Blastomyces dermatitidis is one of the dimorphic fungi known to cause disease primarily within certain endemic areas scattered throughout the world. The endemic regions of North America blastomycosis include the southeastern, south central, and midwestern states (especially areas bordering the Mississippi and Ohio Rivers and the Great Lakes), adjacent areas of Canada, and a small area in upper New York State and Canada that follows the St. Lawrence River. Within these areas, blastomycosis has occurred sporadically or in outbreaks. Infection is acquired via inhalation of airborne spores from disturbed contaminated soil. Although B dermatitidis is highly infectious, symptomatic disease seems to develop in less than half of those infected.
Blastomycosis is a great masquerader. It can present in several ways, as either acute or chronic disease, and involve one or several body sites, the most common being the lungs, followed by skin and bone. Pulmonary blastomycosis can produce a wide variety of radiographic findings, ranging from lobar consolidation to miliary infiltration to large masses. Thus, pulmonary blastomycosis can present in a manner indistinguishable from bacterial pneumonia, tuberculosis, histoplasmosis, ARDS, or bronchogenic carcinoma. The diversity of potential misdiagnoses underscores the importance of recognizing this infection in order to avoid errors in treatment and potential complications from unnecessary procedures or progression of the untreated infection. To make a definitive diagnosis, B dermatitidis must be isolated in culture, or observed in a smear or histologic specimen. Clinician awareness of the many faces of this disease will enhance the selection of proper diagnostic specimens and laboratory methods to identify the pathogen.
In this issue of CHEST (see page 768), Martynowicz and Prakash describe how pulmonary blastomycosis presented and was diagnosed at the Rochester Mayo Medical Center over a 20-year period. Included in their analysis are patients with a wide spectrum of clinical pictures, primarily acute pneumonitis or pulmonary mass or nodule, but also chronic pulmonary infiltration, ARDS, and empyema. The confusing nature of this infection is illustrated by one patient with a blastomycotic mass who had been treated with radiotherapy for presumed bronchogenic carcinoma before the infectious nature of the lesion was recognized.
The difficulty of diagnosing pulmonary blastomycosis lies not in growing the organism in culture or identifying the organism in smears and biopsies. Unlike the other endemic fungi, B dermatitidis evokes a purulent inflammatory response so that adequate sputum or pus drainage specimens can usually be obtained. These specimens have had a high yield of recovery in culture. (1,2) The ease of diagnosis should be further augmented by the large size of the tissue yeast phase organism (5 to 20 [micro]m in diameter), and the easily recognizable "figure-of-eight morphology" with a characteristic thick, doubly refractile cell wall, and single wide-based budding.
Martynowicz and Prakash do report a high diagnostic yield by culture of noninvasive respiratory specimens (86% per patient), including sputum, tracheal secretions, and gastric washings. The diagnostic yield of bronchoscopy, 92%, was of a similar magnitude to that of the noninvasive specimens. In one case, the sputum culture finding was positive despite a nondiagnostic bronchoscopy. Previous series (1,2) have concurred that most cases of acute pulmonary blastomycosis can be confirmed via sputum culture, although examination of several specimens may be needed. Other investigators (3) have indicated lack of significant difference in diagnostic yield between sputum analysis and bronchoscopy. So the recommended sequence of workup is simple: (1) initially, obtain multiple sputum samples for stain and culture; (2) proceed to bronchoscopy for patients with negative sputum specimen findings, those who are unable to produce sputum, or to exclude another infection or malignancy; and 3) proceed to more invasive procedures, like thoracoscopy or thoracotomy, only in the occasional patient in whom sputum and bronchoscopy are nondiagnostic.
Unfortunately, there can be a glitch in this straightforward diagnostic approach. The time it takes to culture and identify B dermatitidis may vary from 5 to 30 days, so that a period of 4 to 5 weeks is required before the culture finding can be considered negative. This may be too long to wait to avoid aggressive antibiotic treatment for presumed bacterial pneumonia or tuberculosis or a thoracotomy to excise presumed lung cancer. Thus, although the positive culture finding is an invaluable piece of confirmatory diagnostic information, the ability to make an immediate diagnosis may have greater impact on patient management, if not via an early positive culture finding, by smears, histology, or serologic tests. This is where clinician vigilance in endemic areas can make a difference.
