Clinical significance of the neutrophil‐to‐lymphocyte ratio on the prognosis of critically ill elderly patients with sepsis— a retrospective study based on the MIMICV‐IV database

Abstract Introduction This study aimed to elucidate the relationship between neutrophil‐to‐lymphocyte ratio (neutrophil–lymphocyte ratio, NLR) and 28‐day mortality in critically ill elderly sepsis patients. Methods Retrieval of information on patients > 65 years of age who meet the diagnostic criteria for sepsis in sepsis 3.0 using the Medical Information Marketplace for Intensive Care‐IV (MIMIC‐IV) database. Optimal NLR truncation was identified using the X‐tile software and corresponding Kaplan–Meier (K‐M) survival curves were plotted. The ability of the NLR in combination with quick Sequential Organ Failure Assessment (qSOFA) and both alone to predict 28‐day mortality in critically ill elderly septic patients was also assessed using the receiver operating characteristic curve. Least absolute shrinkage and selection operator (LASSO) and logistic regressions were applied to factors with p < 0.05 to establish a 28‐day mortality model. Area under the curve (AUC), calibration curve, and decision curve analysis were utilized to assess nomogram predictive accuracy. Results Among 2032 elderly sepsis patients, 44.6% expired. The K–M survival analysis suggested that NLR ≥ 13.6 significantly increased the risk of death. The AUC of the NLR combined with the qSOFA score to predict 28‐day mortality in critically ill elderly septic patients was 0.764, which was significantly higher than that of the NLR (0.728) and qSOFA (0.644) alone and stronger than the predictive power of the SOFA (0.738) but weaker than that of the Simplified Acute Physiology Score ( p = 0.785). A nomogram of 28‐day mortality in critically ill elderly septic patients was constructed using regression analysis of the collected variables, culminating in the inclusion of five factors: age, NLR, lactate, mechanical ventilation, and qSOFA. This nomogram showed good discrimination and calibration in predicting mortality in critically ill elderly septic patients. Conclusion This study showed that low levels of NLR (< 13.6) were significantly associated with better survival. The ability of the NLR combined with the qSOFA score to predict 28‐day mortality in elderly patients with severe sepsis was significantly higher than when the two were used alone and was also stronger than the predictive ability of the SOFA but weaker than that of the SAPSII. The nomogram constructed using the LASSO regression and logistic regression analysis effectively integrates multiple vital factors, showing good predictive ability and clinical practicability, and providing a powerful tool for evaluating the prognosis of critically ill elderly patients with sepsis.