescein. Changes in fluorescence are then measured by a fluorescence reader and quantified as a change in the relative fluorescence units. In patients with chronic granulomatous disease, no oxidative burst is seen in response to stimulation with phorbol myristate acetate, and therefore there is little or no change in the relative fluorescence.
Commentary
Physicians are often confronted with the question of whether a premature disorder is the cause of a patient's symptoms; coronary artery disease as the cause of chest pain in a 36-year-old patient is one such example. The opposite phenomenon — considering illnesses that should have presented years earlier — is a less familiar scenario. In this case, perhaps if the patient's physicians had further considered disorders with delayed presentation earlier, the patient's condition might have been diagnosed sooner.
As the discussant points out, recurrent pneumonia in an otherwise healthy adolescent patient merits further investigation. Even though this young man had had two previous episodes of pneumonia, the reappearance of extensive lung disease and the subacute presentation should have raised the possibility of an underlying inherited disease at an earlier stage of his evaluation.
The clinical team, however, can hardly be faulted for not quickly recognizing an atypical presentation of chronic granulomatous disease. The disease was first described in 1957 as a "fatal granulomatosus of childhood."1,2 It soon became clear that neutrophils from affected patients, although capable of phagocytosis, were unable to generate active microbicidal oxygen species necessary to eradicate infection.3,4 The most common defect present in the X-linked recessive condition affects gp91phox, an integral membrane protein of NADPH oxidase.5 Three additional forms of the disease are due to autosomal recessive defects in other major components of the oxidase, p22phox, p47phox, and p67phox.5,6
It has been estimated that chronic granulomatous disease has an incidence of 1 per 200,000 births in the United States.5 Of 368 patients enrolled in a national registry of patients with chronic granulomatous disease, 70 percent had the X-linked recessive form and 22 percent had an autosomal recessive form; the genetic basis of the disease could not be determined in the remaining 8 percent.5 Eighty-five percent of the patients in these cases were male; 83 percent were white, 11 percent were black, 2 percent were Asian, less than 1 percent were Native American, 1 percent were of mixed ancestry, and 3 percent were of unknown race or ethnic group, on the basis of demographic information obtained from the physicians who enrolled patients in the registry. The mean age at diagnosis was three years in patients with the X-linked recessive form and eight years in those with autosomal recessive disease.5,7
Infection represents the most common clinical finding in patients with chronic granulomatous disease. The leading infections are pneumonia, abscess formation, suppurative adenitis, and osteomyelitis.5,7 Noninfectious complications related to granuloma formation have also been described and include colitis, gastric-outlet obstruction, and skin ulceration.6,8
The diagnosis of chronic granulomatous disease has historically been confirmed by the nitroblue tetrazolium test. Functioning neutrophils reduce the nitroblue tetrazolium dye from a clear yellow, water-soluble compound to a dark blue precipitant (formazan) in the presence of oxygen species produced during the respiratory burst.6,9 More sensitive tests are now available that involve probes whose fluorescent or chemiluminescent properties are altered by their reaction with reactive oxidants, such as the dihydrorhodamine-123 fluorescence test.6,10 After confirmation of the diagnosis with functional assays, the genotype can be determined by immunoblotting or direct sequencing.
The outcome for patients has improved substantially as compared with outcomes in earlier cohorts, in which only 70 percent of children with chronic granulomatous disease survived beyond the age of eight years.11,12 Currently, median survival is into the second decade of life.5 This improvement in survival is attributed primarily to widespread antimicrobial prophylaxis with trimethoprim–sulfamethoxazole.11,12 Adjunctive use of itraconazole has been effective as prophylaxis against fungal pathogens.12 Whereas treatment with interferon gamma reduces the frequency and severity of infections, when used alone or in conjunction with antimicrobial prophylaxis, its cost and side effects have limited its use.13 Newer methods, such as gene therapy and bone marrow transplantation, may one day represent curative options.14
Chronic granulomatous disease is only one example of a genetic disorder that may present later than usually described; other genetic disorders — for example, Tay–Sachs disease and cystic fibrosis — may do the same. This case reminds us to keep in mind conditions that show up earlier or later than expected, however unfashionable their arrival may seem.
Dr. Saint is the recipient of a Career Development Award from the Health Services Research and Development Program of the Department of Veterans Affairs and a Patient Safety Developmental Center Grant from the Agency for Healthcare Research and Quality (P20-HS11540).
