Evaluation of New Diagnostic Biomarkers in Pediatric Sepsis: Matrix Metalloproteinase-9, Tissue Inhibitor of Metalloproteinase-1, Mid-Regional Pro-Atrial Natriuretic Peptide, and Adipocyte Fatty-Acid Binding Protein

Elevated plasma concentrations of matrix metalloproteinase-9 (MMP-9), tissue inhibitor of metalloproteinase-1 (TIMP-1), mid-regional pro-atrial natriuretic peptide (mrProANP), and adipocyte fatty-acid-binding proteins (A-FaBPs) have been investigated as biomarkers for sepsis or detection of acute neurological injuries in adults, but not children. We carried out a single-center, prospective observational study to determine if these measures could serve as biomarkers to identify children with sepsis. A secondary aim was to determine if these biomarkers could identify children with neurologic complications of sepsis. A total of 90 patients <= 18 years-old were included in this study. 30 with severe sepsis or septic shock were compared to 30 age-matched febrile and 30 age-matched healthy controls. Serial measurements of each biomarker were obtained, beginning on day 1 of ICU admission. In septic patients, MMP9-/TIMP-1 ratios (Median, IQR, n) were reduced on day 1 (0.024, 0.004-0.174, 13), day 2 (0.020, 0.002-0.109, 10), and day 3 (0.018, 0.003-0.058, 23) compared with febrile (0.705, 0.187-1.778, 22) and healthy (0.7, 0.4-1.2, 29) (p< 0.05) controls. A-FaBP and mrProANP (Median, IQR ng/mL, n) were elevated in septic patients compared to control groups on first 2 days after admission to the PICU (p < 0.05). The area under the curve (AUC) for MMP-9/TIMP-1 ratio, mrProANP, and A-FaBP to distinguish septic patients from healthy controls were 0.96, 0.99, and 0.76, respectively. MMP-9/TIMP-1 ratio was inversely and mrProANP was directly related to PIM-2, PELOD, and ICU and hospital LOS (p < 0.05). A-FaBP level was associated with PELOD, hospital and ICU length of stay (p < 0.05). MMP-9/TIMP-1 ratio associated with poor Glasgow Outcome Score (p < 0.05). A-FaBP levels in septic patients with neurological dysfunction (29.3, 17.2-54.6, 7) were significantly increased compared to septic patients without neurological dysfunction (14.6, 13.32-0.6, 11). MMP-9/TIMP-1 ratios were significantly lower, while A-FaBP and mrProANP were higher in septic patients compared to the control groups. Each biomarker was associated with hospital morbidity and length of stay. These results suggest that these biomarkers merit further prospective study for the early identification of children with sepsis.