Article Text

S13 Pharmacokinetics and pharmacodynamics of antimicrobials in critically ill patients with lower respiratory tract infections. are ‘one size fits all’ doses appropriate?
  1. IB Oldfield1,
  2. K Kipper1,
  3. CI Barker1,
  4. BJ Philips1,
  5. M Cecconi2,
  6. A Rhodes2,
  7. A Johnston1,
  8. JF Standing3,
  9. EH Baker1,
  10. M Sharland1,
  11. DO Lonsdale1
  1. 1Institute for Infection and Immunity, St George’s, University of London, London, UK
  2. 2St George’s University Hospitals NHS Foundation Trust, London, UK
  3. 3Infectious Diseases and Microbiology Unit, University College London, Institute of Child Health, London, UK


Introduction Respiratory infection is a common cause of severe sepsis.1 Current therapeutic guidelines emphasise the importance of early initiation of antibiotic therapy, but make no recommendations on dose. Recent studies have suggested that some critically ill patients fail to achieve sufficient plasma antibiotic concentrations to treat infection effectively.2

We determined whether critically-ill patients with respiratory infection achieved pharmacokinetic/pharmacodynamic (PK/PD) targets during antibiotic treatment and investigated factors associated with failure to meet these targets.

Methods This was a subgroup, interim analysis of an ongoing study, ABDose. Participants were adults in intensive care receiving piperacillin-tazobactam or co-amoxiclav for respiratory infection. Demographics and measures of organ function were recorded. Antibiotic concentrations were measured, at steady-state, in plasma at 50% and 100% of the dosing interval. Efficacy of beta-lactam antibiotics is dependent upon time above minimum inhibitory concentration(MIC). We chose PK/PD targets of antibiotic concentration >MIC and a more conservative >4 × MIC of likely pathogen or microbiological isolate (when available). These targets have been used previously.2 During 28-day follow up, need for additional antibiotics was recorded.

Results 24 participants (median age 61, IQR [50–70] years), received co-amoxiclav (n = 7), piperacillin-tazobactam (n = 15) or both (n = 2). At 100% of the dosing interval, 12 achieved plasma antibiotic concentrations >MIC and 8 achieved >4×MIC. Participants who did not achieve PK/PD targets were younger (48 [39–59] years vs 68 [61–80] years, p = 0.002*) and had a higher eGFR (131 ± 58 mls/min/1.73m2 vs 64 ± 28 mls/min/1.73m2, p = 0.004*) than those who did. Antibiotic concentrations were correlated with age and negatively correlated with eGFR (Figure 1). All participants failing to achieve antibiotic concentrations >4 × MIC at 100% of the dosing interval required further courses of antibiotics during follow-up compared to 50% of patients achieving this target (p = 0.02*).

Conclusion In critically-ill patients with respiratory infection, uniform dosing of beta-lactam antibiotics does not consistently achieve PK/PD targets required for optimal efficacy. Younger patients, with better renal function may be under-dosed. These interim findings identify a need for further work to determine whether personalised dynamic dosing regimens could improve outcomes for patients with severe respiratory infection. Population PK modelling and further covariate analysis is planned within ABDose.


  1. Mayr FB, et al. Epidemiology of severe sepsis. Virulence 2014;5:4–11.

  2. Roberts JA, et al. DALI: defining antibiotic levels in intensive care unit patients: are current β-lactam antibiotic doses sufficient for critically ill patients?. Clin Infect Dis 2014;58:1072–83.

Abstract S13 Figure 1

Antimicrobial concentration measured at 50% of the dosing interval plotted against age (left) and eGFR (right). Line of best fir and associated coefficients suggest correlation with age and negative correlation with eGFR

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.