Source Information
From the Departments of Internal Medicine (M.L., D.D., S.S.) and Pediatrics (M.L.), University of Michigan, Ann Arbor; the Department of Medicine, Beth Israel Deaconess Medical Center and Harvard Medical School, Boston (S.W.); and the Ann Arbor Veterans Affairs Health Services Research and Development Center of Excellence, and the Patient Safety Enhancement Program, Ann Arbor Veterans Affairs Medical Center and University of Michigan Health System (S.S.) — all in Ann Arbor.
Address reprint requests to Dr. Lukela at the Department of Internal Medicine, Division of General Medicine, 3119R Taubman Center, Ann Arbor, MI 48109-0376, or at mlukela@umich.edu.
References
Berendes H, Bridges RA, Good RA. A fatal granulomatosus of childhood: the clinical study of a new syndrome. Minn Med 1957;40:309-312.
Bridges RA, Berendes H, Good RA. A fatal granulomatous disease of childhood: the clinical, pathological, and laboratory features of a new syndrome. AMA J Dis Child 1959;97:387-408.
Baehner RL, Nathan DG. Leukocyte oxidase: defective activity in chronic granulomatous disease. Science 1967;155:835-836.
Quie PG, White JG, Holmes PG, Good RA. In vitro bactericidal capacity of human polymorphonuclear leukocytes: diminished activity in chronic granulomatous disease of childhood. J Clin Invest 1967;46:668-679.
Winkelstein JA, Marino MC, Johnston RB Jr, et al. Chronic granulomatous disease: report on a national registry of 368 patients. Medicine (Baltimore) 2000;79:155-169.
Segal BH, Leto TL, Gallin JI, Malech HL, Holland SM. Genetic, biochemical, and clinical features of chronic granulomatous disease. Medicine (Baltimore) 2000;79:170-200.
Johnston RB Jr. Clinical aspects of chronic granulomatous disease. Curr Opin Hematol 2001;8:17-22.
Barton LL, Moussa SL, Villar RG, Hulett RL. Gastrointestinal complications of chronic granulomatous disease: case report and literature review. Clin Pediatr (Phila) 1998;37:231-236.
Baehner RL, Boxer LA, Davis J. The biochemical basis of nitroblue tetrazolium reduction in normal human and chronic granulomatous disease polymorphonuclear leukocytes. Blood 1976;48:309-313.
Jirapongsananuruk O, Malech HL, Kuhns DB, et al. Diagnostic paradigm for evaluation of male patients with chronic granulomatous disease, based on the dihydrorhodamine 123 assay. J Allergy Clin Immunol 2003;111:374-379.
Cale CM, Jones AM, Goldblatt D. Follow up of patients with chronic granulomatous disease diagnosed since 1990. Clin Exp Immunol 2000;120:351-355.
Mouy R, Fischer A, Vilmer E, Seger R, Griscelli C. Incidence, severity, and prevention of infections in chronic granulomatous disease. J Pediatr 1989;114:555-560.
Gallin JI, Malech HL, Weening RS, et al. A controlled trial of interferon gamma to prevent infection in chronic granulomatous disease. N Engl J Med 1991;324:509-516.
Malech HL. Progress in gene therapy for chronic granulomatous disease. J Infect Dis 1999;179:Suppl 2:S318-S325.(Michael Lukela, M.D., Dav)
An 18-year-old man presented with shortness of breath, a cough that was productive of clear sputum, and a two-week history of pleuritic chest pain. He also reported night sweats, fever, and fatigue, but no hemoptysis, weight loss, recent travel, or new environmental exposures.
The combination of progressive shortness of breath and pleuritic chest pain suggests either of two pathogenetic sequences. In one scenario, the primary process could have originated at the pleural surface and produced either a pleural effusion or a pneumothorax large enough to cause dyspnea. Alternatively, the process could have originated in the pulmonary parenchyma (causing dyspnea) and extended to the pleural surface (causing pleuritic chest pain). In either case, the presence of night sweats, fever, and fatigue suggests an underlying infection or a noninfectious inflammatory process, although a neoplasm (such as lymphoma) or even recurrent pulmonary embolism is also possible.