Martynowicz and Prakash found that KOH smears of respiratory specimens were infrequently obtained in their patients with pulmonary blastomycosis, and when they were, they had a relatively low diagnostic yield. The low yield of KOH smears has been reported by others, (2,3) but there is evidence that it can be increased with the examination of multiple specimens.3 In endemic areas, patients with compatible radiographic features who are able to produce sputum may benefit by submitting several sputum specimens for KOH preparation.
Papanicolau staining of sputum and other respiratory specimens is another quick and reliable means of making an immediate diagnosis. The fungi are seen as large pink to red structures bearing their typical morphology. The main difficulty with Papanicolaou smears is that they are usually performed to screen for malignancy, and even the characteristic appearance of B dermatitidis may be overlooked unless the cytotechnician is trained to look for them. In the series by Martynowicz and Prakash, Papanicolaou cytologic preparations of respiratory specimens were seldom performed, although other series have reported a high diagnostic yield. (3) These authors have proposed the introduction of routine Papanicolaou smears in patients from endemic areas with pulmonary nodules or masses or undiagnosed pneumonia. This would seem to be a prudent approach, at least for patients able to produce adequate sputum for these preparations. Awareness and training of laboratory personnel would be another part of the equation, which would hopefully amount to a better chance of noninvasive diagnosis of pulmonary blastomycosis.
If bronchoscopy or thoracoscopy/thoracotomy must enter the diagnostic sequence, a couple of caveats should be kept in mind. First, lidocaine treatment has been shown to inhibit the growth of B dermatitidis in culture. Thus, limiting the concentration of lidocaine to 1 g/dL has been recommended in suspected cases to reduce the antifungal effect. (4) Second, B dermatitidis is frequently missed in standard hematoxylin and eosin-stained histologic specimens, so special stains, usually silver or periodic acid-Schiff stains, are often needed to identify the organism.
A serologic test, usually being inexpensive and noninvasive, might be an ideal way to screen for blastomycosis and direct the workup for possible active cases in endemic areas. Regrettably, Martynowicz and Prakash, like others, have found the currently available commonly used serologic techniques to have inadequate sensitivity for such an application. Yet work is ongoing in this area, and in the future, novel serologic or molecular biological techniques may make the job of uncovering veiled B dermatitidis infections easier.
Blastomycosis is not a common infection. The Centers for Disease Control and Prevention estimates the incidence in endemic areas to be one to two cases per 100,000 population. Yet, the consequences of missing the diagnosis can be significant, including not only an unfavorable clinical outcome, but also substantial distress to the patient as he or she faces potentially unnecessary tests and procedures while having to cope with the uncertainty of diagnosis. Understanding the various forms of this treatable infection, especially by clinicians who practice in endemic areas, is the first step to smooth diagnosis and treatment. Then, suspicion in patients whose exposure history and disease presentation put blastomycosis into the differential diagnosis should direct collection of adequate, and perhaps, multiple sputum specimens to be cultured and stained for fungal detection, including by Papanicolaou stain. Specimens from bronchoscopy and other more invasive procedures should also be appropriately collected and processed. With the proper knowledge and watchfulness, hopefully the true identity of this great masquerader will, in more cases than not, be promptly exposed.
Dr. Wallace is Associate Professor, Department of Medicine, Olive View-UCLA Medical Center.
Correspondence to: Jeanne Wallace, MD, FCCP, Department of Medicine, Olive View-UCLA Medical Center, 14445 Olive View Dr, Sylmar, CA 91011; e-mail: firstname.lastname@example.org
(1) Areno JP IV, Campbell GD Jr, George RB. Diagnosis of blastomycosis. Semin Respir Infect 1997; 12:252-262
(2) Patel RG, Patel B, Petrini MF, et al. Clinical presentation, radiographic findings, and diagnostic methods of pulmonary blastomycosis: a review of 100 cases. South Med J 1999; 92:289 -295
(3) Tumbull ML, Chesney TM. The cytological diagnosis of pulmonary blastomycosis. JAMA 1981; 245:836-838
(4) Taylor MR, Lawson LA, Boyce JM, et al. Inhibition of Blastomyces dermatitidis by topical lidocaine. Chest 1983; 84:431-432
COPYRIGHT 2002 American College of Chest Physicians
COPYRIGHT 2002 Gale Group