Format

Send to

Choose Destination
Respir Med. 2010 Sep;104(9):1319-25. doi: 10.1016/j.rmed.2010.03.001.

Sniff nasal inspiratory pressure versus IC/TLC ratio as predictors of mortality in COPD.

Author information

1
Royal Brompton Hospital, London SW3 6NP, UK. a.moore@ic.ac.uk

Abstract

BACKGROUND:

Hyperinflation is a recognized adverse prognostic factor in COPD. As the sniff inspiratory nasal pressure (SnIP) principally reflects the severity of hyperinflation in COPD, we hypothesized that it might also be a predictor of mortality. We therefore compared the SnIP to the inspiratory capacity-to-total lung capacity (IC/TLC) ratio as predictors of mortality in advanced COPD.

METHODS:

A retrospective mortality analysis of 110 patients with COPD (mean FEV(1) 1.01litres, 37% predicted; 66% male) was performed. All patients had SnIP and lung volume measurements performed. The power of each test to predict mortality was determined, and predicted survival curves were created for both the SnIP and IC/TLC ratio.

RESULTS:

37 patients (34%) died during the study period (29 male, 8 female). Mortality rates were analysed with a Chi(2) test; there was a significant trend towards male death (mortality rate male vs. female; 39.7% vs. 21.6% respectively; chi(2)p=0.058, Chi 3.6). ROC curves demonstrated that both SnIP and IC/TLC ratio are predictors of mortality, but analysis by Cox proportional hazards suggested the SnIP has a stronger predictive power (SnIP vs. IC/TLC ratio; p=0.017 vs 0.525; HR 0.97 vs 0.99 respectively), and analysis of the area under ROC curves (AUC) suggest that SnIP is a better discriminator than IC/TLC ratio (AUC SnIP vs IC/TLC; 0.679 vs 0.618).

CONCLUSIONS:

The SnIP conveys at least as much predictive power for mortality in COPD as hyperinflation determined by IC/TLC ratio. This test is cheaper, quicker and easier than measuring lung volumes by plethysmography.

PMID:
20399631
DOI:
10.1016/j.rmed.2010.03.001
[Indexed for MEDLINE]
Free full text

Supplemental Content

Full text links

Icon for Elsevier Science
Loading ...
Support Center