The patient had had two episodes of community-acquired pneumonia as a teenager; the last episode occurred two years before the current illness. Other than having a cyst removed from his neck as a child, he reported no serious illnesses or operations. He was taking no medications. There was no family history of cystic fibrosis, lung disease, or immune deficiencies. His immunizations were current. He was a recent high-school graduate who lived with his parents in suburban Michigan. He said that he did not use alcohol, tobacco, or illicit drugs, and he had no pets.
The patient's history of previous episodes of community-acquired pneumonia raises the question of an underlying susceptibility to bacterial or other infections, either because of a systemic problem associated with impaired pulmonary defense mechanisms (such as an immune deficiency, ciliary dysfunction, or cystic fibrosis) or because of focal endobronchial disease predisposing him to postobstructive infection (especially if the earlier bouts of pneumonia were in the same anatomic location). Although what he described as a neck cyst was probably unrelated to the current symptoms, the possibility of an underlying anatomic problem or complication following excision or that the cyst had been an infected lymph node (as in cases of tuberculous lymphadenitis), rather than a sterile cyst, should also be considered.
The patient was a thin white man in no acute distress. His temperature was 36.6°C, the heart rate 88 beats per minute, the respiratory rate 18 breaths per minute, the blood pressure 106/56 mm Hg, and the oxygen saturation 98 percent while he was breathing ambient air. His height was 163 cm, and his weight 55 kg. He had no cervical lymphadenopathy. He had bronchial breath sounds at both lung bases and egophony in his right mid-chest area. His back and both thighs were noted to have small acneiform lesions; he had no clubbing. The remainder of his physical examination was normal.
The white-cell count was 7.0 per cubic millimeter, with 74 percent neutrophils, 16 percent lymphocytes, and 10 percent monocytes. The hematocrit was 40 percent, and the platelet count 344,000 per cubic millimeter. The levels of aspartate and alanine aminotransferase were normal at 30 and 22 units per liter, respectively. The serum level of lactate dehydrogenase was 325 units per liter (normal range, 60 to 200). The results of an electrolyte panel and urinalysis were normal, as was the level of creatinine. A chest radiograph showed air-space opacities in the right middle lobe, lingula, and both lower lobes (Figure 1).
Figure 1. Chest Radiographs.
A posteroanterior view (Panel A) and a lateral view (Panel B) show bilateral air-space opacities in the right middle lobe, lingula, and both lower lobes.
The extensive air-space opacities and the consolidation suggest that the process causing pleuritic chest pain started in the pulmonary parenchyma and extended to the pleural surface. In addition, the multifocal nature of the opacities on the chest radiographs is inconsistent with a focal endobronchial obstruction leading to recurrent postobstructive pneumonia. My primary diagnostic considerations remain infection and noninfectious inflammatory disease, specifically bronchiolitis obliterans with organizing pneumonia or hypersensitivity pneumonitis, since both can have patchy airspace opacities that mimic bacterial pneumonia. The subacute clinical presentation, which took place over a number of weeks, combined with constitutional symptoms is particularly suggestive of bronchiolitis obliterans with organizing pneumonia, whereas the history of what was reported as two previous episodes of community-acquired pneumonia raises the possibility of recurrent episodes of hypersensitivity pneumonitis misdiagnosed as pneumonia.
The patient was admitted to the hospital and was given ceftriaxone and azithromycin. A test for the human immunodeficiency virus was negative, as were the results of tuberculin skin testing. Quantitative levels of immunoglobulins and of complement were normal. Tests for the presence of antinuclear antibody and antineutrophil cytoplasmic antibody were both negative; the Westergren erythrocyte sedimentation rate was 57 mm per hour. Several blood and sputum cultures obtained during febrile periods revealed no growth. Serologic tests for the presence of cytomegalovirus, Epstein–Barr virus, herpes simplex virus, and parvovirus B19 were negative. Immunodiffusion studies for histoplasmosis, coccidioidomycosis, and blastomycosis were nonreactive. A transthoracic echocardiogram showed no abnormalities.
