1048. Serum (1,3)-beta-D-glucan (BDG) Has A Good Negative Predictive Value For Candidemia In Patients With Hematologic Malignancies
Session: Poster Abstract Session: Fungal Diagnosis
Saturday, October 22, 2011
Room: Poster Hall B1
Background: The serum BDG assay is a marker of invasive fungal infections (IFIs) including candidemia with variable but overall good test performance. We evaluated the performance of serial BDG as a marker of candidemia among patients with hematological malignancies undergoing chemotherapy

Methods: A retrospective review of the medical records was performed on all patients with candidemia between June 2008 and September 2010. An episode referred to any 10-day period during which testing included ≥ 2 BDG and 2 ≥ blood cultures. Episodes with other known causes of elevated BDG were excluded. BDG level was determined by the Fungitell Assay (Associates of Cape Cod Inc., MA, USA). Two analyses were performed: one focused on episodes (episode analysis) and another included BDG obtained within 48 hours from blood cultures during all episodes (overall analysis). Statistical analyses included area under the curve of the receiver operating characteristics (AUC ROC) and the likelihood ratio (LR) using two BDG cut off values of ≥80 and ≥ 60 pg/mL. Fisher’s exact test was used for all analyses.

Results: Eighty episodes occurred in 44 patients; 45 episodes with candidemia and 35 control episodes in which patients with resolved candidemia served as their own controls. Over analysis included 420 BDG tests and 420 blood cultures. All patients received fluconazole prophylaxis. Candida spp. during the episodes were C. kruseii (39%), C. glabrata (29%), C. tropicalis (10%), C. parapsilosis (8%), C. albicans (8%), C lipolytica (3%), and C. dublinensis (1%), C. lusitanae (1%) and C. species (1%). Table 1 shows the performance of BDG assay.

Conclusion: In this patient population, a negative BDG at the cut-off point of ≥ 80 pg/mL is a useful test to exclude candidemia; however, a positive test should be interpreted with caution.

Table 1. BDG test performance as a marker of candidemia

Sensitivity

(%, 95% CI)

Specificity

(%,  95% CI)

PPV

(%, 95% CI)

NPV

(%, 95% CI)

ROC

LR+

LR-

Episode analysis

BDG > 60pg/ml

82

(0.67-0.91)

45

(0.29-0.63)

66%

(0.52-0.77)

66%

(0.44-0.83)

0.663

1.51

0.38

Episode analysis BDG> 80 pg/ml

69

(0.53-0.81)

57

(0.39-0.73)

67

(0.52-0.80)

59

(0.40-0.75)

0.631

1.0

0.5

Overall analysis

BDG> 80pg/ml

75

(0.64-0.84)

54

(0.49-0.59)

28

(0.22-0.34)

90

(0.85-0.94)

0.589

1.6

0.46


Subject Category: M. Mycology including clinical and basic studies of fungal infections

Zainab Shahid, M.D.1, Naveen Sanath kumar, M.D1, Lakshmikanth Katragadda, M.D.1, Alejandro Restrepo, M.D.1, Senu Apewokin, MD1, Monica Grazziutti, MD1, Marcio Nucci, MD2 and Elias Anaissie, MD1, (1)The Myeloma Institute for Research and Therapy/University of Arkansas for Medical Sciences, Little Rock, AR, (2)Department of Internal Medicine, Federal University of Rio de Janeiro, Rio de Janeiro - RJ , Brazil

Disclosures:

Z. Shahid, None

N. Sanath kumar, None

L. Katragadda, None

A. Restrepo, None

S. Apewokin, None

M. Grazziutti, None

M. Nucci, Pfizer: Board Member, Consultant, Grant Investigator, Scientific Advisor and Speaker's Bureau, Consulting fee, Educational grant, Research grant and Speaker honorarium
Merck: Board Member, Consultant, Grant Investigator and Scientific Advisor, Consulting fee, Educational grant and Speaker honorarium
Astellas: Board Member, Consultant and Scientific Advisor, Consulting fee and Speaker honorarium

E. Anaissie, None

See more of: Fungal Diagnosis
See more of: Poster Abstract Session

Findings in the abstracts are embargoed until 12:01 a.m. EST Thursday, Oct. 20 with the exception of research findings presented at IDSA press conferences.