Distinction between bacterial and viral infections by serum measurement of human neutrophil lipocalin (HNL) and the impact of antibody selection

• Published in: Journal of immunological methods Vol. 432; pp. 82 - 86
• Main Authors: Yu, Zhang, Jing, Huang, Hongtao, Pan, Furong, Jia, Yuting, Jin, Xu, Shengyuan, Venge, Per
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 01.05.2016
• Subjects: Acute Infection; Adult; Aged; Antibodies, Monoclonal - immunology; Area Under Curve; Bacteria; Bacterial Infections - blood; Bacterial Infections - diagnosis; Bacterial Infections - microbiology; Biomarkers - blood; Blotting, Western; Case-Control Studies; CRP; Diagnosis; Diagnosis, Differential; Enzyme-Linked Immunosorbent Assay; Epitopes; Female; Humans; Lipocalin; Lipocalins - blood; Lipocalins - immunology; Male; Middle Aged; Neutrophil; Predictive Value of Tests; Protein Multimerization; ROC Curve; Up-Regulation; Virus Diseases - blood; Virus Diseases - diagnosis; Virus Diseases - virology; Young Adult; Acute Infection; CRP; Diagnosis; Neutrophil; Lipocalin
• ISSN: 0022-1759; 1872-7905; 1872-7905
• DOI: 10.1016/j.jim.2016.02.014

The distinction between acute infections of bacterial or viral causes is clinically important, but often very difficult even for experienced doctors. Previous studies indicated that serum measurements of HNL (Human Neutrophil Lipocalin) might be a superior diagnostic means in this regard, but also indicated that the antibody conformation of the HNL assay might have an impact on the diagnostic performance. The aim of the present report was to examine this further. Several different (n=24) HNL ELISA assays were developed using different combinations of monoclonal and polyclonal HNL antibodies. Sera were collected from healthy persons (n=188) and from 155 patients with acute infections before any antibiotics treatment. The patients were diagnosed as having bacterial (n=69) or viral causes (n=86) of their infections. Plasma and serum were also examined by Western blotting using HNL-specific polyclonal antibodies. The optimal assay format for the distinction between bacterial and viral infection resulted in an area under the receiver operating characteristics curve (AuROC) for S-HNL of 0.98. (95% CI 0.94–1.00) as compared to 0.83 (0.76–0.88) for blood neutrophil counts and 0.69 (0.61–0.76) for S-CRP. Results also showed that different assay formats of HNL identified monomeric and dimeric HNL differently, the monomeric HNL being elevated in viral infections and the dimeric HNL being elevated in bacterial infections. We conclude that serum measurement of HNL is a superior diagnostic means to distinguish between acute infections caused by bacteria or virus. For optimal clinical performance the immunoassay should address conformational epitopes in the dimeric HNL. •HNL (human neutrophil lipocalin) is present in blood in several different forms with different cellular origins•The secretion of HNL from neutrophils is increased during bacterial infections•Serum levels of HNL discriminate with high accuracy between patients infected by bacteria or virus•For the optimal diagnostic discrimination the antibody configuration of the immunoassay should target the dimeric form of HNL