Cytology of bronchoalveolar lavage in some rare pulmonary disorders: pulmonary alveolar proteinosis and amiodarone pulmonary toxicity

Bronchoalveolar lavage as a diagnostic procedure: a review of known cellular and molecular findings in various lung diseases

Bronchoalveolar lavage (BAL) is a commonly used procedure in the evaluation of lung disease as it allows for sampling of the lower respiratory tract. In many circumstances, BAL differential cell counts have been reported to be typical of specific lung disorders. In addition, more specific diagnostic tests including molecular assays such as polymerase chain reaction (PCR) or enzyme-linked immunosorbent assay, special cytopathologic stains, or particular microscopic findings have been described as part of BAL fluid analysis. This review focuses on common cellular and molecular findings of BAL in a wide range of lung diseases. Since the performance of the first lung irrigation in 1927, BAL has become a common and important diagnostic tool. While some pulmonary disorders have a highly characteristic signature of BAL findings, BAL results alone often lack specificity and require interpretation along with other clinical and radiographic details. Development of new diagnostic assays is certain to reinforce the utility of BAL in the future. Our review of the BAL literature is intended to serve as a resource to assist clinicians in the care of patients with lung disorders.

Expansion of Pulmonary CD8+CD56+ Natural Killer T-Cells in Hypersensitivity Pneumonitis

BACKGROUND

Natural killer T (NKT) cells, a newly identified subgroup of T cells with immunoregulatory function, may be implicated in the pathogenesis of interstitial lung disease (ILD).

METHODS

We used multiparameter flow cytometry with antibodies to CD3, CD4, CD8, CD14, CD19, CD45, CD16/56, CD56, CD161, and Valpha24 invariant T-cell receptor (TCR) in BAL fluid (BALF) to examine the frequency and distribution of pulmonary NKT cells in several cases of ILD. We included 57 patients with sarcoidosis and 17 patients with hypersensitivity pneumonitis.

RESULTS

We found significantly higher frequencies of pulmonary NKT cells in patients with hypersensitivity pneumonitis in comparison to the other study patients with ILD (median proportion of NKT cells, 11%; range, 3 to 38%; vs 3%; range, 0 to 16%; p < 0.0001). In contrast, there was no difference in the proportion of conventional natural killer cells. We found that a major subset of NKT cells in the BALF of patients with hypersensitivity pneumonitis was a CD8+CD56+ population that did not express the invariant TCR.

CONCLUSIONS

These results suggest the involvement of NKT cells in the pathogenesis of hypersensitivity pneumonitis.

Pulmonary alveolar proteinosis: a spectrum of cytologic, histochemical, and ultrastructural findings in bronchoalveolar lavage fluid

