Screening with serum galactomannan (sGM) in haematological patients on micafungin prophylaxis: does the test remain useful for diagnosis of invasive aspergillosis?

Antonio Vena, Patricia Muñoz, Fritz Cajuste, Ana Alvarez-Uria, Roberto Alonso, Pablo Martín-Rabadán, Teresa Pelaez Garcia, Javier Gómez-Castellá, Maricela Valerio Minero, Emilio Bouza Santiago

Abstract: 

Background: Invasive aspergillosis (IA) remains a significant cause of morbidity and mortality in
hematological patients. In order to achieve an early diagnosis, systematic screening with sGM is
performed in many hematology departments. However, recent studies have shown that posaconazole
prophylaxis reduces very significantly the positive predictive value of GM (11%) (Duarte et al, 2014),
leading to many unnecessary CTs and antifungal therapies. Micafungin is also used for active antimould
prophylaxis, especially in patients with mucositis. Our aim was to investigate the diagnostic
performance of sGM assay in high risk hematology patients receiving micafungin prophylaxis.
Material/methods: Retrospective 5-year study including hematological patients who received
micafungin prophylaxis during high-risk episodes. Indication of sGM testing was classified as:
prompted by clinical suspicion (“diagnostic driven“) or twice weekly screening in afebrile patients
(“surveillance GM”). The clinical situation oft he patients was classified as follows: True positive-TP
(positive GM test results in the context of breakthrough IA), False positive-FP (positive GM test in
patients who remained on micafungin prophylaxis and had no evidence of IA), True negative-TN
(episodes with all consecutive GM tests negative and no diagnostic features of IA), False negative-FN
(negative GM test results in the context of breakthrough IA) and non-evaluable (patients received
antifungal treatment and IA could not be excluded or confirmed with certainty).
Results: During the study period, 149 patients (60.4% male) with a median age of 43 years (range,5-
76), underwent micafungin prophylaxis during 208 episodes at risk of IA (40 AML chemotherapy, 129
allogeneic HCT, and 39 GVHD). A total of 1617 serum GM tests were performed in these patients,
with a median of 8 GM tests per risk episode (range, 2–40). Of these, 1405 samples were negative,
and 212 serum GM tests (13.1%) were positive, corresponding to 54 episodes (27.9%). Overall, 62
(29.7%) episodes were not evaluable because patients received antifungal treatment for other fungal
infections (3 episodes) or due to persistent fever (59 episodes). Of the remaining 146 episodes, 4
corresponded to TP GM test results in the context of breakthrough probable IA (incidence of
breakthrough IA under micafungin prophylaxis 2.7%). Among the remaining episodes, 111 (76%) were
considered as TN and 31 (21.2%) as FP. No FN episodes were detected. All except one episode of
FP result occurred among surveillance GM tests, leading to high-resolution CT scans in 25% of them,
all of which were negative. PPV/NPV of GM test used in surveillance or diagnostic driven approaches
were 3.1% /100% and 75% and 100%, respectively.
Conclusions: GM surveillance of asymptomatic patients receiving micafungin prophylaxis is not
recommended, because all results will be either negative or false positive. However, sGM may still be
helpful for the diagnosis of breakthrough IA in symptomatic patients during prophylaxis.

2016

Poster: 

AttachmentSize
Image icon EV0759.png2 MB

abstract No: 

#3679

Full conference title: 

26th European Congress of Clinical Microbiology and Infectious Diseases
    • ECCMID 26th (2016)