Beta-lactam carryover in arterial and central venous catheters is negligible.

Therapeutic drug monitoring is used for aminoglycosides and vancomycin, and has been proposed for β-lactam antibiotics. Clinical blood samples in the ICU are often obtained via an existing vascular catheter rather than fresh needle phlebotomy. If antibiotics had previously been infused through a vascular catheter then used for blood sampling, carryover of antibiotic from the infusion to the sample might result in misleading assessments of target attainment. To address this concern we conducted a series of in vitro measurements of carryover for three commonly used antibiotics. We infused piperacillin-tazobactam, meropenem, and cefepime at pharmacologic concentrations through commonly used vascular catheters at our hospital and flushed the catheters. We then aspirated warmed citrated bovine blood through each catheter and measured antibiotic concentrations in each aspirate. Carryover was below the limits of detection for piperacillin-tazobactam, meropenem, and vancomycin. Cefepime carryover, in contrast, was not negligible and needs to be investigated more fully. Carryover from prior infusions does not appear to jeopardize measurements of piperacillin-tazobactam, meropenem, or vancomycin in commonly used vascular catheters at our institution. Caution in interpreting samples obtained for cefepime measurements appears advised until more data is available.