Heparin interference in whole blood sodium measurements in a pediatric setting Paul M. Yip a,b, ⁎ , Man Khun Chan a , Nuala Zielinski a , Khosrow Adeli a,b a Division of Clinical Biochemistry, The Hospital for Sick Children, Toronto, ON, Canada M5G 1X8 b Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, ON, Canada M5G 1L5 Received 20 July 2005; received in revised form 30 November 2005; accepted 5 December 2005 Available online 26 January 2006 Abstract Objectives: In a pediatric setting, the incomplete filling of heparinized syringes is not an uncommon occurrence and has led to reports of falsely low hyponatremia in our institution. Little is known about heparin interference on sodium determination in whole blood, and our study aimed to investigate this interference due to excessive concentrations of heparin in pediatric specimens. Design and methods: Three different types of syringes were filled with various amounts of blood to mimic greater than normal concentrations of heparin. Specimens were analyzed on an ABL 725 blood gas analyzer, and corresponding plasma fractions were analyzed on a VITROS 950 chemistry system. In a separate study, paired patient samples consisting of a capillary tube and microtainer clot were similarly analyzed. Results: The presence of lithium heparin at 100 units/mL in blood caused a significant negative bias of 2–3 mmol/L in sodium concentration with the ABL 725, but no significant bias occurred when the corresponding plasma fraction was analyzed on the VITROS 950. For syringes containing electrolyte-balanced heparin, a similar negative bias was observed for blood but was not significant. Capillary tubes contained high concentrations of heparin (N100 units/mL) even when completely filled. Sodium results from capillary samples averaged 3.4 mmol/L lower than the corresponding serum values. These effects were independent of the sodium concentration across a wide range. Conclusions: Small blood volumes collected with heparinized sampling devices in pediatric samples lead to excess heparin that may significantly affect sodium determinations and spur false reports of critical hyponatremia. © 2006 The Canadian Society of Clinical Chemists. All rights reserved. Keywords: Heparin; Sodium; Hyponatremia; Pediatrics; Syringes; Blood specimen collection; Blood gas analysis Introduction Heparin continues to be a highly reliable anticoagulant used for whole blood anticoagulation despite a long history of pre- analytical interferences due to dilution error, ion binding (notably calcium) and electrolyte distortion due to the release of bound ions [1–3]. While the current usage of dry heparin has virtually eliminated errors arising from sample dilution, the availability of various new preparations of heparin intended to overcome such problems has been met with mixed success [4]. Although its ability to interfere with ionized calcium measure- ments has been recognized, lithium heparin remains widely used as it is generally less expensive than electrolyte-balanced preparations of heparin. Little attention has been given to interference by heparin on sodium determinations, particularly in whole blood. Previous reports have noted that, with concentrations of heparin at 100 U/ mL, clinically significant negative biases are observed [5,6]. Con- cerns over the interference in ionized calcium determinations have greatly overshadowed apparent hyponatremia in cases of excessive heparin. Modest levels of heparin have been found to interfere with ionized calcium leading to proposals that final concentrations should not exceed 10–15 U/mL for lithium heparin [1,3,4], and several manufacturers now supply preparations which contain less heparin. In pediatric settings, however, the risk of sample clotting is much higher than the general population, and the use of collection devices with reduced heparin may not be practical. Blood gas analysis plays an important role in the critical care setting where fast turnaround time is required. Since small sample Clinical Biochemistry 39 (2006) 391 – 395 ⁎ Corresponding author. Toronto Medical Laboratories, University Health Network, 200 Elizabeth Street, Eaton 3-412, Toronto, ON, Canada M5G 2C4. Fax: +1 416 586 1426. E-mail address: paul.yip@utoronto.ca (P.M. Yip). 0009-9120/$ - see front matter © 2006 The Canadian Society of Clinical Chemists. All rights reserved. doi:10.1016/j.clinbiochem.2005.12.006