Predictive value of procalcitonin for infection of patients with type‑2 diabetes mellitus

Abstract

The aim of the present study was to investigate the predictive value of procalcitonin (PCT) for infection in patients with type 2 diabetes mellitus (T2DM). A retrospective analysis of 178 patients with T2DM who were divided into non-infection, local infection and sepsis groups was conducted; in conjunction with 33 healthy control patients. Clinicopathological characteristics and inflammatory indicators were compared between the four groups. Patients in the non-infection group exhibited significantly higher PCT levels compared with healthy controls (P=0.002). In addition, PCT, C-reactive protein (CRP), white blood cell count and neutrophil percent were significantly different amongst patients with T2MD across different infection groups (all P<0.001) with the following rank order: Sepsis group > local infection group > non-infection group (all P<0.05). In addition, the following observations were made: i) PCT and CRP demonstrated larger areas under the curve (AUC) for predicting local infection (0.804 and 0.741, respectively); ii) PCT displayed lower sensitivity of only 21.8% at its classical cutoff value (0.500 ng/ml) whereas CRP exhibited higher sensitivity and specificity at 64.1 and 76.1%, respectively, at its classical cutoff value (10.0 mg/l); and iii) PCT exhibited the largest AUC (0.914) for predicting sepsis with high sensitivity and specificity (86.4 and 84.5%, respectively) at its cutoff value (0.990 ng/ml). Patients with T2DM without infection demonstrated higher baseline PCT levels. The present study clarified the value of PCT in predicting infection of T2DM patients. The application of PCT to predict local infection in patients with T2DM was identified to be inferior to CRP, but its ability to predict sepsis was concluded to be the best when compared with CRP, white blood cell count and neutrophil percent.