Cytologic Evaluation of 200 Consecutive Bronchoalveolar Lavage Specimens with Grocott’s Methenamine Silver Stain and Correlation with Microbiologic Studies

Document Type


Publication Date


Publication Title

Journal of the American Society of Cytopathology



Cytologic examination and microbiologic testing of bronchoalveolar lavage (BAL) specimens are important in the diagnosis and treatment of pulmonary infections. We investigated the value of cytomorphologic findings, including Grocott’s methenamine silver (GMS) stain, relative to concurrent fungal cultures.

Materials and Methods

Blinded to clinical and microbiologic data, Papanicolaou- and GMS-stained slides from 200 consecutive BALs were examined by three authors. The number of high power fields (HPFs) containing fungal organisms and semi-quantitative assessment of proportions of epithelial and inflammatory cells were averaged and correlated with fungal culture results, which were considered the gold standard for the presence of fungus for statistical analysis.


When any of the three examiners identified fungal organisms, sensitivity of 56.7% and positive predictive value (PPV) of 47.2% were found. PPV increased to 61.0% when all agreed on fungal presence. A negative GMS by all examiners showed a specificity of 72.9% and negative predictive value (NPV) of 79.7%. Using different cutoffs for the number of HPFs with fungal organisms on smears (see Table 1), sensitivity of 52.7% (cutoff >=1) and PPV of 100% (cutoff >=20) were found. For cytospins, sensitivity of 100% (cutoff >=1) and PPV of 50% (cutoff >=20) were found. With different cutoffs for percentage of neutrophils (see Table 2) sensitivity of 56.7% (cutoff >=50%) and PPV of 39.5% (cutoff >=50%) were found. With different cutoffs for percentage of squamous cells (see Table 3), sensitivity of 20.0% (cutoff >=25%) and PPV of 57.1% (cutoff >=25%) were found.


When quantifying the number of HPFs with fungal organisms on GMS-stained smears, cutoff values of >=5, >=10, and >=20 HPFs showed high PPV (>90%) for subsequent positive fungal culture; however, with lower sensitivity, specificity, and NPV. Routine GMS staining on all BAL specimens is likely not warranted given time, cost, and labor considerations.





First Page