Sniff nasal inspiratory pressure as a predictor of 1-year mortality after exacerbation of COPD

Abstract

Introduction: The maximal sniff nasal inspiratory pressure (SNIP) is widely used as a simple assessment of inspiratory muscle strength. Both SNIP and the BODE index have been reported to be able to predict mortality in stable COPD (Moore AJ, et al. 2010 and Celli BR, et al. 2004). Aims: This study aimed to compare SNIP to the BODE index as predictors of 1-year mortality in patients with COPD after an exacerbation. Methods: A prospective mortality analysis of 48 patients admitted for exacerbation of COPD [39 male, age median 78 (IQR 72-82) years, length of stay 19 (12-35) days] was performed (Ethics 15R149). All patients underwent pulmonary rehabilitation during hospitalization. At discharge, SNIP and the BODE index were measured. Mortality was followed up for 1-year after discharge. Predictive ability for mortality was tested using multivariate Cox proportional hazard (PH) model adjusted for age and gender, and discriminatory power by the area under the receiver operating characteristic curve (AUC). Results: 7 patients (15%) died during the study period. At discharge, SNIP and the BODE index were 47 (35-60) cmH2O and 6 (4-7) [FEV1 44 (34-55) % predicted, BMI 21 (18-24) kg/m2, 6-min walk distance 220 (164-279) m, mMRC dyspnea scale 3 (2-3)], respectively. In a Cox PH model including SNIP and the BODE index, only SNIP was a significant predictor of mortality [the hazard ratio: 0.935, 95% confidence interval (CI) 0.882-0.992, p=0.026]. The AUC of SNIP was 0.787 (95% CI 0.588-0.987, p=0.016), and a cut-off value was 35 cmH2O (sensitivity 71%, specificity 83%). Conclusions: We found that SNIP may be more beneficial than the BODE index as a predictor of 1-year mortality after exacerbation of COPD.