Impiego dell’indice itterico come esame di primo livello per l’identificazione dei campioni con concentrazioni anormali di bilirubinemia

Suitability of icteric index (II) as front-line test for the identification of blood samples with abnormal total bilirubin (TB) concentrations. The use of II as a front-line test for the preliminary identification of blood samples with abnormal TB concentrations was recently proposed. However, laboratories should validate this approach on their own analyzers. In this study we validated the diagnostic accuracy of II on the Abbott Architect c16000 platform. TB concentrations (diazo-based colorimetric assay) and corresponding II values (derived from absorbance measurements of samples diluted with saline) in heparinised plasma and serum samples were collected for a 3-month period. Linear regression analysis (LRA) (II vs. TB) was performed for both samples. The diagnostic performance of II to discriminate between abnormal (>1.2 mg/dL) and physiological TB concentrations was evaluated using the ROC curve analysis. The optimal II cut-off was selected at a negative predictive value (NPV) >99% for detection of abnormal TB values. TB and relative II were obtained from 18,486 serum and 3700 plasma samples. LRA showed a strong correlation between II and TB (serum: r2=0.951; plasma: r2=0.941). ROC curve analysis gave the following areas under the curve: serum, 0.948 (CI: 0.945-0.951), and plasma, 0.922 (CI: 0.913-0.930), showing the high accuracy of II for detecting abnormal TB on both sample types. An II ≤0.8 reliably excluded abnormal (>1.2 mg/dL) TB concentrations (serum, prevalence 25.4%: sensitivity 99.6%, NPV 99.7%; plasma, prevalence 16.7%: sensitivity 98.6%, NPV 99.4%). In our laboratory the use of an II value ≤0.8 as front-line test should allow the accurate “zero-cost” detection of samples with normal TB concentrations avoiding TB measurement in ∼35% of serum and ∼40% of plasma samples.