Diagnostic value of serum leptin and a promising novel diagnostic model for sepsis

作者:Chen, Mingyi; Wang, Bin; Xu, Yaping; Deng, Zihui; Xue, Hui; Wang, Luhuan; He, Lei*
来源:Experimental and Therapeutic Medicine, 2014, 7(4): 881-886.
DOI:10.3892/etm.2014.1506

摘要

Diagnosis of sepsis in critically ill patients is important to reduce morbidity and mortality. The present study was conducted to determine the role of serum leptin in the early diagnosis of sepsis and to establish a diagnostic model for sepsis. A retrospective study was conducted of 331 patients from an intensive care unit. All patients underwent consistent blood collection at 6:00 a.m. every morning after fasting. Serum leptin concentrations and additional markers of sepsis were compared between the sepsis group (n=128) and the non-sepsis group (n=203). Septic patients displayed significantly higher leptin serum concentrations compared with those of the non-sepsis group (mean concentration, 11.67 versus 4.824 mg/dl; P<0.001). The leptin levels in male patients were higher than those in female patients, particularly in the sepsis group. The accuracy of serum leptin levels in distinguishing septic patients from non-septic patients was 76%, and the area under the receiver operating characteristic (ROC) curve of serum leptin was <= 0.8. Additional markers of inflammation in the sepsis group were also significantly higher than those in the non-sepsis group. Positive correlations were identified between leptin and body temperature, heart rate and creatinine levels. Therefore, a prognostic model comprising a combination of leptin with temperature, platelet count, white blood cell count and heart rate was evaluated as an effective logistic regression model for the diagnosis of sepsis. The logistic regression output cut-off value was 0.46 and the area under the ROC curve was 0.953 (P<0.0001). It may be concluded that leptin is a valuable marker in the diagnosis of sepsis and the proposed prognostic model is an effective logistic regression model for the diagnosis of sepsis. The prognostic model is able to aid the differentiation of septic patients from non-septic patients.