Applications Of Serum Human Neutrophil Lipocalin Levels In Acute Bacterial Infection

Objective: This study is in order to investigate the clinical application value of the ser um HNL in acute bacterial infection through th e tests of serum HNL levels in healthy control group and case groups. Meanwhile, to compare the serum HNL levels with serum PCT, serum CRP and WBC, and to analysis the diagnostic efficiency of these four measurements. To investigate the kinetic of serum HNL on the basis of dynamically detections of HNL, the value of monitoring the efficacy of antibiotics was observed as well.Methods: This study included a total of 193 se rum samples including 133 bacterial infections and 60 viral inf ections, which were obtaine d from hospitalized patients with acute infection between Apr 2014 and Jan 2015 at the First Bethune Hospital of Jilin University, Jilin, Changchun. The b acterial group included 93 cases of local bacterial group and 40 cases of sepsis group. 60 healthy people in the corresponding period were enrolled as the control group. A ll cases were determ ined the levels of serum HNL, PCT, CRP and WBC before treatment. Patients of local bacterial infection group were determ ined the levels of serum HNL, PCT and CRP after the use of antibiotics for 48 h and 72 h. The analysis was conducted by the SPSS program(version 18.0). A P value of less than 0.05 was considered statistically significant.Results: 1.In pre-antibiotics treatment, the m edians of serum HNL levels are 58.42μg/L(control group), 90.41μg/ L(viral group), 221.87μg/L(local bacterial group) and 309.94μg/L(sepsis group), resp ectively. There were significant differences between control group and other three case groups( P<0.001). There were significant differences between any two groups of the three case groups( P<0.001). The results of serum PCT and CRP indicated that the dif ferences between viral groups and control groups we re statistically significant( P<0.05). The differences between local bacterial groups or sepsis groups and control groups were statistically significant(P<0.01). There were significant dif ferences between any two groups of the three case groups( P<0.01). In the results of WBC, there were significant differences between viral groups or local bacterial groups and control groups(P<0.05). The dif ference between sepsis group and control group were statistically significant(P<0.01). There were significant dif ferences between any two groups of the three case groups(P<0.01). 2.The ROC curves of any of the four markers only and joint detection for the diagnosis of acute bacterial infection indicated that the optim al cut-off value of discrimination between acute viral and bacterial infec tions occurred at 142.39μg/L for serum HNL with a AUC of 0.975( 95%CI: 0.954-0.996), superior to AUCs of serum PCT, serum CRP and WBC which were 0.947( 95%CI: 0.910-0.984), 0.849(95%CI: 0.785-0.913)and 0.760( 95%CI: 0.679-0.841). W hen the optim al cut-off value occurred at 142.39μg/L, serum HNL had the sensitivity, specificity and Youden index of 94.6%, 93.3% and 0.879, respectively, superior to those of serum PCT(90.3%, 90%, 0.803), CRP(74.2%, 86.7%, 0.609) and WBC(78.5%, 66.7%, 0.452). The analysis of the four measurements revealed that there was no significant differences between serum HNL and PCT(P>0.05) for the diagnosis of acute bacterial infection, while serum HNL was better than serum CRP or W BC with statistically significance(P<0.05). While, the joint detection had the biggest AUC of 0.983 with the sensitivity and specificity of 97.8% and 98.3%。 3.Among patients with local bacterial infections, before treatment serum HNL levels elevated to 221.38(169.77,302.43) μg/L, and started to reduced after treatment for 48 h and continued to reduce fu rther on 72 h after treatm ent, 162.40(127.08,198.95) μg/L and 1 12.89(79.34,132.96) μg/L, respectively. The differences of HNL 、PCT and CRP were highly statis tically significant at all detection points(P<0.01), and the dif ferences of WBC had no statistically significance(P>0.05). The proportions of the se rum HNL levels above the optim al cut-off value varied from 94.6%(88/93) to 50.5%(47/93) afte r treatment for 48h and to 12.9%(12/93) after treatm ent for 72 h. Serum PCT was similar to HNL, the results above the optimal cut-off values varied from 91.4%(85/93) to 58.1%(54/93) and 20.4%(19/93) quickly. On the contrary, the concentrations of serum CRP stayed increased even 48 h or 72 h after antibiotics treatment, there were even 68.8%(64/93) cases above the optim al cut-off value afte r treatment for 72 h. The proportions of WBC results above the optim al cut-off value were 78.5%(73/93), 70%(65/93) and 66.7%(62/93), respectively.Conclusion: 1. Detection of serum HNL level is another new method which can be used to diagnose acute bacterial inf ections and the levels of serum HNL can reflect the severity of acute bacterial infections. 2. The diagnostic efficiency of serum HNL is superior to W BC, serum PCT and CRP. 3. Dynamically detection of serum HNL may monitor the disease progressions of acute b acterial infections and p rovide guidelines for clinicians to adjust the antibiotics in time.