Transtracheal aspiration in the diagnosis of pulmonary blastomycosis (17 cases: 2000–2005)

Abstract

Blastomyces dermatitidis is a common etiologic agent of fungal pneumonia in dogs. Definitive diagnosis is based on cytologic demonstration of the organism in affected tissues. Fluid obtained through transtracheal aspiration has previously been reported to have a low diagnostic yield for B. dermatitidis organisms. This retrospective study identified B. dermatitidis organisms in 76% of samples when transtracheal aspiration was performed in 17 nonsedated dogs with pulmonary blastomycosis. Transtracheal aspiration is a noninvasive and simple procedure that should be considered as an early diagnostic test whenever blastomycosis is a differential diagnosis in dogs with pulmonary disease.

Introduction

Blastomycosis is caused by infection with fungal spores of Blastomyces dermatitidis. Infection most commonly occurs through inhalation of conidiophores, with subsequent respiratory colonization (1). Infective conidia are phagocytized by alveolar macrophages and transformed from the mycelial phase to the yeast phase. Infection may be confined to the lung, or it may disseminate through hematogenous or lymphatic routes to distant sites (1,2).

Clinical signs of B. dermatitidis infection reflect the inflammatory and multisystemic nature of the disease. Anorexia, weight loss, and fever are common, with 40% to 60% of affected dogs having a fever of 39.4°C (103°F) or greater (2). Pulmonary lesions occur in 65% to 85% of cases and may be clinically silent or, more often, associated with respiratory signs, including exercise intolerance, tachypnea, and cough (1,2). Lymphadenopathy occurs in 40% to 60% of affected dogs (1–3). Ocular involvement may occur with or without concurrent pulmonary disease, resulting in uveitis, chorioretinitis, or panophthalmitis in 20% to 50% of affected dogs. Granulomatous or ulcerative cutaneous lesions occur in 30% to 50% of cases, particularly involving the nasal planum, the face, and nailbeds (1–3).

Accurate and rapid diagnosis of pulmonary blastomycosis is important due to the life threatening nature of severe disease and improved survival with early diagnosis (2,4). Routine laboratory testing does not allow for definitive diagnosis, but it may be supportive of systemic inflammatory disease. Most affected dogs exhibit a moderate leukocytosis, hyperglobulinemia, and mild hypoalbuminemia (2). Thoracic radiographs typically reveal diffuse or nodular interstitial and bronchointerstitial lung changes (1,2). Less commonly, well-defined solitary or multiple nodules or masses will be identified within the pulmonary parenchyma. Tracheobronchial lymphadenopathy may occur (1). Characteristic radiographic lesions in a dog with fever and an inflammatory leukogram support a diagnosis of fungal disease; however, definitive diagnosis can only be made by identifying organisms retrieved from affected tissues by aspiration or biopsy.

Diagnostic techniques employed to collect samples from the respiratory tract for cytologic examination include transtracheal aspirates (TTA), bronchoalveolar lavage (BAL), and fine-needle pulmonary aspirates. Examination of samples obtained by using TTA has been reported to be an insensitive procedure for the diagnosis of pulmonary blastomycosis, with a diagnostic yield of less than 50% (1–6). The low diagnostic yield (% positive results) of TTA has been speculated to be due to the interstitial location of the organism (1).

The purpose of this retrospective study was to investigate the diagnostic utility of cytologic evaluation of samples, obtained by using TTA, to definitively diagnose B. dermatitidis in dogs with pulmonary blastomycosis. This study was undertaken, despite literature reports of poor diagnostic utility, because of the authors’ clinical impression that TTA could be a useful procedure for the diagnosis of pulmonary blastomycosis.

Materials and methods

Medical records of all dogs evaluated at the Veterinary Teaching Hospital, Western College of Veterinary Medicine, University of Saskatchewan, during a 5-year period (between January 2000 and December 2005) with a diagnosis of blastomycosis were reviewed to obtain data for this retrospective study. Data collected included radiographic features of lung lesions in each case, nonpulmonary organs affected, method of diagnosis, and prior medical therapy. Fluid samples obtained by TTA were examined by a clinical pathologist who rated them as positive (B. dermatitidis organisms present) or negative.

