Heart-type fatty acid-binding protein as a prognostic factor in patients with severe sepsis and septic shock.

OBJECTIVE: This study was performed to evaluate whether heart-type fatty acid-binding protein (H-FABP) could predict 28-day mortality in patients with severe sepsis and septic shock.

METHODS: We performed a prospective observational study and included consecutive patients with severe sepsis and septic shock. Patients' demographic data, Acute Physiology and Chronic Health Evaluation (APACHE) II score, and the blood test results including H-FABP concentrations were compared between the 28-day survivors and nonsurvivors. The association between the concentration of H-FABP and survival was analyzed with multivariate logistic regression and Cox proportional hazards regression analyses. The prognostic performance of H-FABP was compared with those of the APACHE II score and albumin using the area under the receiver operating characteristic curve.

RESULTS: Of the 99 patients, 38 (38%) died. The mortality rate increased with increasing H-FABP concentration. In multivariate logistic regression analyses, H-FABP greater than 40 ng/mL was an independent predictor of mortality compared with H-FABP less than 7 ng/mL (odds ratios, 9.23; 95% confidence interval, 1.29-65.86). By Cox proportional hazards analysis, H-FABP greater than 40 ng/mL was associated with a 5.57-fold increased risk for death during the 28-day follow-up period (hazard ratio, 5.57; 95% confidence interval, 1.20-25.80). The area under the receiver operating characteristic curve of H-FABP was 0.739 (95% confidence interval, 0.640-0.839), which was comparable with those of the APACHE II score and albumin.

CONCLUSION: The H-FABP was an independent prognostic factor and could be a useful biomarker for 28-day mortality in patients with severe sepsis and septic shock.