Background: Bronchoalveolar lavage galactomannan testing (BAL-GM) is recommended for diagnosing invasive pulmonary aspergillosis in hematological malignancy patients. While antifungal therapy is known to reduce serum GM sensitivity, its impact on BAL-GM performance is less well-established. Objective: To determine whether antifungal therapy reduces the diagnostic performance of BAL-GM for invasive aspergillosis in patients with hematological malignancies. Methods: Pubmed, Embase, SCOPUS and Google Scholar were searched. Two investigators independently extracted data and assessed quality using QUADAS-2. Bayesian bivariate meta-analysis models were fitted using bamdit package in R to calculate pooled sensitivity, specificity, and AUC. Subgroup analyses compared diagnostic performance between patients with and without antifungal therapy. Results: 124 studies were included. BAL galactomannan showed sensitivity of 0.82 (97.5% CI: 0.76-0.87) and specificity of 0.90 (97.5% CI: 0.85-0.93), with an AUC of 0.67 (97.5% CI: 0.64-0.70). In patients without antifungal therapy, BAL galactomannan demonstrated comparable performance with sensitivity of 0.84 (97.5% CI: 0.34-1.0), specificity of 0.90 (97.5% CI: 0.82-0.96), and AUC of 0.70 (97.5% CI: 0.67-0.72). Discussion: Antifungal therapy does not substantially impair BAL galactomannan performance, with comparable diagnostic accuracy observed between treated and untreated populations. The substantial uncertainty identified indicates that the clinical utility of galactomannan testing during antifungal therapy remains an unresolved question. Overall, BAL galactomannan testing should not be abandoned in patients receiving antifungal therapy but requires careful clinical interpretation when negative.
Presupposti: Il test del galattomannano su liquido di lavaggio broncoalveolare (BAL-GM) è raccomandato per la diagnosi di aspergillosi polmonare invasiva nei pazienti con neoplasie ematologiche. Benché sia noto che la terapia antifungina riduca la sensibilità del GM sierico, l’impatto di tale terapia sulla performance del BAL-GM è meno chiaro. Obiettivi: Valutare se la terapia antifungina riduce l'utilità clinica del test BAL-GM nella diagnosi di aspergillosi invasiva nei pazienti con neoplasie ematologiche. Metodi: Sono stati consultati PubMed, Embase, SCOPUS e Google Scholar. Due ricercatori hanno estratto indipendentemente i dati e valutato la qualità degli studi utilizzando QUADAS-2. Sono stati applicati modelli bayesiani bivariati per la meta-analisi utilizzando il pacchetto bamdit in R per calcolare sensibilità, specificità e AUC aggregate. Le analisi per sottogruppi hanno confrontato le prestazioni diagnostiche tra pazienti sottoposti a terapia antifungina e non. Risultati: Sono stati inclusi 124 studi. GM-BAL ha mostrato una sensibilità di 0,82 (97.5% CI: 0,76-0,87) e una specificità di 0,90 (97.5% CI: 0,85-0,93), con un AUC di 0,67 (97.5% CI: 0,64-0,70). Nei pazienti non sottoposti a terapia antifungina, il GM-BAL ha dimostrato prestazioni comparabili con sensibilità di 0,84 (97.5% CI: 0,34-1,0), specificità di 0,90 (97.5% CI: 0,82-0,96) e AUC di 0,70 (97.5% CI: 0,67-0,72). Conclusioni: La terapia antifungina non compromette la performance diagnostica del test GM-BAL. La sostanziale incertezza identificata nello studio indica che l'utilità clinica del test del galattomannano durante la terapia antifungina rimane una questione irrisolta. Nel complesso emerge che non è opportuno abbandonare il GM-BAL nei pazienti che ricevono terapia antifungina, tuttavia un test negativo richiede un'attenta interpretazione clinica poiché non esclude la malattia.
Bayesian Meta-Analysis of Galactomannan Diagnostic Performance Across Clinical Scenarios
CENTONZE, MARCO
2024/2025
Abstract
Background: Bronchoalveolar lavage galactomannan testing (BAL-GM) is recommended for diagnosing invasive pulmonary aspergillosis in hematological malignancy patients. While antifungal therapy is known to reduce serum GM sensitivity, its impact on BAL-GM performance is less well-established. Objective: To determine whether antifungal therapy reduces the diagnostic performance of BAL-GM for invasive aspergillosis in patients with hematological malignancies. Methods: Pubmed, Embase, SCOPUS and Google Scholar were searched. Two investigators independently extracted data and assessed quality using QUADAS-2. Bayesian bivariate meta-analysis models were fitted using bamdit package in R to calculate pooled sensitivity, specificity, and AUC. Subgroup analyses compared diagnostic performance between patients with and without antifungal therapy. Results: 124 studies were included. BAL galactomannan showed sensitivity of 0.82 (97.5% CI: 0.76-0.87) and specificity of 0.90 (97.5% CI: 0.85-0.93), with an AUC of 0.67 (97.5% CI: 0.64-0.70). In patients without antifungal therapy, BAL galactomannan demonstrated comparable performance with sensitivity of 0.84 (97.5% CI: 0.34-1.0), specificity of 0.90 (97.5% CI: 0.82-0.96), and AUC of 0.70 (97.5% CI: 0.67-0.72). Discussion: Antifungal therapy does not substantially impair BAL galactomannan performance, with comparable diagnostic accuracy observed between treated and untreated populations. The substantial uncertainty identified indicates that the clinical utility of galactomannan testing during antifungal therapy remains an unresolved question. Overall, BAL galactomannan testing should not be abandoned in patients receiving antifungal therapy but requires careful clinical interpretation when negative.| File | Dimensione | Formato | |
|---|---|---|---|
|
Centonze_Marco_Tesi.pdf
Accesso riservato
Dimensione
4.98 MB
Formato
Adobe PDF
|
4.98 MB | Adobe PDF |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/93309