The Neutrophil/Lymphocyte Ratio Could Predict Noninvasive Mechanical Ventilation Failure in Patients with Acute Exacerbation of Chronic Obstructive Pulmonary Disease: A Retrospective Observational Study.

2021 
Purpose To determine the effectiveness of neutrophil/lymphocyte ratio (NLR), compared to traditional inflammatory markers, for predicting noninvasive mechanical ventilation (NIMV) failure in acute exacerbation of chronic obstructive pulmonary disease (AECOPD) patients. Patients and Methods We conducted this retrospective observational study including 212 AECOPD patients who required NIMV during hospitalization from January 2015 to December 2020 in the department of respiratory and critical care medicine of Beijing Chao-Yang Hospital (west campus). We reviewed the medical record to determine if NIMV succeeded or failed for each patient, and compared NLR with traditional markers (leukocyte, C-reactive protein [CRP] and procalcitonin [PCT]) between NIMV failure and NIMV success group. Receiver-operating characteristic (ROC) curve and multivariate logistic regression analysis were used to assess the accuracy of these markers for predicting NIMV failure. Results A total of 38 (17.9%) patients experienced NIMV failure. NLR was a more sensitive biomarker to predict NIMV failure (AUC, 0.858; 95% CI 0.785-0.931) than leukocyte counts (AUC, 0.723; 95% CI 0.623-0.823), CRP (AUC, 0.670; 95% CI 0.567-0.773) and PCT (AUC, 0.719; 95% CI 0.615-0.823). There was statistically positive correlation between NLR and leukocytes count (r=0.35, p 8.9 (odds ratio, 10.783; 95% CI, 2.069-56.194; P=0.05) was an independent predictor of NIMV failure in the multivariate logistic regression model. Conclusion NLR may be an effective marker for predicting NIMV failure in AECOPD patients, and the patients with NLR>8.9 should be handled with caution since they are at higher risk of NIMV failure and require intubation. Further study with a larger sample size and with more data is necessary to confirm our study.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    42
    References
    0
    Citations
    NaN
    KQI
    []