Transtracheal aspirates had been performed according to a standardized protocol by a variety of VTH clinicians, including veterinary students, interns, residents in internal medicine, and board-certified internists. The TTA procedure had been completed on awake (nonsedated) dogs in order to maintain a strong cough reflex and provide the best possible sample. Slightly different techniques and equipment had been used to perform the TTA in small dogs (< 15 kg) versus large dogs (> 15 kg). In small dogs, a 16–20 gauge through-the-needle catheter (Intra-cath Becton; Dickinson, Franklin Lakes, New Jersey, USA) was used; in large dogs, a sterile 70-cm (28-in), 3.5 french polypropylene catheter was passed through a 14-gauge over-the-needle catheter (Medicut catheter; Sherwood Medical Industries, St. Louis, Missouri, USA). The dog was restrained in a sternal or sitting position with the neck extended dorsally. The region over the larynx was clipped and aseptically prepared. A lidocaine block (0.5 mL of 2% lidocaine) was completed. The needle was inserted through the cricothyroid ligament into the tracheal lumen and the catheter advanced to approximately the level of the tracheal bifurcation. Sterile saline aliquots [0.25–0.50 mL/kg body weight (BW)] were then infused and aspirated as the dog coughed. Fluid samples obtained by TTA were used to prepare direct smears and concentrated slides (Shandon Cytospin 2; Thermo Fischer Scientific; Waltham, Massachussets, USA), which were then stained using Wright’s-Giemsa stain and examined by the duty pathologist.

Results

During the 5-year study period, the medical records of 29 cases of blastomycosis were identified and reviewed. Twenty-six of the 29 cases had radiographic evidence of pulmonary involvement at the time of diagnosis. One case with pulmonary disease was primarily being evaluated for dermal complications of itraconazole therapy. One of the 3 dogs without pulmonary lesions had disease limited to the central nervous system, which had been diagnosed on postmortem examination; another had ocular and skin lesions; and the 3rd had bone lesions. Eleven of the 26 dogs with pulmonary blastomycosis had no identifiable extrapulmonary involvement while the other 15 had evidence of concurrent ocular (n = 10), lymph node (n = 1), joint (n = 1), skin and ocular (n = 2), and skin (n = 1) involvement.

Diagnosis of blastomycosis had been made through cytologic identification of organisms in lymph node aspirates, fine needle aspirates of cutaneous lesions, transtracheal aspirates, fine needle lung aspirates, vitreal aspirates, ocular histopathologic examination, or on postmortem examination. In 9 dogs with pulmonary blastomycosis, the diagnosis had been made by sampling extrapulmonary tissues or on postmortem examination, eliminating the need for collection of a respiratory sample. Transtracheal aspiration was performed in 17/26 cases with pulmonary parenchymal disease; in 13/17 (76.4%) of these cases, a definitive diagnosis of blastomycosis had been made, based on cytologic demonstration of fungal organisms in the TTA sample. No complications of TTA were reported in the medical records of any dog.

Thoracic radiographs were reviewed from the 17 dogs with pulmonary blastomycosis that underwent TTA. A nodular-interstitial pattern of pulmonary parenchymal involvement (7/17) was most common, while radiographs from the other dogs had been classified as a miliary interstitial pattern (3/17), a diffuse interstitial pattern (3/17), a focal region of consolidation (3/17), or a solitary nodule (1/17) (Table 1). Transtracheal aspirate cytologic examination yielded a diagnosis in 66% to 100% of the dogs with different radiographic classifications of pulmonary disease (Table 1).

Table 1.

