The Association Between the Neutrophil-to-Lymphocyte Ratio and Mortality in Patients with Acute Respiratory Distress Syndrome: A Retrospective Cohort Study

Weijing Li,Xiaolin Ai,Yuenan Ni,Zengpanpan Ye,Zongan Liang
DOI: https://doi.org/10.1097/shk.0000000000001136
2018-01-01
Shock
Abstract:Purpose: Systemic inflammation relates to the initiation and progression of acute respiratory distress syndrome (ARDS). As neutrophil-to-lymphocyte ratio (NLR) has been shown to be a prognostic inflammatory biomarker in various diseases, in this study, we sought to explore whether NLR is a prognostic factor in patients with ARDS. Methods: A retrospective study was performed on patients diagnosed as ARDS admitted to the intensive care unit (ICU). We calculated the NLR by dividing the neutrophil count by the lymphocyte count and categorized patients into four groups based on quartile of NLR values. The association of NLR quartiles and 28-day mortality was assessed using multivariable Cox regression. Secondary outcomes included ICU mortality and hospital mortality. Results: A total of 224 patients were included in the final analysis. The median (interquartile range) NLRs from first quartile to fourth quartile were as follows: 6.88 (4.61-7.94), 13.06 (11.35-14.89), 20.99 (19.09-23.19), and 39.39 (32.63-50.15), respectively. The 28-day mortalities for the same groupswere as follows: 10.7%, 19.6%, 41.4%, and 53.6% (P<0.001). Cox regression analysis showed NLR was a significant risk factor predicting 28-day mortality (first quartile, reference group; second quartile, adjusted hazard ratio [HR] = 1.674, 95% confidence interval [CI], 0.462-6.063, P = 0.432; third quartile, HR = 5.075, 95% CI, 1.554-16.576, P = 0.007; fourth quartile, HR = 5.815, 95% CI, 1.824-18.533, P = 0.003). Similar trends were observed for ICU mortality and hospital mortality. Conclusions: High NLR was associated with the poor outcome in critically ill patients with ARDS. The NLR therefore seems to be a prognostic biomarker of outcomes in critically ill patients with ARDS. Further investigation is required to validate this relationship with data collected prospectively.
What problem does this paper attempt to address?