Although patients with a clinical picture and a chest radiograph such as this patient had are typically given a diagnosis of bacterial pneumonia and started empirically on antibiotic treatment, the normal leukocyte count and the subacute presentation argue against this diagnosis. Consequently, my preference would have been to withhold empirical antibiotic therapy, unless the patient appeared particularly ill. One nonbacterial infection I would consider is blastomycosis, and the negative results on the immunodiffusion study do not dissuade me from this possibility, given the relatively poor diagnostic sensitivity of the test. The chest radiograph in a patient with pulmonary blastomycosis often shows one or more areas of consolidation that may mimic the appearance of a bacterial pneumonia — similar to the abnormalities seen in this patient.
Furthermore, if the acneiform lesions noted on the physical examination appear to be quite new, they could represent an early cutaneous manifestation of Blastomyces dermatitidis, whose name reminds us of the potential for involvement of the skin. However, I remain concerned about bronchiolitis obliterans with organizing pneumonia. As with blastomycosis, the areas of consolidation on the chest radiograph mimic the signs of pneumonia, but the clinical onset of bronchiolitis obliterans with organizing pneumonia is typically more gradual than the onset of bacterial pneumonia. The absence of a response to antibiotic treatment is often a clue to the diagnosis of bronchiolitis obliterans with organizing pneumonia.
High-resolution computed tomographic (CT) scans of the chest (Figure 2) revealed extensive air-space opacification involving the right lower lobe, accompanied by additional patchy and confluent areas of air-space opacification in the right middle lobe, lingula, and left lower lobe. There was also mild bronchiectasis. No lymphadenopathy was present. Examination by bronchoscopy showed normal airways; bronchoalveolar-lavage fluid was negative for acid-fast bacilli, as determined by staining, and for Pneumocystis carinii, as determined by polymerase-chain-reaction (PCR) assay. A transbronchial biopsy revealed nonspecific plasma-cell and lymphocytic infiltrates. The patient's fever abated and his dyspnea improved while he was receiving ceftriaxone and azithromycin. He was discharged to his home without antibiotics, having completed a 10-day course of treatment as an inpatient.
Figure 2. High-Resolution CT Scan of the Chest.
A representative CT image shows extensive air-space opacification involving the right lower lobe, accompanied by additional patchy and confluent areas of opacification in the lingula and the left lower lobe and by mild bronchiectasis. An area of opacification was also present in the right middle lobe (not shown).
Although the dense areas of consolidation in both lungs, as seen on the CT scans, are consistent with a diagnosis of bacterial pneumonia, they are also consistent with either blastomycosis or bronchiolitis obliterans with organizing pneumonia. In particular, the peripheral nature of some of the areas of consolidation may be indicative of bronchiolitis obliterans with organizing pneumonia. The presence of plasma-cell and lymphocytic infiltrates in the absence of neutrophils argues against a diagnosis of bacterial pneumonia, and the absence of eosinophils makes a diagnosis of either acute or chronic eosinophilic pneumonia improbable.
Although the patient is reported to have improved while receiving antibiotic agents, it is unclear whether the improvement was spontaneous and coincidental with institution of this treatment or was actually due to antibiotic action. At the moment, my leading diagnosis would be bronchiolitis obliterans with organizing pneumonia, but respiratory infections can be the precipitant for an inflammatory response with the pattern of bronchiolitis obliterans with organizing pneumonia, so that the diagnoses of infection and bronchiolitis obliterans with organizing pneumonia are not mutually exclusive.
During the two weeks after discharge, the patient became febrile, with a temperature as high as 40°C. He was readmitted to the hospital and given treatment with cefepime, clindamycin, and azithromycin. He showed little improvement after five days of hospitalization. Since the previous transbronchial biopsy had been nondiagnostic, open-lung biopsy with mediastinoscopy was performed and revealed necrotizing granulomas with atypical lymphocytic infiltration. Immunophenotyping showed no evidence of lymphoma. Staining for organisms was also negative. Initial cultures of the tissue specimen obtained during the lung biopsy showed no growth.
Despite the presumed improvement associated with the first regimen of antibiotic agents, it is clear that the clinical problem was far from resolved, and the biopsy to obtain tissue was appropriate. The finding of necrotizing granulomas with atypical lymphocytic infiltration on histopathological examination is not diagnostic, but it makes some earlier diagnostic considerations less probable and at the same time raises new possibilities. Mycobacterial or fungal infections, especially blastomycosis, remain a concern, despite the negative tuberculin skin test and the negative serologic studies for fungus. The histopathological features were not suggestive of bronchiolitis obliterans with organizing pneumonia, so among the noninfectious inflammatory (or even neoplastic) disorders, I would now be more concerned about those entities with prominent atypical lymphocytic or necrotizing granulomatous inflammation, such as lymphomatoid granulomatosis. The patient's history of persistent cough, dyspnea, and fever is consistent with this disorder, although imaging studies typically show nodules or masses, rather than air-space opacification. Nevertheless, at this point, my leading diagnosis would be either lymphomatoid granulomatosis or an infectious process, either fungal (especially blastomycosis) or mycobacterial.
