Cancer Diagnosis Program | Biorepositories and Biospecimen Research Branch

Search BRD Papers

The influence of hemolysis in patient samples on biochemical tests analyzed using Roche Cobas® 8000 Analyzer.

Author(s): Hung YE, Chiu YI, Lin YC, Shiesh SC, Cheng CL, Hsueh KY, Lin WL

Publication: Ann Clin Biochem, 2025, Vol. , Page 45632251356827

PubMed ID: 40551378 PubMed Review Paper? No

Purpose of Paper

Conclusion of Paper

Studies

1. Study Purpose

   This study compared levels of 23 clinical chemistry analytes in case-matched hemolyzed and non-hemolyzed blood specimens that were retrospectively collected. This study included 678 lithium heparin blood specimens that were rejected from clinical analysis due to hemolysis (hemolysis index (HI)  ≥15) and case-matched subsequent blood specimens that were collected within 4 h of the rejected specimen (diagnosis not specified). All specimens were collected in lithium heparin gel separator tubes and transported to the laboratory within 1 h of blood collection. Plasma was separated (details not provided). Icterus (480/505 nm), lipemia (660/700 nm) and hemolysis (570/600 nm) were assessed by a spectrophotometer.  All specimens had an icteric index of ≤4 and lipemic index ≤150.  Hemolysis was characterized as mild if the hemolysis index was between 15-100, moderate if the hemolysis index was 101-300 or heavy if the hemolysis index was 301-1000. Levels of albumin, alanine aminotransferase (ALT), alkaline phosphatase (ALP), ammonia, α-amylase, aspartate aminotransferase (AST), direct bilirubin, total bilirubin blood urea nitrogen, calcium, creatine kinase (CK), chloride, creatinine, ɣ glutamyltransferase (GGT), iron, lactate dehydrogenase (LDH), lipase, magnesium, phosphate, potassium, sodium, total protein, and uric acid were quantified using a Roche Cobas 8000 Analyzer. For each analyte, the bias was calculated by subtracting the value of the non-hemolyzed specimen from that of the hemolyzed specimen. The reference change value (RCV) was calculated based on analytical and biological variation. A difference was considered to be clinical significant when the bias exceeded 10% and the RCV.

   Summary of Findings:

   Of the 678 hemolyzed specimens, 25 (3.7%) specimens were classified with mild hemolysis (HI of 15–100), 490 (72.3%) with moderate hemolysis (HI of 101–300) and the remaining 163 (24%) with heavy hemolysis (HI of 301–1000). The lipemia index was significantly higher in case-matched specimens with a hemolysis index of 301-1000 or 101-300 than those with an HI<15 (P<0.001, both) despite no visible turbidity, which the authors state is indicative that hemolysis may falsely elevate the lipemia index.  The mean positive biases in ammonia, AST, creatine CK, LDH, potassium and negative biases in direct bilirubin, and GGT were considered clinically significant, as they exceeded the RCV and 10% bias threshold in hemolyzed specimens compared to non-hemolyzed specimens.  The bias in AST, LDH and potassium exceeded the RCV in blood specimens with mild, moderate, and heavy hemolysis (HI ≥15). The negative bias in direct bilirubin exceeded the RCV when hemolysis was moderate or heavy (HI ≥100). The biases in ammonia, CK, and GGT exceeded the RCV when hemolysis was heavy (HI ≥300), but ammonia was not assessed in specimens with mild or moderate hemolysis (HI 15-100). In linear regression plots, a significant correlation between the percentage bias and the hemolysis index was noted for AST (R²=0.3814, P<0.001), direct bilirubin (R²=0.0764, P<0.05), CK (R²=0.1530, P<0.001), LDH (R²=0.6524, P<0.001), potassium (R²=0.5630, P<0.001), and sodium (R²=0.5414, P<0.001). While not clinically significant, the bias in sodium exceeded the RCV; the biases in alanine aminotransferase (ALT), iron, and phosphorus exceeded the 10% threshold; and the biases in alkaline phosphatase (ALP), α-amylase, total bilirubin, magnesium, and total protein exceeded the manufacturer’s acceptance criteria threshold for a HI <1000. The bias in albumin, blood urea nitrogen, calcium, chloride, creatinine, lipase, and uric acid remained within the RCV, 10% threshold and the manufacturers’ acceptance criteria regardless of the degree of specimen hemolysis. 

   Biospecimens

   • Bodily Fluid - Plasma
   • Bodily Fluid - Blood

   Preservative Types

   • None (Fresh)

   Diagnoses:

   • Not specified

   Platform:

   Analyte	Technology Platform
   Electrolyte/Metal	Clinical chemistry/auto analyzer
   Protein	Clinical chemistry/auto analyzer
   Lipid	Spectrophotometry
   Protein	Spectrophotometry
   Small molecule	Clinical chemistry/auto analyzer

   Pre-analytical Factors:

   Classification	Pre-analytical Factor	Value(s)
   Biospecimen Aliquots and Components	Hemolysis	A range of hemolysis scores investigated

   View more