Web site designed and maintained by Chris Cairns  © SICS EBM Group 2004                                  

The pulmonary artery catheter in the critically ill ICU patients

Citation/s: Harvey S, Harrison DA, Singer M et al. Assessment of the clinical effectiveness of the pulmonary artery catheter; a randomised controlled trial. Lancet 2005; 366: 472- 477.

Lead author's name and fax: Harvey S et al (ICNARC), correspondence to Mervyn Singer: m.singer@ucl.ac.uk

Three-part Clinical Question: Does the utilisation of the pulmonary artery catheter in critically ill ICU patients improve outcome.

Search Terms: Swan-Ganz catheter, pulmonary artery catheter, intensive care, monitoring

The Study: Single-blinded randomised controlled trial with intention-to-treat.
 

The Study Patients: Critically ill patients admitted to ICU who were considered by treating clinician as someone who should be managed with a PAC..

Control group (N = 507; 507 analysed): These patients were not permitted to be managed with a pulmonary artery catheter (PAC) but could be either managed without a pulmonary artery catheter (strata A [107 patients]) or with an alternative cardiac output monitor (strata B[400 patients]).
 

Experimental group (N = 506; 506 analysed): These patients had a PAC inserted but the timing of insertion and subsequent management were at the discretion of the treating clinician.

The Evidence:

EBM Comments:

1. Do the methods allow the adequate testing of the hypothesis- Yes. This is a well-designed randomised controlled trial with low risk of bias, despite reduction in sample size and early inspection of control group outcomes. Weaknesses relate to lack of blinding which would have been impossible and the control group being allowed to utilise other form of cardiac output device. This is therefore a study of PAC or not, and not a study of PAC versus other monitor or of cardiovascular flow measurement versus no measurement. The authors also added a cost-effectiveness analysis and suggested that the cost per QALY for removing the PAC from clinical practice would be £2980. This analysis is limited in its validity and does not represent a proper cost-effectiveness analysis. Saying that, in light of the lack of effectiveness of PAC the result is likely to be approximately correct.

2. Do the statistical tests correctly test the results to allow differentiation of statistically significant result- Yes they do. All statistical analysis is performed correctly.

3. Are conclusions valid in light of results:- Yes they are. They state that there is no overall evidence of benefit or harm for the PAC. They also state that other cardiac output technologies need to be assessed. Finally they state that although skilled use of the PAC may offer advantage the deceasing use of this device may make retention of skills more difficult.

4. Did results get omitted, and why- No, there was appropriate intention-to-treat analysis and no patients were omitted.

5. Did they suggest areas of further research- Yes, they suggest evaluation of other cardiac output technologies.

6. Did they make recommendations based on results and were they appropriate- They did suggest that optimum management protocols to identify the groups of patients that would gain from management with a PAC before the technology becomes redundant.

7. Is this study relevant to my clinical practise? Yes

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

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

10. What grade of recommendation can I make when this study is considered along with other available evidence- A. Other randomised evidence demonstrates similar results although this disagrees with the results of Conner’s retrospective work that provoked much of the debate.

11. Should I change my practice because of these results- The PAC is already declining in use through out the world. This study suggests that there is no advantage in PAC use at least compared to other cardiac output devices. Sadly, it does not compare PAC versus no cardiac output device as clinicians clearly did not have equipoise on this issue. You could take a number of conclusions from this study:

    • If you use the PAC you could argue for continuing its’ use as it does not cause harm (10% complication rate mainly related to central venous access rather than PAC itself) or excess mortality. Clearly there are costs associated with its continued use but these are not much different from the majority of non-invasive technologies and may cancel out (although note cost per QALY for withdrawal £2980).

    • If you don’t use the PAC but do use other cardiac output technologies you could argue to continue with your current strategy. It is likely that in the future we will have to do another RCT looking at these devices versus no device as this reviewer does not believe that any form of cardiac output device is effective.

    • Finally, you could argue that you can continue using PAC in patients with complex haemodynamic problems and all other patients use no monitoring device. In this case you will have to work hard to maintain clinical skills and should audit your complication rate as experience levels fall. It could also be argued that you should insert the catheter, perform a range of cardiovascular interventions and then remove the catheter shortly afterwards to minimise complication rates. In other words more for diagnosis than monitoring. This reviewer favours this final approach.

12. Should I audit my current practice because of these results- Yes you should. It could also be argued that you should audit every use of the PAC the look at trends in mortality and complications and this role could be performed by central audit groups such as SICSAG and ICNARC.

Appraised by: Brian H Cuthbertson ICU, Aberdeen Royal Infirmary Aberdeen; 23 September 2005
Email: b.h.cuthbertson@abdn.ac.uk

Reviewed & Edited by SJM & CC

Kill or Update By: 23rd September 2008

Citation: EBM Critical Appraisals. Scottish Intensive Care Society EBM Group. Cuthbertson B. 2005. : Harvey S, Harrison DA, Singer M et al. Assessment of the clinical effectiveness of the pulmonary artery catheter; a randomised controlled trial. Lancet 2005; 366: 472- 477.

Printer friendly view

©SICS EBM 2005