In light of the pathological findings on lung biopsy, the patient was given a preliminary diagnosis of either bronchiolitis obliterans with organizing pneumonia or a lymphoproliferative disorder of unclear cause; he was started on oral prednisone (60 mg daily). He noted moderate improvement in his dyspnea and cough. He was discharged to his home, still taking corticosteroids, and with close outpatient follow-up included in his treatment plan.
I am concerned about the empirical use of corticosteroids in this setting, without a definite diagnosis. Through the nonspecific suppression of the inflammatory response, corticosteroids can result in initial symptomatic improvement in many clinical conditions, but the ultimate control or cure of the underlying problem requires additional or alternative treatment. The early improvement in the patient's dyspnea and cough after the institution of corticosteroids does not assure me that the presumed diagnosis of either bronchiolitis obliterans with organizing pneumonia or a lymphoproliferative disorder is correct. The possibility of an unrecognized infection remains, and the use of corticosteroids in the absence of appropriate antimicrobial coverage carries a risk, since these drugs will suppress his cellular immune response.
The patient continued with the oral prednisone at 60 mg daily. He was seen in the clinic three weeks after his discharge, and while he was there, the return of his dyspnea, fevers as high as 39.4°C, and cough with pleuritic chest pain were noted. He was admitted again to the hospital, and a review of the culture from the lung-biopsy specimen obtained three weeks earlier revealed growth of Burkholderia cepacia. He was started on intravenous ceftazidime and levofloxacin. A repeated high-resolution CT scan of the chest showed consolidation in both lower lobes and the right middle lobe, findings that were unchanged from the previous examination.
We now have an organism growing in culture from a lung-biopsy specimen, and the results are surprising. B. cepacia can be an environmental contaminant, a colonizer, or a true cause of disease; in this particular case, I would worry that it is a real pathogen. However, it typically occurs in the setting of certain disorders characterized by impaired pulmonary defense mechanisms, specifically in patients with cystic fibrosis or chronic granulomatous disease. Although both of these diseases are usually diagnosed during childhood, they can sometimes be found in late adolescence or early adulthood. Recurrent pneumonia and mild bronchiectasis could be sequelae of cystic fibrosis, and the necrotizing granulomas found on biopsy could indicate coexistent infection with nontuberculous mycobacteria, a well-recognized complication of cystic fibrosis. Alternatively, recurrent B. cepacia pneumonia associated with chronic granulomatous disease is also possible, in which case the granulomatous inflammation found on biopsy is a characteristic of the underlying disease.
Careful review of this patient's family history again revealed no relatives with cystic fibrosis or immune deficiencies. The sweat chloride test was normal, with a sweat weight of 22 g and a sweat chloride level of 24 mEq per liter (normal range, 1 to 39). A PCR assay for an abnormal cystic fibrosis transmembrane regulator (CFTR) gene was negative for a standard panel of 25 genes. The patient remained febrile while receiving intravenous antibiotics; the results of a repeated bronchoscopy showed normal airways. Bronchoalveolar lavage was performed, and the fluid was negative for bacteria and viruses on culture, for acid-fast bacilli on staining, and for Pneumocystis carinii on PCR assay.
Although a normal result from a chloride sweat test and negative test results for the standard panel of abnormal CFTR genes, taken alone, can each miss a small percentage of cases of cystic fibrosis, the fact that both of these tests were negative makes a diagnosis of cystic fibrosis extremely unlikely. Therefore, I believe that the unexpected growth of B. cepacia, in the setting of a negative workup for cystic fibrosis, leads us to chronic granulomatous disease as the most likely disorder responsible for this patient's clinical presentation, course, and histologic findings. Whereas we often consider an unusual presentation of a common disease to be more likely than a typical presentation of an uncommon disease, I think we may be dealing here with an even less likely scenario: an unusual presentation of an uncommon disease.
