1. Introduction
Notably, the neutrophil-to-lymphocyte ratio (NLR) has been reported to be a prognostic predictor in sepsis recently.[ 1 ] However, there are still some debates on it. A previous study found no significant relationship between NLR and prognosis of mortality in critical patients with sepsis .[ 2 ] Another study reported that the nonsurvivors had a relatively lower NLR, with a 0.1-fold reduction in mortality for every 10-unit increase in sepsis .[ 3 ] The other study reported that a low NLR was associated with early death (<5 days after the onset of septic shock ) in septic shock patients; the NLR of the early-death patients at admission was significantly lower than that of late-death patients and survivors. But NLR increment was associated with late death (>5 days after the onset of septic shock ) in septic shock patients.[ 4 ] To explore the influence factors of NLR, clinical data were extracted from the electronic database of eICU to investigate the predictive capability of NLR in different situations.
2. Research presentation
2.1. Ethical approval
This study has been approved by the ethics committee of the Chinese PLA General Hospital (S2017-054-01). The requirement to obtain written informed consent from each patient was waived because this was an observational retrospective study.
2.2. Patients and methods
The retrospective study was based on the eICU Collaborative Research Database, which consists of data from 139,367 patients (200,859 ICU visits) who were admitted to 200 different ICUs located throughout the United States in 2014 and 2015.[ 5 ] Septic patients were selected from the eICU database. The NLRs were compared between the survival and nonsurvival groups. Then, receiver operating characteristic (ROC) curve was used, and the area under the curve (AUC) of NLR was calculated under different conditions as shown in Figure S1, https://links.lww.com/IDI/A23
Python 3.0 (Wilmington, DE, USA) was used for statistics. Continuous variables distributed normally were summarized as mean ± SD. Continuous variables distributed nonnormally were summarized as a median (lower quartile–upper quartile). Propensity score matching analysis was used to generate equally weighted cohorts. The accuracy of NLR predicting in-hospital mortality was calculated using ROC curves and AUC. A P value less than 0.05 was considered statistically significant.
3. Results
A total of 10,840 ICU admissions were analyzed. After propensity score matching analysis, 1096 nonsurvivor and survivor patients were matched, and data showed that NLR was still higher in the nonsurvivor group than that in the survivor group (17.97 [8.94–31.85] vs. 15.27 [7.92–26.28], P = 0.008) (Table S1, https://links.lww.com/IDI/A24
First, the AUC of the NLR increased from 0 day to the sixth day and decreased from the sixth day to the eighth day [Figure 1 ]. Data showed that the AUC of the sixth day was the highest (0.74 ± 0.03), with P < 0.01 when compared with that of 0 to the second day. Second, the ROC curve was used to evaluate the Simplified Acute Physiological Score 3 (SAPS3 ) as a predictor of in-hospital mortality in sepsis . Septic patients were divided into two groups depending on the cutoff value of SAPS3 (≤52 and >52). The AUCs of NLR in sepsis patients with different SAPS3 groups are shown in Figure 2 . The AUC of the NLR in the ≤52 group was significantly higher than that in the >52 group (0.78 ± 0.05 vs. 0.69 ± 0.03, P = 0.007). Third, patients with sepsis were divided into the shock group and nonshock group. There was significant difference in the AUC of the NLR between the nonshock group and shock group (0.72 ± 0.04 vs. 0.65 ± 0.05, P < 0.01) [Figure 3 ].
Figure 1: The AUC of the NLR from 0 to the eighth day after admission. The AUC of the NLR increased from 0 day to the sixth day, and decreased from the sixth day to the eighth day. The NLR of the 0 day was obtained from the records within 3 days before admission, but was the nearest one to the time point of admission. Data showed that the AUC of the sixth day was the highest (0.74 ± 0.03), with P < 0.01 when compared with 0 to the second day. There was no significant difference between the sixth day and the third, fourth, fifth, seventh, or eighth day. **P < 0.01, compared with the AUC of the sixth day. AUC: Area under curve; NLR: Neutrophil-to-lymphocyte ratio .
Figure 2: The AUC of the NLR on day 6 after admission in different SAPS3 groups. Septic patients were divided into two groups depending on the cutoff value of SAPS3 (≤52 and >52). The NLR of the sixth day after admission was selected. The AUC of the NLR in the ≤52 group (n = 2871) was significantly higher than that in the >52 group (n = 2270). AUC: Area under curve; NLR: Neutrophil-to-lymphocyte ratio . **P < 0.01.
Figure 3: The AUC of the NLR on day 6 after admission in the nonshock and shock groups of patients with sepsis . Patients with sepsis were divided into shock group and nonshock group. The NLR of the sixth day was selected. There was a significant difference in the AUC of the NLR on day 6 after admission between the nonshock group (n = 4216) and shock group (n = 925). AUC: Area under curve; NLR: Neutrophil-to-lymphocyte ratio . **P < 0.01.
4. Discussion
Neutrophil-to-lymphocyte ratio is reported to be a useful tool to distinguish those sepsis patients with high risk of death. However, association between NLR and sepsis prognosis was impacted by some confounding parameters. Thus, this retrospective study was conducted, and influencing factors including admission time, SAPS3 , and septic shock were analyzed based on the eICU database.
