BACKGROUND: The aim of this study was to assess the use of neutrophil distribution width (NDW) and to compare it to C-reactive protein (CRP) and procalcitonin (PCT), in the detection of early sepsis in the intensive care unit. METHODS: Subjects (N = 166) were divided into 4 groups: healthy, acute inflammatory non-infectious (AINI), localized infection, and systemic infection, according to clinical history and cultures. NDW, CRP, and PCT were compared among the different groups using multivariate analysis of variance (MANOVA). Diagnostic efficacy was assessed using receiver operating characteristic curves and areas under the curves (AUC). RESULTS: The lowest mean(NDW) was found in the healthy group (n = 41), followed by the AINI (n = 20), localized infection (n = 55), and systemic infection (n = 50) groups. AUC(NDW) was 0.877 for infected (localized + systemic) vs non-infected (healthy + AINI) groups, and 0.965 for systemic infection vs non-infected groups. A cut-off of 21.9 resulted in 90% sensitivity, 92% specificity, 90% positive predictive value, and 92% negative predictive value (AUC(NDW) = 0.965, 95% confidence interval 0.935-0.995). According to MANOVA, only NDW was able to differentiate an acute inflammatory process from early infection in postoperative patients, but not healthy from AINI subjects. CONCLUSIONS: NDW had the highest diagnostic accuracy and is available with the complete blood count with differential (CBC). It may be a promising parameter to aid in the diagnosis of acute infection in adults, provided the possibility of haematological disorders is first ruled out.
BACKGROUND: The aim of this study was to assess the use of neutrophil distribution width (NDW) and to compare it to C-reactive protein (CRP) and procalcitonin (PCT), in the detection of early sepsis in the intensive care unit. METHODS: Subjects (N = 166) were divided into 4 groups: healthy, acute inflammatory non-infectious (AINI), localized infection, and systemic infection, according to clinical history and cultures. NDW, CRP, and PCT were compared among the different groups using multivariate analysis of variance (MANOVA). Diagnostic efficacy was assessed using receiver operating characteristic curves and areas under the curves (AUC). RESULTS: The lowest mean(NDW) was found in the healthy group (n = 41), followed by the AINI (n = 20), localized infection (n = 55), and systemic infection (n = 50) groups. AUC(NDW) was 0.877 for infected (localized + systemic) vs non-infected (healthy + AINI) groups, and 0.965 for systemic infection vs non-infected groups. A cut-off of 21.9 resulted in 90% sensitivity, 92% specificity, 90% positive predictive value, and 92% negative predictive value (AUC(NDW) = 0.965, 95% confidence interval 0.935-0.995). According to MANOVA, only NDW was able to differentiate an acute inflammatory process from early infection in postoperative patients, but not healthy from AINI subjects. CONCLUSIONS: NDW had the highest diagnostic accuracy and is available with the complete blood count with differential (CBC). It may be a promising parameter to aid in the diagnosis of acute infection in adults, provided the possibility of haematological disorders is first ruled out.