Protective ventilatory strategy in severe ARDS

 

 

Protective ventilation strategy reduced 28 day mortality but not hospital mortality in patients with severe ARDS with to Murray Lung Injury Score > 2.5 Level of Evidence 1-

 

 

Citation/s:
Amato MBP et al. Effect of protective-ventilation strategy on mortality in the Acute Respiratory Distress Syndrome. N Engl J Med. 1998;338:347-54

Lead author's name and fax:

MBP Amato , Respiratory intensive care unit, Pulmonary Division, Hospital das Clinicas, University of Sao Paulo, Brazil

 

Three-part Clinical Question:

 

1.    Patients- ARDS according to a Murray Lung Injury Score of > 2.5 and a PAOP < 16

2.      The treatment- Protective ventilation strategy vs Conventional ventilation

3.      The outcome- Primary 28 day mortality Secondary Hospital mortality, barotrauma, weaning rates

 

Search Terms:

ARDS- therapy Limited to Prospective study / or randomised study / or random allocation; double blind method

 

The Study:
Non-blinded concealed randomised controlled trial without intention-to-treat.

The Study Patients:

Patients with ARDS according to Murray criteria > 2.5, of all aetiologies, requiring ventilatory support in ICU for < 1 week

Control group (N = 29; 25 analysed): Targeting PaCO2 35-38mmHg regardless of airway pressure, A FiO2 of < = 0.6 with adequate systemic oxygen delivery (not defined), Stepwise PEEP algorithm, tidal volume 12ml/kg, respiratory rate 10-24 per minute, square wave flow 50-80l/min

Experimental group (N = 24; 23 analysed): Tidal volume 6ml/kg, respiratory rate < 30/min, permissive hypercapnia up to 80mmHg (10.5kPa) keeping pH > 7.2 with sodium bicarbonate infusion, Driving pressure (Pplat - PEEP < 20cmH2O and peak pressure < 40cmH2O, pressure limited modes only, PEEP set at 2cm above Pflex.

The Evidence:

Outcome	Time to Outcome	CER	EER	RRR	ARR	NNT
Mortality	28 day	0.379	0.708	-87%	-0.329	-3
	95% Confidence Intervals:			-154% to -20%	-0.583 to -0.075	-13 to -2
Mortality	Hospital	0.448	0.708	-58%	-0.260	-4
	95% Confidence Intervals:			-115% to -1%	-0.517 to -0.003	-295 to -2
Barotrauma	During ventilation	0.069	0.41	-494%	-0.341	-3
	95% Confidence Intervals:			-809% to -179%	-0.558 to -0.124	-8 to -2

 

 

Non-Event Outcomes	Time to outcome/s	Control group	Experimental group	P-value
Weaning	28 days	19	7	0.001

 

Comments:
A small RCT without blinding or intention to treat analyses that was stopped at 5th interim analysis due to improved outcome in treatment group in terms of primary outcome measure (28 day mortality). The trial failed to show a difference in hospital mortality (approached significance). This trial also included the results of a previous study which was published under the same first authors name 3 years previously.

 

EBM summary questions:

1. Do the methods allow the adequate testing of the hypothesis- Yes, this is a quite well designed study, despite it inability to eliminate bias due to its lack of blinding and intention to treat, and its methods allow testing of the hypothesis.

2. Do the statistical tests correctly test the results to allow differentiation of statistically significant result- No power calculation was performed /. Presented for this study. The authors introduced a interim analysis system which was performed five times and eventually leads to the trial being stopped due to an improvement in the primary outcome measure. However, this premature stopping of the study lead to a hospital mortality only showing a trend to improvement and thus an important result may have been lost. Otherwise stats were appropriate.

3. Are conclusions valid in light of results:- They conclude a reduction in 28 day mortality (their primary outcome measure) with the protective ventilatory strategy as well as a reduction in the incidence of barotraumas. They also conclude that they were unable to show a benefit in terms of hospital mortality and give many reasons why this should be the case. There conclusions are valid within the limitations of their results.

4. Did results get omitted, and why- Yes, 4 patients in the protective group and one in the conventional group were not analysed due to protocol violations. There was no intention-to-treat analysis.

5. Did they suggest areas of further research- No they do not.

6. Did they make recommendations based on results and were they appropriate- No

7. Is this study relevant to my clinical practice- The ARDS definition was not the American –European Criteria now so widely used and accepted and there group represents a severe group of ARDS patients. They also included ARDS patients with diagnosis that are very uncommon in western Europe (Leptospirosis is the commonest individual cause). The control group mortality of 71% would also be consider high by modern quoted mortality rates from clinical trials although it is not that much higher than the Scottish ARDS study results. So there are some problems with comparing these results with our practice.

8. What level of evidence does this study represent- 1 negative

9. What grade of recommendation can I make on this result alone- B

10. What grade of recommendation can I make when this study is considered along with other available evidence- A

11. Should I change my practice because of these results- No not on these results alone

12. Should I audit my current practice because of these results- Yes, ideal opportunity in light of more current evidence to re-audit Scottish outcomes for ARDS with volume limited strategies.

 

Appraised by: Brian H Cuthbertson Intensive Care Unit, Aberdeen Royal Infirmary Aberdeen; Monday, November 04, 2002
Email: b.h.cuthbertson@abdn.ac.uk
Kill or Update By: 3rd November 2005

 

Printer-friendly format