Background Invasive pulmonary aspergillosis (IPA) is a severe fungal infection with high morbidity and mortality, primarily affecting immunocompromised individuals and patients with chronic underlying diseases. Current diagnostic methods, including histopathological examination, culture, and imaging studies, have significant limitations. Plasma Aspergillus -specific IgG antibody (Asp IgG) detection has recently gained attention for its high sensitivity in non-neutropenic patients and convenient blood sampling. This study aims to systematically evaluate the diagnostic value of plasma Asp IgG detection for IPA. Methods This single-center, retrospective, and controlled study included 167 non-neutropenic patients with suspected IPA from July 2022 to March 2025, divided into IPA (n = 61) and non-IPA groups (n = 106). Missing bronchoalveolar lavage fluid (BALF) galactomannan (GM) data were handled using multiple imputation with predictive mean matching. The diagnostic values of tests for Asp IgG, serum GM, and BALF GM were evaluated, along with the potentials of combined testing strategies. Results At a cut-off value of ≥80 AU/mL, Asp IgG showed higher sensitivity than serum GM and BALF GM (65.6% vs. 29.5% and 47.5%, P < 0.001), but lower specificity (62.3% vs. 97.0% and 95.3%, P < 0.001). A receiver operating characteristic curve showed that Asp IgG had an optimal diagnostic value when the cut-off value was 92.74 AU/mL, and the sensitivity and specificity were 63.9% and 70.8%, respectively. At the same cut-off value, combined Asp IgG and GM testing improved diagnostic efficacy, with the "OR" rule increasing sensitivity to 90.2% and the "AND" rule raising specificity to 98.1%. Acinetobacter baumannii infection was an independent risk factor for Asp IgG false positivity in the non-IPA group (OR = 3.929, P = 0.026). Conclusion Asp IgG testing exhibits high sensitivity for IPA diagnosis, making it suitable for non-neutropenic and critical patients. Combined testing strategies can optimize diagnostic performance based on clinical needs.
Ainiwaer et al. (Mon,) studied this question.