Detection of galactomannan antigenemia in patients receiving piperacillin-tazobactam and correlations between in vitro, in vivo, and clinical properties of the drug-antigen interaction

Recent case reports describe patients receiving piperacillin-tazobactam who were found to have circulating galactomannan detected by the double sandwich enzyme-linked immunosorbent assay (ELISA) system, leading to the false presumption of invasive aspergillosis. Since this property of piperacillin-tazobactam and galactomannan ELISA is not well understood, we investigated the in vitro, in vivo, and clinical properties of this interaction. Among the 12 reconstituted antibiotics representing four classes of antibacterial compounds that are commonly used in immunocompromised patients, piperacillin-tazobactam expressed a distinctively high level of galactomannan antigen in vitro (P = 0.001). After intravenous infusion of piperacillin-tazobactam into rabbits, the serum galactomannan index (GMI) in vivo changed significantly (P = 0.0007) from a preinfusion mean baseline value of 0.27 to a mean GMI of 0.83 by 30 min to slowly decline to a mean GMI of 0.44 24 h later. Repeated administration of piperacillin-tazobactam over 7 days resulted in accumulation of circulating galactomannan to a mean peak GMI of 1.31 and a nadir of 0.53. Further studies revealed that the antigen reached a steady state by the third day of administration of piperacillin-tazobactam. Twenty-six hospitalized patients with no evidence of invasive aspergillosis who were receiving antibiotics and ten healthy blood bank donors were studied for expression of circulating galactomannan. Patients (n = 13) receiving piperacillin-tazobactam had significantly greater mean serum GMI values (0.74 +/- 0.14) compared to patients (n = 13) receiving other antibiotics (0.14 +/- 0.08) and compared to healthy blood bank donors (0.14 +/- 0.06) (P<0.001). Five (38.5%) of thirteen patients receiving piperacillin-tazobactam had serum GMI values >0.5 compared to none of thirteen subjects receiving other antibiotics (P = 0.039) and to none of ten healthy blood bank donors (P = 0.046). These data demonstrate that among antibiotics that are commonly used in immunocompromised patients, only piperacillin-tazobactam contains significant amounts of galactomannan antigen in vitro, that in animals receiving piperacillin-tazobactam circulating galactomannan antigen accumulates in vivo to significantly increased and sustained levels, and that some but not all patients receiving this antibiotic will demonstrate circulating galactomannan above the threshold considered positive for invasive aspergillosis by the recently licensed double sandwich ELISA.