A study of prognostic factors affecting the incidence and outcome of invasive fungal infections occurring in neutropenic and allogeneic transplant patients receiving prior antifungal agents

J. De la Serna (1), J. Lopez (2), I. Jarque (3), J.M. Moraleda (4), M. Cuétara (5), A. Garcí­a (6), I. Olazabal (7), L. Yánez (8), P. Martí­nez (1)

Author address: 

(1)Hospital 12 de Octubre (Madrid, ES); (2)Hospital Ramon y Cajal (Madrid, ES); (3)Hospital La Fe (Valencia, ES); (4)Hospital V Arrixaca (Murcia, ES); (5)Hospital Severo Ochoa (Madrid, ES); (6)Hospital de la Princesa (Madrid, ES); (7)Hospital


Mold-active antifungal agents are useful for the prevention or treatment of suspected fungal infections in patients with hematologic malignancies. However they may fail due to breakthrough invasive fungal infections (B-IFI). Several host and antifungal related factors may underlie this lack of effi cacy. In this retrospective study we aimed to identify the risk factors for incidence and outcome of B-IFI. Defi nitions were based in 2008 EORTC/MSG criteria. Outcome was recorded by response and survival at the end of antifungal treatment and after 90 days. We recorded 62 consecutive cases of B-IFI in neutropenic or allogeneic transplant patients who were receiving mold-active antifungal agents and 122 controls matched for disease and type of treatment from the same institutions. Their main characteristics are shown in Table 1. Cases had a higher median age (p=0.009). Disease diagnosis, status and allogeneic transplants were well balanced. More cases had acute GVHD (p=0.039). The use of antifungal agents was not uniform. Prophylaxis was azol-based in 62,2% and 88,5% of cases and controls, respectively. Empiric therapy was based in equinocandins in 56,9% in both groups, and pre-emptive therapy equally distributed (50%) between liposomal amphotericin B and azol. There were 37 (59.7%) and 25 (40.3%) cases of probable and proven B-IFI. Overall, 29 (46.8%), 22 (35.5%) and 11 (17.7%) cases occurred after prophylaxis, empiric and pre-emptive antifungal treatment. Long-lasting neutropenia (p=0.0002) and acute GVHD (p=0.039) were associated with B-IFI. Proven B-IFI were due to yeast species, fusarium and aspergillus in 13 (52%), 5 (20%) and 4 (16%) cases, respectively. Multivariate analysis showed that the probability of B-IFI was lower in patients receiving mold-active antifungals for prophylaxis or empiric therapy than in the pre-emptive setting (OR 0.07) and refractory AML (OR 4.95) and acute GVHD (OR 4.95) were main risk factors. B-IFI patients were treated with more than one antifungal in 37 cases (59,7%). Overall, 35 (56,5%) and 27 (43.5%) cases had a favourable outcome at the end of treatment and at 90 days. Survival at 90 days was 48.3%. Both advanced age and refractory AML are predictors of death in patients with B-IFI (p=0.02 and 0.05) respectively. Conclusion: This study shows that the pattern of breakthough IFI is moving to yeasts and the higher risks are in the more advanced use of prior antifungals, long lasting neutropenia and acute GVHD.

abstract No: 


Full conference title: 

Annual Meeting of the EBMT, 37th
    • EBMT 37th (2011)