Accuracy of tobramycin levels obtained from central venous access devices in patients with cystic fibrosis is technique dependent.

Peter J. Mogayzel, Emily Pierce, Joanne Mills, Annette McNeil, Karen Loehr, Rebecca Joplin, Steve McMahan, Kathryn A. Carson

Research output: Contribution to journalArticlepeer-review

9 Scopus citations


OBJECTIVE: Patients with cystic fibrosis (CF) frequently require prolonged aminoglycoside antibiotic therapy administered via central venous access devices (CVADs). Therefore, the effect of the volume of saline flush used prior to drawing blood on tobramycin levels obtained from CVADs was investigated to determine the potential for inappropriate drug dosing decisions. PATIENTS AND METHODS: This was a prospective study comparing tobramycin levels in blood samples obtained simultaneously from CVADs and from venipuncture. Two saline flushing protocols were tested. During Phase I, CVADs were flushed with 3 mL of normal saline. The saline flush volume was increased to 10 to 20 mL during Phase II. All patients had CF and ranged in age from 10 months to 20 years. RESULTS: Twenty-eight paired peak levels were collected from 19 patients during Phase I of the study, and 30 paired samples were obtained from 20 patients using the revised collection protocol during Phase II. The median (range) peak tobramycin level obtained from CVADs using the original flushing protocol was 13.1 (3.4 to 52.0) compared to 9.5 (3.1 to 46.4) microg/mL for those drawn peripherally. Use of CVAD-obtained levels would have caused inappropriate dosing changes in 43% of cases. Increasing the saline flush volume improved the accuracy of the tobramycin levels obtained from CVADs. Median peak tobramycin levels using the modified flushing protocol were 17.3 (3.6 to 47.0) and 19.0 (3.3 to 42.1) microg/mL obtained from a CVAD or venipuncture, respectively. The intraclass correlation coefficient for this protocol was 0.90 (lower 95% confidence limit = 0.80). Additionally, 87% of these CVAD-obtained levels provided accurate information for clinical decision making. CONCLUSIONS: Peak tobramycin levels drawn from CVADs flushed with a small flush volume are falsely elevated compared to those obtained by peripheral venipuncture. The use of a larger saline flush greatly improves the accuracy of these levels. Therefore, care should be taken to base clinical decisions on tobramycin levels obtained by peripheral venipuncture or following an appropriate large volume saline flush.

Original languageEnglish (US)
Pages (from-to)464-468; quiz 468-469
JournalPediatric Nursing
Issue number6
StatePublished - Nov 2008

ASJC Scopus subject areas

  • Pediatrics


Dive into the research topics of 'Accuracy of tobramycin levels obtained from central venous access devices in patients with cystic fibrosis is technique dependent.'. Together they form a unique fingerprint.

Cite this