Erratum: Microbiological Laboratory Testing in the Diagnosis of Fungal Infections in Pulmonary and Critical Care Practice. An Official American Thoracic Society Clinical Practice Guideline

There are errors in the ATS clinical practice guidelines (1) published in the September 1, 2019, issue of the Journal. Because of a problem with the processing of an Excel file, some incorrect values were included in the publication.

On page 540, third column, the negative predictive value for a single positive blood Aspergillus PCR test should read 98%, and the negative predictive value for two consecutive positive blood Aspergillus PCR tests should read 96%; the negative predictive value for a BAL Aspergillus PCR test should read 98%. Reflecting the corrected BAL test performance, the document should read that the test would be expected to yield 87 true positives, 847 true negatives, 13 false negatives, and 53 false positives.

On page 542, first column, the positive predictive value and negative predictive value for a (1→3)-β-d-glucan test with a positivity threshold of 80 pg/ml should read 19% and 97%, respectively. On page 543, third column, the negative predictive value for a Histoplasma antigen test should read 98% when assuming a prevalence of 10%, and should read 94% when assuming a prevalence of 25%. The yield would be 204 true positives and 46 false negative (instead of 46 false positives).

The authors of the document stress that none of these corrections alter any of the recommendations included in this ATS clinical practice guideline.