In this present study, NLR in the survivors was lower than that in the nonsurvivors. Based on this result, further analysis was carried on to investigate the impact of different factors. First, our data showed that the highest AUC of the NLR was on the sixth day. Too early (0–3 day) or too late (eighth day) stage was not ideal for NLR as a predictive tool of prognosis in sepsis . Neutrophils are activated by proinflammatory cytokine and endotoxin stimuli at the early stage of sepsis . However, apoptosis of neutrophils and lymphocytes increases gradually during sepsis , which affects the NLR as a result.[ 6 ] Therefore, the predictive capability of the NLR might vary in sepsis . Then, data showed that the AUC of the NLR in the SAPS3 ≤52 group was significantly higher than that in the SAPS3 >52 group. NLR had a better predicting capability in patients with a SAPS3 ≤52. The cutoff value of 52 was determined by ROC curve analysis of SAPS3 . SAPS3 >52 meant a higher risk of death. The response of neutrophils and lymphocytes was depressed by severe infection in those septic patients with SAPS3 >52. In parallel, sepsis shock also impacted the AUC of the NLR significantly. There was a significant difference in the AUC of the NLR between septic patients and septic shock patients. Septic patients with shock were probably more severe, and their SAPS3 may be higher than that of septic patients without shock . Results about the influence of shock were consistent with the data of SAPS3 . In brief, NLR showed a statistically better predictive capability of in-hospital mortality in less severe patients. Severe infection may result in depressed immunologic function. This might impair the sensitivity of NLR response and lead to a decreasing predictive capability of NLR.
Nowadays, there is a growing interest about NLR. NLR is derived from routine blood test and reported to be significantly related to mortality of sepsis , which makes NLR an economical, readily available, and reliable predictive tool. Neutrophils represent innate inflammation, whereas lymphocytes are the major cells of the adaptive immune system.[ 7 ] Both are activated promptly in large amounts of circumstances, such as trauma, surgical stress, tumor, bacteremia and infection, shock , and sepsis . NLR reflects the interaction between innate (neutrophil granulocytes) and adaptive immune system (lymphocytes). Not only physiological process (wound repair and healing), but also pathophysiological response (excessive inflammation, cytokine storm, or prolonged immunosuppression or residual inflammation), could be expressed by NLR. NLR is widely taken as a useful marker of prognosis in a great deal of diseases, such as rheumatism, tumor, stroke, diabetes, pneumonia, coronary heart disease, sepsis , and so on.[ 7 ] A study found that NLR was a potential predictor of acute respiratory distress syndrome. An NLR >14.35 strongly indicated a benefit from corticosteroid therapy for acute respiratory distress syndrome patients.[ 8 ] NLR is thought to be a valuable prognostic tool in septic patients now. The absolute value and the dynamic changes are both crucial in NLR, and both objectively reflect the severity of sepsis . This was verified again in our present study.
There are several limitations to this present study. First, the data are derived from only one database, which limits the application of the conclusion drawn in this present study. Second, the nature of the retrospective cohort research decreases the grade of evidence produced in this study. Propensity score matching analysis was applied to decrease the influence of confounding parameters, which partly made up the deficiency of cohort study. Finally, this is a secondary analysis based on the eICU database. The detailed information of medication is not provided in the electronic database. Therefore, further confirmatory study is needed to overcome these shortages.
Taken together, a more appropriate strategy of NLR to be an outcome predictor of sepsis is suggested for physicians in critical care medicine.
Funding
This study was funded with support from Beijing Natural Science Foundation (No. 7222162).
Author Contributions
Qing Zhao and Yubao Ma took part in the study design, performed statistical analyses and drafted the manuscript. Jianguo Xiao had the access permission to eICU and extracted data for the present study. Hui Liu conceived of the study, participated in the design and revised the manuscript critically for important intellectual content. All authors read and approved the final manuscript.
Conflicts of Interest
None.
Data Sharing Statement
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
References
1. Liu S, Wang X, She F, et al. Effects of
neutrophil-to-lymphocyte ratio combined with interleukin-6 in predicting 28-day mortality in patients with
sepsis . Front Immunol 2021;12:639735. doi: 10.3389/fimmu.2021.639735.
2. Salciccioli JD, Marshall DC, Pimentel MA, et al. The association between the
neutrophil-to-lymphocyte ratio and mortality in critical illness: an observational cohort study. Crit Care 2015;19(1):13. doi: 10.1186/s13054-014-0731-6.
3. Ni J, Wang H, Li Y, et al. Neutrophil to lymphocyte ratio (NLR) as a prognostic marker for in-hospital mortality of patients with
sepsis : a secondary analysis based on a single-center, retrospective, cohort study. Medicine 2019;98(46):e18029. doi: 10.1097/MD.0000000000018029.
4. Riché F, Gayat E, Barthélémy R, et al. Reversal of neutrophil-to-lymphocyte count ratio in early versus late death from septic
shock . Crit Care 2015;19:439. doi: 10.1186/s13054-015-1144-x.
5. Pollard TJ, Johnson AEW, Raffa JD, et al. The eICU Collaborative Research Database, a freely available multi-center database for critical care research. Sci Data 2018;5:180178. doi: 10.1038/sdata.2018.178.
6. Hsu YC, Yang YY, Tsai IT. Lymphocyte-to-monocyte ratio predicts mortality in cirrhotic patients with septic
shock . Am J Emerg Med 2021;40:70–76. doi: 10.1016/j.ajem.2020.11.071.
7. Wittermans E, van de Garde EM, Voorn GP, et al. Neutrophil count, lymphocyte count and
neutrophil-to-lymphocyte ratio in relation to response to adjunctive dexamethasone treatment in community-acquired pneumonia. Eur J Intern Med 2022;96:102–108. doi: 10.1016/j.ejim.2021.10.030.
8. Wang Q, Xie T, Gao R, et al.
Neutrophil-to-lymphocyte ratio is a powerful predictor of adult patients with acute respiratory distress syndrome who might benefit from corticosteroid therapy. Am J Transl Res 2021;13(10):11556–11570.
