The following article requires a subscription:



(Format: HTML, PDF)

Background. Galactomannan is an Aspergillus-specific polysaccharide released during aspergillosis and is detected by the quantitative serum galactomannan index (GMI) test. Preclinical and preliminary clinical reports have suggested a good correlation between GMI and aspergillosis outcome.

Methods. We reviewed the literature to assess the strength of correlation between GMI and aspergillosis outcome using the [kappa] correlation coefficient. We included 27 studies that enrolled patients with hematological cancer and proven or probable aspergillosis and that used sequential GMI testing. We examined the 3 following outcomes: survival (survival vs. death), global outcome (survival vs. death [including autopsy findings]), and autopsy outcome (autopsy findings only).

Results. Overall, 257 patients fulfilled criteria for proven or probable aspergillosis and were eligible for outcome evaluation. Correlation between GMI (within <=1 week before outcome) and defined outcomes was excellent, with [kappa] correlation coefficients of 0.8737 (95% confidence interval, 0.8140-0.9333; P<.001) and 0.9123 (95% confidence interval, 0.8617-0.9629; P<.001) for survival and global outcome, respectively. Most importantly, the [kappa] correlation coefficient for autopsy outcome was high (0.8498; 95% confidence interval, 0.5608-1.000; P<.001). Furthermore, the [kappa] correlation coefficient for all outcomes was comparable across age groups (pediatric and adult patients) and treatment modalities, including allogeneic transplantation. This strong correlation is also supported by extensive preclinical data and recent clinical reports.

Conclusions. We conclude that serum GMI is a good marker of aspergillosis outcome.

(C) Copyright Oxford University Press 2008.