Clinical utility of the bronchial washing-to-serum galactomannan antigen ratio for the diagnosis of pulmonary aspergillosis.

INTRODUCTION

The galactomannan (GM) antigen test using bronchial washing samples (BW) is useful for diagnosing pulmonary aspergillosis (PA); however, the reported optimal cut-off level varies, and the levels are affected by serum GM antigen levels. Thus, we hypothesized that the BW-to-serum GM antigen ratio, particularly a ratio > 1.0 (indicating higher BW levels than serum levels), is useful for diagnosing PA.

METHODS

This retrospective study enrolled 53 patients with suspected PA, including chronic necrotizing PA, aspergilloma, and allergic bronchopulmonary aspergillosis, whose GM antigen levels in BW and serum were measured from April 2013 to August 2023 at Hiroshima University Hospital.

RESULTS

The ratio was significantly higher in patients with PA than in those without (median 3.89, interquartile range [IQR] 1.65-15.6 vs. median 0.58, IQR 0.24-0.96). The area under the curve (AUC) analyzed using the ratio for diagnosing PA was 0.926 (95%CI, 0.821-0.971). The prevalence of PA was higher in patients with a ratio > 1.0 than in those with a ratio < 1.0 (69.2% vs. 3.7%, respectively). Subgroup analyses revealed that the AUC of the ratio was higher in patients with serum GM antigen levels >0.5 than in those with levels <0.5 [0.993 (95%CI, 0.898-0.999) vs. 0.869 (95%CI, 0.665-0.957)]. In each patient group with serum GM antigen levels higher and lower than 0.5, a ratio of 1.0 could stratify the prevalence of PA (87.5% vs. 3.7% and 61.1% vs. 0.0%, respectively).

CONCLUSION

The BW-to-serum GM antigen ratio, particularly the cut-off level of 1.0, could aid in diagnosing PA.