Speaker
Description
Hemolysis is a common preanalytical source of interference that affects laboratory test results, particularly potassium (K) and lactate dehydrogenase (LDH) measurements [1,2]. This study investigated the impact of hemolysis on K and LDH results in routine laboratory practice and evaluated the performance of a hemolysis index (HI)-based correction method [3].
A retrospective analysis was performed using anonymized data generated by a Roche Cobas analyzer. Potassium, LDH, sodium (Na), and corresponding HI values were evaluated. Linear regression analyzed was used to assess the relationship between hemolysis and analyte concentrations. Regression-based correction models were subsequently applied to estimate and reduce hemolysis-induced analytical bias. The correction models were derived from routine clinical samples and therefore reflect real-world laboratory conditions.
A total of 111 serum samples were included in the study. Higher HI values were significantly associated with increased potassium concentrations (R² = 0.224, p < 0.001) and LDH activity (R² = 0.088, p = 0.002), whereas sodium showed no significant correlation with HI (R² = 0.002, p = 0.623). Potassium concentrations differed significantly across hemolysis categories (ANOVA, p < 0.001), increasing from 4.03 mmol/L at HI 0–10 to 4.63 mmol/L at HI >60. LDH activity increased from 279.9 to 402.6 U/L across the same categories. Application of the correction models substantially reduced these differences, with potassium differences decreasing from 0.60 to 0.04 mmol/L and LDH differences from 122.7 to 4.0 U/L.
These findings confirm the significant impact of hemolysis on potassium and LDH measurements and indicate that locally derived HI-based correction models may improve the interpretation of hemolyzed samples in routine laboratory practice.
Keywords: hemolysis, hemolysis index, potassium, LDH, analytical interference, clinical biochemistry, regression analysis, correction model
References:
1. Lippi G, Plebani M. Clin Chem Lab Med. 2012;50(3):393–401.
2. Koseoglu M, Hur A, Atay A, Cuhadar S. Biochem Med (Zagreb). 2011;21(1):79–85.
3. Lippi G, Cadamuro J, von Meyer A, Simundic AM. Clin Chem Lab Med. 2018;56(5):718–727.