Pulmonary alveolar proteinosis (PAP) is defined as abundant extracellular proteinaceous periodic acid-Schiff (PAS)-positive material which represents surfactant distending alveolar spaces. While this lesion is defined by histologic findings, there are characteristic radiologic features and cytologic findings in bronchoalveolar lavage (BAL) specimens that together may provide a confident diagnosis. The BAL specimens from all patients for which a diagnosis of PAP was made or suggested on either cytologic or biopsy specimens at University of North Carolina Hospitals from 1990-1999 were reviewed. There were 23 cytologic specimens from 11 patients. Patient ages ranged from 6 wk to 76 yr. All 23 specimens had slides prepared for Papanicolaou stain, 22 specimens (all patients) had Diff-Quik stains, 10 specimens (6 patients) had PAS stains, and 8 specimens (5 patients) had lipid stains. Nine patients had lung biopsies in addition to cytologic specimens. The clinical charts of all patients were reviewed. Twenty-one cytologic specimens were described as cloudy or milky, and 2 were bloody. By chart review and/or biopsy results, 8 patients were felt to have definite PAP. The initial lavage specimens from 6 of these patients showed classic cytologic findings of PAP, consisting of paucicellular specimens dominated by adundant extracellular granular to globular material which was basophilic on Diff-Quik stain, pale to focally eosinophilic on Pap stain, and PAS-positive, diastase-resistant. Five of these patients had biopsies; 3 showed PAP, and 2 were insufficient. Later BAL specimens after therapeutic lavage from these patients were often less characteristic, with scant extracellular material present. The other 2 patients with PAP clinically and by biopsy had atypical cytologic findings, with one showing numerous macrophages with scant PAS-positive material and abundant lipid mimicking lipid pneumonia, and one showing moderate eosinophils in addition to the extracellular proteinacous material. The remaining 3 patients were felt not to have PAP clinically or by biopsy (1 lymphocytic interstitial pneumonitis, 1 rheumatoid lung, and 1 hemosiderosis), and their BAL specimens predominantly contained macrophages with rare proteinaceous extracellular globules. Electron microscopy was performed in 5 patients (4 considered to have PAP, and 1 with lymphocytic interstitial pneumonitis) and in all cases showed whorled myelin figures characteristic of surfactant. The PAP cases and the non-PAP case had identical ultrastructural findings. We conclude that BAL specimens with classic cytologic features and supporting clinical and radiographic evidence may be diagnosed as PAP. Atypical specimens should be approached with caution, and may represent either PAP or other pulmonary diseases with secondary accumulation of surfactant. Cytology specimens taken subsequent to therapeutic lavage from PAP patients may also not be diagnostic.

Acute respiratory failure caused by secondary alveolar proteinosis in a patient with acute myeloid leukemia: a case report

Pulmonary alvelolar proteinosis (PAP) is a rare cause of chronic respiratory failure due to progressive alveolar accumulation of a periodic acid-schiff (PAS) positive proteinaceous material. In some cases, the rapid accumulation of intra-alveolar material leads to acute respiratory failure (ARF). We report the causative role of secondary PAP in the case of a 26-year-old man with acute myeloid leukemia who developed fever, increased serum lactate dehydrogenase level and ARF, and required mechanical ventilation. The diagnosis of PAP was established by the examination of material obtained by bronchoalveolar lavage (BAL). Respiratory improvement occurred several days after the patient had recovered from neutropenia. This report underlines the importance of the early diagnosis of PAP as a potential cause of ARF in leukemic patients. Adequate stain on BAL fluid provides the diagnosis and avoids repeated invasive procedures and inappropriate treatments.

Pulmonary alveolar proteinosis: diagnosis using routinely processed smears of bronchoalveolar lavage fluid

AIMS

For the diagnosis of pulmonary alveolar proteinosis from bronchoalveolar lavage specimens it is normally necessary to make an ultrastructural examination. However, this is thought to be impractical for bronchoalveolar lavage specimens that have been routinely fixed in ethanol. In the present study, bronchoalveolar lavage cytology smears on slide glasses were examined directly ultrastructurally to make a diagnosis of pulmonary alveolar proteinosis.

METHODS

Bronchoalveolar lavage smears from three pulmonary alveolar proteinosis patients were stained with Papanicolaou and periodic acid-Schiff (PAS) for identification of amorphous globular structures. Subsequently, they were refixed with glutaraldehyde and osmium tetroxide, and embedded in epoxy resin. Ultrathin sections were cut and examined ultrastructurally.

RESULTS

Papanicolaou stained specimens from pulmonary alveolar proteinosis patients contained scattered amorphous or granular globules, 20-50 microns in diameter, which were PAS positive. Ultrastructural examination of the globules revealed multilamellated structures, characteristic of pulmonary alveolar proteinosis, in all cases.

CONCLUSIONS

In general, it is thought that the morphological diagnosis of pulmonary alveolar proteinosis from bronchoalveolar lavage specimens requires both cytological and ultrastructural examination. However, the amorphous globules evident on cytology smears proved to contain multilamellated structures so that they can themselves be used as diagnostic evidence.