639. Ex-Vivo Evidence of the ‘Skip Well’ Phenomenon in Patients Receiving Polymyxin B (PB)
Session: Poster Abstract Session: Pharmacokinetics and Adverse Drug Reactions
Friday, October 21, 2011
Room: Poster Hall B1
Background:  PB is used to treat infections due to multidrug-resistant (MDR) Gram negative pathogens. A “skip well” phenomenon, growth in higher after no growth in lower concentrations, has been observed during in vitro PB testing. No clinical correlate to this phenomenon has been described.

Methods:  Clinical data and peak and trough plasma samples were obtained from consenting patients receiving PB therapy. Plasma bactericidal activity was determined against 2 prototypical MDR pathogens (Acinetobacter baumannii and blaKPC+-Klebsiella pneumoniae) and the patients’ clinical isolates (PB MICs 0.25-2 µg/ml). For comparison, bactericidal activity was also determined using known concentrations of PB in control plasma.

Results: Nine patients were enrolled. All received from 1-4 additional antibiotics that generally lacked activity against the clinical and prototypical strains. Peak plasma samples were bactericidal at dilutions ranging from 1:1 to 1:128 (median 1:16) and trough plasma samples from <1:1 to 1:64 (median 1:8). However, growth in lower dilutions was seen in peak and/or trough plasma samples from 7 of 9 patients. Most isolates recovered in these wells had a twofold increase in the PB MIC. The skip well phenomenon was not seen in plasma from the 2 patients who received rifampin vs. all 7 of the other patients (P=0.03). Against both prototypical test strains, control plasma was bactericidal in dilutions down to ~ 0.25 µg/ml of PB. The addition of fixed concentrations (10 µg/ml) of meropenem and/or vancomycin (which were administered with PB in most patients) did not change bactericidal activity in the control samples. Four of the 9 patients were judged to have had a clinical failure after 14 days of therapy. The outcomes did not correlate with plasma bactericidal activity, dose of PB administered or with renal function.

Conclusion: Plasma bactericidal activity from patients receiving PB did not correlate with the dose or underlying renal function. The “skip well” phenomenon was frequently observed and may provide a window of opportunity for the emergence of resistance and contribute to poor clinical outcomes. The clinical significance of this phenomenon and the potential to prevent it with rifampin therapy deserve further study.

Subject Category: A. Antimicrobial agents and Resistance

Purba Gupta, MD1, Martin Bäcker, MD2, Monica Ghitan, MD1, Khawar Khurshid, MD1, Manali Pednekar, MD1, Farah Chinwalla, DO1, Roopali Sharma, PharmD2, Jerome Salvani, MD2, David Landman, MD2 and John Quale, MD2, (1)Maimonides Med Ctr, Brooklyn, NY, (2)SUNY Downstate Medical Center, Brooklyn, NY


P. Gupta, None

M. Bäcker, None

M. Ghitan, None

K. Khurshid, None

M. Pednekar, None

F. Chinwalla, None

R. Sharma, None

J. Salvani, None

D. Landman, None

J. Quale, None

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.