Impact of bronchoalveolar lavage galactomannan on the outcome of patients at risk for invasive pulmonary aspergillosis

Abstract

Background: Invasive pulmonary aspergillosis (IPA) is an important cause of morbidity and mortality among immunocompromised patients especially in neutropenic and patients treated with immunosuppressive drugs. New diagnostic tools have been developed to improve treatment and outcome. Compared with serum galactomannan, bronchoalveolar lavage galactomannan (BAL GM) detection has higher sensitivity (81% vs. 71%) and comparable specificity (87.6% vs. 89%). No study has correlated this test result to clinical outcome. Material and Method: A prospective non-randomised study was conducted from March to December 2008 in adult patients who were suspected to have invasive pulmonary aspergillosis (IPA). Serum galactomannan levels were measured and bronchoscopy was performed to obtained BAL fluid for direct examination, culture, and measurement of galactomannan level. Response to treatment and mortality within 6-weeks of follow-up were compared between positive and negative BAL GM groups. Factors influencing outcome were also analysed. Results: There were 30 patients with 3 probable, 11 possible and 17 no IPA. Other causative organisms can be identified in 8 of 17 patients in the no IPA group. Overall, BAL GM at the 0.5 cut-off yielded a 46% positive result compared with 13% of serum GM (p = 0.005). There was no significant difference in positive result between BAL GM at 1.0 cut-off and serum GM. By using BAL GM as a mycological criteria, 54% of possible IPA was upgraded to probable IPA. Neither BAL GM nor serum GM results were associated with clinical response and mortality. Recovery of neutropenia was the only factor associated with response to treatment and outcome (p = 0.003). Conclusion: BAL GM detection has a higher positive rate than serum GM in patients at risk for IPA. It is helpful in diagnosis and categorization of IPA, but its impact on clinical outcome cannot be demonstrated in this study.