The Value of Neutrophil-To-Lymphocyte Ratio for Evaluating Blood Stream Infection Caused by Carbapenem-Resistant Klebsiella pneumoniae: A Retrospective Cohort Study

Abstract

Background: The neutrophil-to-lymphocyte ratio (NLR) is a useful marker of inflammation. However, the prognostic function of the NLR in patients with carbapenem-resistant Klebsiella pneumoniae (CRKP) blood stream infection (BSI) remains largely unknown. The aim of this study was to explore the potential relationship between the NLR and mortality in these patients.

Methods: We performed a retrospective cohort study based on data retrieved from the computerized patient record system in a tertiary hospital from 1 January 2017 to 31 October, 2020. A total of 134 inpatients with CRKP BSI were enrolled in this study, including 54 fatal cases and 80 survival cases, 28 days after the onset of CRKP BSI. A logistic analysis was performed to assess the association between the NLR on the 4th day and 28-day mortality. Multivariate analyses were used to control for the confounders.

Results: The overall 28-day mortality rate of patients with a CRKP BSI episode was 40.3% (54/134). We conducted a multivariate analysis of the data of 134 patients and found that the NLR on the 4th day [odds ratio (OR) 1.148, 95% confidence interval (CI) 1.076-1.225, p < 0.001] and antibiotic exposure before BSI onset (OR 3.847, 95% CI 1.322-11.196, p = 0.013) were independent risk factors for 28-day mortality of patients with CRKP BSI, while appropriate initial therapy (AIT, OR 0.073, 95% CI 0.017-0.307, p < 0.001) was an independent protective factor. Among patients treated with AITs, the Cox proportional hazards regression analysis revealed a significant difference in prognosis (p = 0.006) between the ceftazidime/avibactam contained (CAZ) group and non CAZ-AVI groups. After dividing the non CAZ-AVI group into the tigecycline (TGC), colistin (COL), and TGC + COL groups, there were no differences between the CAZ-AVI group and the TGC group (p = 0.093), but CAZ-AVI group showed lower 28-day mortality than COL (p = 0.002) and TGC + COL (p = 0.002) groups. Meanwhile, there was no difference in NLR on the 1st day (p = 0.958) of patients in different groups but significant difference in NLR on the 4th day (p = 0.047).

Conclusions: The NLR on the 4th day is a readily available and independent prognostic biomarker for patients with CRKP BSI. This marker may have the potential for use in evaluating the efficacy of different anti-infection therapy strategies at an early stage.

Keywords: blood stream infection; carbapenem-resistant Klebsiella pneumoniae; neutrophil-to-lymphocyte ratio; prognosis; therapy strategies.

Figure 1

Flowchart of the study design.

Figure 2

The ROC curves of NLR and 28-day mortality. ROC, receiver operating characteristic; NLR, neutrophil- to-lymphocyte count ratio; AUC, area under curves.

Figure 3

Kaplan-Meier analysis of NLR on the 4th day. The blue line represents NLR < 12.90. The green line represents NLR ≥ 12.90.

Figure 4

The Cox proportional hazards regression survival analysis. (A) CAZ-AVI vs. non CAZ-AVI (p = 0.006). (B) CAZ-AVI vs. TGC (p = 0.093), CAZ-AVI vs. COL (p = 0.003), CAZ-AVI vs. TGC + COL (p = 0.012). CAZ-AVI, therapy contained ceftazidime/avibactam; TGC, therapy contained tigecycline; COL, therapy contained colistin, TGC + COL, tigecycline and colistin combined therapy.