Neutrophil/lymphocyte ratio is helpful for predicting weaning failure: a prospective, observational cohort study.

Background: To assess the usefulness of the neutrophil/lymphocyte ratio (NLR), a marker of inflammation and/or stress, for predicting weaning failure in patients receiving invasive mechanical ventilation (IMV), compared to levels of leukocytes and C-reactive protein (CRP).

Methods: This observational prospective cohort study was conducted from July 2013 to December 2016 in an intensive care unit in China, enrolling 269 consecutive patients receiving IMV. Patients underwent a spontaneous breathing trial (SBT) if they were ready to wean, and underwent extubation if they passed the SBT. The evaluated markers were measured immediately prior to SBT, and compared between weaning-failure and weaning-success patients. Receiver-operating characteristic (ROC) curve and logistic regression analyses were used to evaluate the ability of these markers to predict weaning failure.

Results: In all, 94 (34.9%) patients failed the weaning process (66 failed SBT and 28 presented with post-extubation respiratory distress). NLR was a better predictor of failure (area under the ROC curve, 0.69; 95% CI, 0.62-0.76) than leukocyte levels (0.60, 0.53-0.67) and CRP values (0.58, 0.51-0.65). NLR >11, leukocyte counts >10.5×10/L, and CRP >58 mg/L prior to weaning had the best combination of sensitivity (73%, 64%, and 63%, respectively), specificity (59%, 55%, and 63%), positive predictive value (49%, 43%, and 48%), negative predictive value (81%, 74%, and 76%), and diagnostic accuracy (64%, 58%, and 63%) for predicting weaning failure. However, only NLR >11 (odds ratio, 5.91; 95% CI, 3.08-11.33; P<0.001) was an independent predictor of weaning failure in the adjusted logistic regression model.

Conclusions: NLR may be a useful marker for predicting weaning failure, and weaning at NLR >11 might be considered with caution. Further study with a larger sample size and with weaning outcome as a variable of concern is warranted. Trial registration: ClinicalTrials.gov identifier: NCT02981589.