Predictive Value Of Peripheral Blood Inflammatory Indexes For Blood Stream Infection In Patients With Urinary Tract Infection

Lizhe An,Qingquan Xu,Qi Wang,Qingya Yang,Zhenjie Zhu,Yang Hong,Xiaobo Huang,Tao Xu
2017-01-01
Abstract:Purpose: To investigate the predictive value of peripheral blood inflammatory indexes for blood stream infection (BSI) in patients with urinary tract infection. Materials and methods: Patients who were suspicious of blood stream infection and received blood culture and routine blood test simultaneously between January 2011 and June 2016 in our department were analyzed retrospectively. A total of 51 patients with positive blood culture were assigned to Group BSI. And the other 238 patients with negative blood culture were assigned to Group control. Peripheral blood inflammatory indexes including white blood cell count (WBC), neutrophil count (NEU), lymphocyte count (LYM), monocyte count (MON), percentage of neutrophil (NEU%), percentage of lymphocyte (LYM%), percentage of monocyte (MON%), neutrophil-to-lymphocyte ratio (NLR), and neutrophil-to-monocyte ratio (NMR) were compared between groups. Then predictive value of indexes that had statistical difference and integrated index of NLR and NMR were evaluated through Receiver Operating Characteristic (ROC) curve. Multivariate Logistic regression was done to identify indexes' contribution to BSI. Results: NEU%, NLR and NMR were significantly higher in Group BSI (P<0.05). However, LYM%, MON%, LYM and MON were significantly higher in Group control (P<0.05). No statistical difference could be seen in WBC and NEU between two groups (P>0.05). AUC of LYM, MON, NEU%, LYM%, MON%, NLR and NMR were 0.632, 0.700, 0.670, 0.633, 0.656, 0.630 and 0.703 respectively. Multivariate analysis showed NLR and NMR were independently related to BSI (OR = 2.463, 95% CI: 1.225 similar to 4.951, P = 0.011) (OR = 3.205, 95% CI: 1.617 similar to 6.351, P = 0.001). Conclusion: For patients with urinary tract infection who were suspicious of BSI, peripheral blood indexes should be supervised. MON and NMR were independent prognostic factors and had higher predictive value for BSI, combining NLR and NLR could improve accuracy.
What problem does this paper attempt to address?