High mobility group box-1 protein in patients with suspected community-acquired infections and sepsis: a prospective study

INTRODUCTION Sepsis is a serious condition with a significant morbidity and mortality. New insight into the immunopathogenesis of sepsis could promote the development of new strategies for diagnosis and therapy. High mobility group box-1 protein (HMGB1) has been known for many years as a nuclear chromosomal protein. Its role as a pro-inflammatory cytokine in sepsis and rheumatoid arthritis has been described recently. The aim of our study was to evaluate HMGB1 as a molecular marker in patients with community-acquired infections. METHODS Patients suspected of having infections/sepsis and admitted to a department of internal medicine were included in the study in a prospective manner. Demographic data, comorbidity, routine biochemistry, microbiological data, infection focus, severity score, and mortality on day 28 were recorded. Plasma and serum were sampled at the time of admission. HMGB1 levels were measured with a commercially available enzyme-linked immunosorbent assay (ELISA). Procalcitonin levels were measured with a TRACE (time-resolved amplified cryptate emission) assay. Lipopolysaccharide-binding protein and interleukin-6 were measured with a chemiluminiscent immunometric assay. Soluble haemoglobin scavenger receptor (sCD163) levels were measured with an in-house ELISA. RESULTS One hundred and ninety-four patients were included in the study. Levels of HMGB1 are presented as medians and interquartile ranges: healthy controls (0.77 ng/ml, 0.6 to 1.46), non-infected patients (1.54 ng/ml, 0.79 to 2.88), infected patients without systemic inflammatory response syndrome (2.41 ng/ml, 0.63 to 3.44), patients with sepsis (2.24 ng/ml, 1.30 to 3.75), and patients with severe sepsis (2.18 ng/ml, 0.91 to 3.85). In a receiver operator characteristic curve analysis discriminating between non-infected patients and all infected patients, the area under the curve for HMGB1 was 0.59 (P < 0.0001). HMGB1 correlated only weakly to levels of white blood cell count, neutrophils, C-reactive protein, interleukin-6, procalcitonin, and lipopolysaccharide-binding protein (P < 0.001). HMGB1 did not correlate to sCD163. CONCLUSION In a cohort of patients with suspected community-acquired infections and sepsis, HMGB1 levels were statistically significantly higher in patients compared to the healthy controls. There was no statistically significant difference between the infected and the non-infected patients. Levels of HMGB1 correlated only very weakly to other pro-inflammatory markers and did not correlate to the anti-inflammatory marker sCD163.