Secondary Logo

Institutional members access full text with Ovid®

Interference Testing of the Creatinine Sensor in the ABL837 FLEX Analyzer

Andersson, Ann-Christin MSc*; Strandberg, Karin MD, PhD*; Becker, Charlotte MD, PhD*; Hägglöf-Persson, Agnetha Biomed Tech*; Lundström, Gunilla Biomed Tech*; Thämlitz, Rolf Med Tech Eng*; Kristensen, Tina MSc; Skurup, Anne MSc

Point of Care: The Journal of Near-Patient Testing & Technology: June 2007 - Volume 6 - Issue 2 - p 139-143
doi: 10.1097/01.poc.0000271400.07312.0e
Original Articles

A newly developed creatinine sensor for the ABL837 FLEX blood gas analyzer (Radiometer Medical ApS, Copenhagen, Denmark) is based on the enzymatic conversion of creatinine by 3 enzymes. This conversion leads to the formation of hydrogen peroxide in the final stage, which is detected by an amperometric electrode. The results of an interference test of the creatinine sensor are presented as well as data on imprecision and bias.

The interference test was performed according to the Clinical and Laboratory Standards Institute (formerly National Committee for Clinical Laboratory Standards) protocol EP7-A2 on serum with a normal creatinine concentration of 50 to 100 μM (0.57-1.13 mg/dL). Fifty-four different endogenous and exogenous substances were tested. Serum samples were spiked with the substance tested (test sample) and were compared with a control sample without the substance. Each substance was analyzed 3 times, and the maximum allowable difference was defined to be 8 μM. The conclusion of the interference test is that no substance demonstrated a higher degree of interference than the allowable 8 μM, although interference from Intralipid at high concentration (10 g/L) cannot be excluded.

For only one analyte, d(+)-glucose, a significant difference existed between test and control sample (P < 0.05). Eight other substances demonstrated a higher SD for the test or control sample than 2.5 μM, which affected the difference between creatinine in control and test sample, although the differences did not in any case exceed 8 μM. The repeatability in the normal as well as in the pathological creatinine range was 1.1% to 1.5% (CV%) in whole blood (n = 45) and 0.9% to 1.1% in plasma (n = 80, duplicate samples for 20 days).

From the *Department of Clinical Chemistry, Lund University, Malmö University Hospital, Malmö, Sweden and †Radiometer Medical ApS, Copenhagen, Denmark.

Analyzer and consumables for this study were supplied by Radiometer Medical ApS (Copenhagen, Denmark).

Reprints: Karin Strandberg, MD, PhD, Department of Clinical Chemistry, Malmö University Hospital, SE-205 02 Malmö, Sweden (e-mail:

© 2007 Lippincott Williams & Wilkins, Inc.