Extent of agreement in gentamicin concentration between serum that is drawn peripherally and from central venous...

At our institution, patients who receive once-daily dosing of gentamicin have serum concentrations determined 3 and 6 hours after dose administration. Patients with single-lumen central venous catheters have the 3-hour samples drawn peripherally. The objective of this study was to evaluate the extent of agreement between peripheral and central venous catheter serum gentamicin concentrations drawn 3 hours after dose administration.

In this prospective, observational study, patients provided both a peripheral and a central blood sample for determination of serum gentamicin concentration. The order of sampling (central venous catheter versus peripheral first) was randomized. Agreement was assessed by determination of the intraclass correlation coefficient and Bland-Altman analysis. The clinically acceptable targets for the lower limit of the intraclass correlation coefficient and Bland-Altman limits of agreement were defined a priori as >0.80 and +/-6%, respectively. Differences between the theoretical dose adjustments using the central venous catheter versus the peripheral sample result were described.

Forty-five pairs of samples were collected: 42 from single-lumen implantable central venous catheters (ports) and 3 from peripherally inserted central venous catheters. The intraclass correlation coefficient was 0.91. However, the Bland-Altman analysis resulted in a mean percentage difference (central venous catheter versus peripheral) of -0.92% and limits of agreement of -27.9% to 26.0%. The gentamicin dose adjustment based on the central venous catheter sample result would have led to clinically significant dose adjustments in 19 (42%) cases, when compared with the peripheral sample result.

These results indicate a lack of agreement between peripheral and single-lumen central venous catheter samples. In particular, ports are not appropriate sites for monitoring serum gentamicin concentrations.