Radiographic abnormalities and diagnostic yield of transtracheal aspiration (TTA) in 17 nonsedated dogs diagnosed as having pulmonary blastomycosis

Radiographic findings	Dogs	TTA diagnostic yield (% positive results)
Nodular, interstitial	7	(71.4%) 5/7
Focal consolidation	3	(66.6%) 2/3
Diffuse miliary interstitial	3	(100%) 3/3
Diffuse interstitial	3	(66.6%) 2/3
Solitary nodule	1	(100%) 1/1

Two of the dogs had been treated with prednisone prior to referral for lung disease and both of these dogs had a large number of organisms visible in their TTA sample.

Discussion

Accurate rapid diagnosis of pulmonary blastomycosis early in the course of the disease is important, so that aggressive therapy can be initiated. Delays in diagnosis and treatment can be critical, with 50% of all dogs that develop severe lung disease prior to treatment succumbing to respiratory failure (2,4).

Dogs with concurrent involvement of other organs may have a quick diagnosis achieved through simple noninvasive testing, such as cytologic examination of aspirates from enlarged lymph nodes or impression smears of ulcerated skin lesions (1,2,6). When the disease is primarily ocular and less invasive techniques have failed, vitreous aspirates or histologic examination of enucleated blind eyes can also be used to reach a diagnosis (2,6,8). Recovery of the organism in samples obtained from the lungs of dogs with pulmonary blastomycosis has been reported to be more problematic (1–6).

Diagnostic samples can be collected from the lung by using a variety of techniques, including TTA, BAL, and transthoracic pulmonary aspiration. Bronchoalveolar lavage collects a more representative sample from deep in the lung parenchyma and is more sensitive than TTA for fungal organism identification, but it requires general anesthesia in a patient already experiencing respiratory difficulty (2). Ultrasound-guided fine needle aspirates of focal pulmonary lesions adjacent to the body wall can be used to reliably identify organisms in dogs with blastomycosis, but when lesions are separated from the body wall by aerated lung, complications, including pneumothorax, hemothorax, and death, although rare, have all been reported (2,7).

The least invasive and safest method for collecting a sample from the respiratory tract of a dog for cytologic examination is TTA. When noninvasive sampling of extrapulmonary sites does not yield a diagnosis in dogs with suspected blastomycosis, TTA is an appropriate initial diagnostic test to collect respiratory samples. Transtracheal aspiration does not require sedation or anesthesia, and complications are minimal. The equipment needed is inexpensive and readily available to all veterinary practitioners. The results of this retrospective study suggest that TTA samples can provide a definitive diagnosis in many dogs with pulmonary blastomycosis.

Previous reports that cytologic examination of TTA samples is not useful in the diagnosis of pulmonary blastomycosis have been based on results in small numbers of dogs and, in some cases, the technique used for TTA has been either suboptimal or not described. In one retrospective study of 115 dogs with blastomycosis, TTA was performed in 10 cases with a diagnostic yield of 30% (4). Evaluation of TTA as a diagnostic tool was not the primary focus of that paper, so the TTA technique used was not described. In another study of 46 dogs with blastomycosis, TTA was found to contribute to the diagnosis in approximately only 5% of cases; it was stated that TTA was rarely performed because a diagnosis was achieved through other methods (6). In yet another study, Hawkins and DeNicola (5) reported that endotracheal washes yielded fluid containing B. dermatitidis organisms in 3/7 animals in which the procedure was performed. Endotracheal washes are performed under general anesthesia through an endotracheal tube, making vigorous coughing and collection of very cellular specimens less likely than with TTA performed in awake dogs.

Although TTA has previously been reported to have a low diagnostic yield for B. dermatitidis organisms, this retrospective study found that B. dermatitidis organisms were identified in 76% of TTA samples. Many clinicians and students had performed the TTA procedure in these dogs and it is possible that poor technique could have interfered with obtaining diagnostic results in the negative cases. The diagnostic success of cytologic examination even in the dog with a solitary nodule and in 2/3 dogs with focal consolidation supports the clinical impression that this test can be useful in animals with varying radiographic features of pulmonary blastomycosis. Transtracheal aspiration in the nonsedated animal is a noninvasive and simple procedure and should be considered as an early diagnostic test whenever blastomycosis is a differential diagnosis in dogs with pulmonary disease. CVJ