A neutrophil oxidative burst assay was performed. The patient's neutrophils showed no evidence of an oxidative burst in response to stimulation, whereas an assay performed concurrently in a normal person as a control showed a normal oxidative burst pattern (Figure 3). The results in the patient were considered diagnostic of chronic granulomatous disease. Another review of his history showed no evidence of recurrent infections or lymphadenopathy as a child, stomatitis, or enteritis. Genetic screening was offered to the patient's family.
Figure 3. Neutrophil Oxidative Burst Assay.
In the assay, neutrophils from the patient and a normal control are exposed to dichlorodihydrofluorescein diacetate (DCFH-DA), and a respiratory burst is stimulated with phorbol myristate acetate. As a result of the respiratory burst, DCFH-DA is oxidized to the green fluorescent compound dichlorodihydrofluorescein. Changes in fluorescence are then measured by a fluorescence reader and quantified as a change in the relative fluorescence units. In patients with chronic granulomatous disease, no oxidative burst is seen in response to stimulation with phorbol myristate acetate, and therefore there is little or no change in the relative fluorescence.
Commentary
Physicians are often confronted with the question of whether a premature disorder is the cause of a patient's symptoms; coronary artery disease as the cause of chest pain in a 36-year-old patient is one such example. The opposite phenomenon — considering illnesses that should have presented years earlier — is a less familiar scenario. In this case, perhaps if the patient's physicians had further considered disorders with delayed presentation earlier, the patient's condition might have been diagnosed sooner.
As the discussant points out, recurrent pneumonia in an otherwise healthy adolescent patient merits further investigation. Even though this young man had had two previous episodes of pneumonia, the reappearance of extensive lung disease and the subacute presentation should have raised the possibility of an underlying inherited disease at an earlier stage of his evaluation.
The clinical team, however, can hardly be faulted for not quickly recognizing an atypical presentation of chronic granulomatous disease. The disease was first described in 1957 as a "fatal granulomatosus of childhood."1,2 It soon became clear that neutrophils from affected patients, although capable of phagocytosis, were unable to generate active microbicidal oxygen species necessary to eradicate infection.3,4 The most common defect present in the X-linked recessive condition affects gp91phox, an integral membrane protein of NADPH oxidase.5 Three additional forms of the disease are due to autosomal recessive defects in other major components of the oxidase, p22phox, p47phox, and p67phox.5,6
It has been estimated that chronic granulomatous disease has an incidence of 1 per 200,000 births in the United States.5 Of 368 patients enrolled in a national registry of patients with chronic granulomatous disease, 70 percent had the X-linked recessive form and 22 percent had an autosomal recessive form; the genetic basis of the disease could not be determined in the remaining 8 percent.5 Eighty-five percent of the patients in these cases were male; 83 percent were white, 11 percent were black, 2 percent were Asian, less than 1 percent were Native American, 1 percent were of mixed ancestry, and 3 percent were of unknown race or ethnic group, on the basis of demographic information obtained from the physicians who enrolled patients in the registry. The mean age at diagnosis was three years in patients with the X-linked recessive form and eight years in those with autosomal recessive disease.5,7
Infection represents the most common clinical finding in patients with chronic granulomatous disease. The leading infections are pneumonia, abscess formation, suppurative adenitis, and osteomyelitis.5,7 Noninfectious complications related to granuloma formation have also been described and include colitis, gastric-outlet obstruction, and skin ulceration.6,8
The diagnosis of chronic granulomatous disease has historically been confirmed by the nitroblue tetrazolium test. Functioning neutrophils reduce the nitroblue tetrazolium dye from a clear yellow, water-soluble compound to a dark blue precipitant (formazan) in the presence of oxygen species produced during the respiratory burst.6,9 More sensitive tests are now available that involve probes whose fluorescent or chemiluminescent properties are altered by their reaction with reactive oxidants, such as the dihydrorhodamine-123 fluorescence test.6,10 After confirmation of the diagnosis with functional assays, the genotype can be determined by immunoblotting or direct sequencing.
