Elsevier

The Lancet

Volume 355, Issue 9219, 3 June 2000, Pages 1931-1935
The Lancet

Articles
Early use of non-invasive ventilation for acute exacerbations of chronic obstructive pulmonary disease on general respiratory wards: a multicentre randomised controlled trial

https://doi.org/10.1016/S0140-6736(00)02323-0Get rights and content

Summary

Background

Within the intensive-care unit, non-invasive ventilation (NIV) can prevent the need for intubation and the mortality associated with severe episodes of chronic obstructive pulmonary disease (COPD). The aim of this study was to find whether the introduction of NIV, early after the admission on a general respiratory ward, was effective at reducing the need for intubation and the mortality associated with acute exacerbations of COPD.

Methods

We did a prospective multicentre randomised controlled study comparing NIV with standard therapy in patients with mild to moderate acidosis. NIV was administered on the ward with a simple non-invasive ventilator and a standardised predefined protocol. Patients were recruited from 14 UK hospitals over 22 months.

Findings

236 patients were recruited, 118 received standard therapy alone and 118 additional NIV. The two groups had similar characteristics at enrolment. The use of NIV significantly reduced the need for intubation as defined by the failure criteria. 32/118 (27%) of the standard group failed compared with 18/118 (15%) of the NIV group (p=0·02). In-hospital mortality was also reduced by NIV, 24/118 (20%) died in the standard group compared with 12/118 (10%) in the NIV group (p=0·05). In both groups pH, PaCO2, and respiratory rate improved at 4 h (p<0·01). However, NIV led to a more rapid improvement in pH in the first hour (p=0·02) and a greater fall in respiratory rate at 4 h (p=0·035). The duration of breathlessness was also reduced by NIV (p=0·025).

Interpretation

The early use of NIV for mildly and moderately acidotic patients with COPD in the general ward setting leads to more rapid improvement of physiological variables, a reduction in the need for invasive mechanical ventilation (with objective criteria), and a reduction in in-hospital mortality.

Introduction

Prospective randomised controlled trials of non-invasive ventilation (NIV) within the intensive-care unit (ICU) have shown reductions in the need for intubation and in-hospital mortality associated with severe exacerbations of chronic obstructive pulmonary disease (COPD).1, 2, 3 The evidence from randomised controlled trials within the accident and emergency department and general respiratory ward is mixed.4, 5, 6, 7 Bott and colleagues, using research staff to initiate ventilation on the ward, showed that NIV improved pH and PaCO2.4 On an intention-to-treat analysis there was no survival benefit from NIV, until those unable to tolerate NIV were excluded. The use of research staff makes it difficult to generalise the results of this study to routine clinical practice. Angus and colleagues randomised 17 patients to either NIV (n=9) or conventional therapy plus doxapram (n=8). There was a non-significant trend to improved survival in the NIV group with three out of eight patients dying with conventional care and nine out of nine surviving with NIV.5 Barbe and colleagues found NIV unnecessary because none of the 24 patients in their study died or required intubation.6 Wood and colleagues found that NIV delayed intubation and increased in-hospital mortality.7 However, of the 27 patients in this study only six had COPD, the groups were not well matched, and the level of pressure support was modest.

The use of NIV in general respiratory wards could theoretically allow the earlier use of NIV during an exacerbation and widen the availability of NIV. This is particularly important in the UK where there is a shortage of ICU and high-dependency bed provision.

We therefore aimed to find out whether NIV was feasible on the ward in non-specialist units and whether it was effective at reducing the need for intubation and in-hospital mortality, compared with standard treatment, in patients admitted with mild-to-moderate acidosis due to an exacerbation of COPD. Patients with acidosis are at higher risk of subsequent intubation and death than patients with a normal pH.8, 9 The null hypothesis was that NIV on the ward would not reduce the need for intubation in this clinical group.8

Section snippets

Patients

Between November, 1996, and September, 1998, adult patients who were in hospital because of an acute exacerbation of COPD were prospectively recruited from 14 hospitals in the UK: Airedale District Hospital, Birch Hill Hospital (Rochdale), Bradford Royal Infirmary, Castle Hill Hospital (Hull), Churchill Hospital (Oxford), Dewsbury and District Hospital, Doncaster Royal Infirmary, Halifax General Hospital, Huddersfield Royal Infirmary, Killingbeck Hospital (Leeds), Leeds General Infirmary,

Characteristics of patients

118 patients were randomly assigned to standard treatment and 118 to NIV (figure 1). The two groups had similar characteristics on admission (table 1). The use of aminophylline and doxapram was not different between the two groups.

Setting

In each hospital between one and three general medical/respiratory wards were identified as sites for NIV. Of the 25 wards involved in the trial, 22 had no experience of NIV and only one was fully experienced. None had previously used the study ventilator. In none of

Discussion

This study shows that the use of NIV on general respiratory wards is both feasible and clinically effective at reducing the demand for invasive ventilatory support and the in-hospital mortality associated with acute ventilatory failure in patients with a clinical diagnosis of COPD.

NIV could be established (defined as > 1 h of use) in 93% of patients, similar to the 87% described by Bott and colleagues,4 and only consumed an additional 26 min of nursing time. In a low nurse-to-patient setting

Cited by (1039)

View all citing articles on Scopus
View full text