A study of biomakers for diagnosis, outcome prediction and antibiotic theraphy guidance in sepsis

Abstract

Sepsis is common in the ICU worldwide and contributes to high mortality. However, timely diagnosis, outcome prediction and antibiotic monitoring in sepsis remains challenging. In Chapter Three, the diagnostic value of model-based insulin sensitivity (SI) for sepsis was studied in 38 non-diabetics on their ICU admission in a cross-sectional study. The findings indicated that baseline SI was significantly lower in sepsis (n = 18) versus non-sepsis (n = 20) (0.996 ± 1.269 versus 5.012 ± 4.930 × 10-4 L/mU/min, P = 0.002), with clinically valid diagnostic performance (AUC 0.814). In Chapter Four, similar methodology was applied to a mixed cohort of 86 diabetic and non-diabetic patients newly admitted to ICU. Although baseline SI was significantly lower in sepsis (n = 41) versus non-sepsis (n = 45) (0.560 ± 0.676 versus 1.097 ± 1.473 × 10-4 L/mU/min, P = 0.037), the biomarker failed to diagnose sepsis in this cohort. Hence, model-based SI may be a useful diagnostic test of sepsis when specifically applied to the non-diabetic ICU patients. In Chapter Five, the prognostic value of a combination of biomarkers in sepsis was explored in a prospective cohort study of 159 ICU patients. It was found that a prediction equation utilizing baseline total leukocytes count, procalcitonin, interleukin-6 and arylesterase activity of paraoxonase-1 predicted 30-day mortality with a remarkable performance (AUC 0.814). Therefore, a multi-marker approach using these biomarkers may be a useful predictor of mortality in sepsis. In Chapter Six, the utility of point-of-care procalcitonin (POCT) to guide duration of antibiotic in the ICU was examined in a randomized-controlled trial. Eighty patients were allocated to either the POCT-guided arm (n = 40) or control arm (n = 40). The mean duration of antibiotic was 6.3 ± 2.1 days in the POCT-guided arm versus 9.1 ± 4.7 days in control arm (P = 0.001), while there was no significant difference in 30-day mortality. Thus, POCT guidance reduced antibiotic duration without compromising mortality in our patients.