The outcome for patients has improved substantially as compared with outcomes in earlier cohorts, in which only 70 percent of children with chronic granulomatous disease survived beyond the age of eight years.11,12 Currently, median survival is into the second decade of life.5 This improvement in survival is attributed primarily to widespread antimicrobial prophylaxis with trimethoprim–sulfamethoxazole.11,12 Adjunctive use of itraconazole has been effective as prophylaxis against fungal pathogens.12 Whereas treatment with interferon gamma reduces the frequency and severity of infections, when used alone or in conjunction with antimicrobial prophylaxis, its cost and side effects have limited its use.13 Newer methods, such as gene therapy and bone marrow transplantation, may one day represent curative options.14
Chronic granulomatous disease is only one example of a genetic disorder that may present later than usually described; other genetic disorders — for example, Tay–Sachs disease and cystic fibrosis — may do the same. This case reminds us to keep in mind conditions that show up earlier or later than expected, however unfashionable their arrival may seem.
Dr. Saint is the recipient of a Career Development Award from the Health Services Research and Development Program of the Department of Veterans Affairs and a Patient Safety Developmental Center Grant from the Agency for Healthcare Research and Quality (P20-HS11540).
Source Information
From the Departments of Internal Medicine (M.L., D.D., S.S.) and Pediatrics (M.L.), University of Michigan, Ann Arbor; the Department of Medicine, Beth Israel Deaconess Medical Center and Harvard Medical School, Boston (S.W.); and the Ann Arbor Veterans Affairs Health Services Research and Development Center of Excellence, and the Patient Safety Enhancement Program, Ann Arbor Veterans Affairs Medical Center and University of Michigan Health System (S.S.) — all in Ann Arbor.
Address reprint requests to Dr. Lukela at the Department of Internal Medicine, Division of General Medicine, 3119R Taubman Center, Ann Arbor, MI 48109-0376, or at mlukela@umich.edu.
References
Berendes H, Bridges RA, Good RA. A fatal granulomatosus of childhood: the clinical study of a new syndrome. Minn Med 1957;40:309-312.
Bridges RA, Berendes H, Good RA. A fatal granulomatous disease of childhood: the clinical, pathological, and laboratory features of a new syndrome. AMA J Dis Child 1959;97:387-408.
Baehner RL, Nathan DG. Leukocyte oxidase: defective activity in chronic granulomatous disease. Science 1967;155:835-836.
Quie PG, White JG, Holmes PG, Good RA. In vitro bactericidal capacity of human polymorphonuclear leukocytes: diminished activity in chronic granulomatous disease of childhood. J Clin Invest 1967;46:668-679.
Winkelstein JA, Marino MC, Johnston RB Jr, et al. Chronic granulomatous disease: report on a national registry of 368 patients. Medicine (Baltimore) 2000;79:155-169.
Segal BH, Leto TL, Gallin JI, Malech HL, Holland SM. Genetic, biochemical, and clinical features of chronic granulomatous disease. Medicine (Baltimore) 2000;79:170-200.
Johnston RB Jr. Clinical aspects of chronic granulomatous disease. Curr Opin Hematol 2001;8:17-22.
Barton LL, Moussa SL, Villar RG, Hulett RL. Gastrointestinal complications of chronic granulomatous disease: case report and literature review. Clin Pediatr (Phila) 1998;37:231-236.
Baehner RL, Boxer LA, Davis J. The biochemical basis of nitroblue tetrazolium reduction in normal human and chronic granulomatous disease polymorphonuclear leukocytes. Blood 1976;48:309-313.
Jirapongsananuruk O, Malech HL, Kuhns DB, et al. Diagnostic paradigm for evaluation of male patients with chronic granulomatous disease, based on the dihydrorhodamine 123 assay. J Allergy Clin Immunol 2003;111:374-379.
Cale CM, Jones AM, Goldblatt D. Follow up of patients with chronic granulomatous disease diagnosed since 1990. Clin Exp Immunol 2000;120:351-355.
Mouy R, Fischer A, Vilmer E, Seger R, Griscelli C. Incidence, severity, and prevention of infections in chronic granulomatous disease. J Pediatr 1989;114:555-560.
Gallin JI, Malech HL, Weening RS, et al. A controlled trial of interferon gamma to prevent infection in chronic granulomatous disease. N Engl J Med 1991;324:509-516.
Malech HL. Progress in gene therapy for chronic granulomatous disease. J Infect Dis 1999;179:Suppl 2:S318-S325.(Michael Lukela, M.D., Dav)