Pulmonary artery catheters

The description of the flow directed pulmonary artery catheter in 1970 and its subsequent commercial development allowed clinicians to measure cardiac output and derive central haemodynamic parameters at the bedside.¹ On the assumption that measuring and manipulating these variables would improve outcomes, this device became widely used in intensive care units. Very soon reports of serious complications including potentially fatal pulmonary artery rupture appeared,² and arguments for and against its use have continued ever since. A recent addition to the debate is a comprehensive evaluation of both the clinical and cost effectiveness of such catheters in adult intensive care units in the United Kingdom.³ The research was commissioned by the UK's National Health Service Health Technology Assessment (HTA) programme; the authors were from the Intensive Care National Audit and Research Centre, practising intensive care clinicians, and academics. Given 35 years of research and debate, clinicians might reasonably ask what is known about the usefulness of pulmonary artery catheters, what does this report add, is further research needed, and should they use these devices?

Research findings to date

The best evidence comes from well designed and conducted clinical trials in the general intensive care population,⁴ and more selected populations of high risk or critically ill patients.^5-7 None of the trials show that using a pulmonary artery catheter benefits patients. A recently published trial in patients with acute lung injury confirmed these findings.⁸ A meta-analysis of 13 trials reported no overall effect of using these devices on mortality or length of hospital stay.⁹

The HTA report includes a systematic review that reaches the same conclusion, a more detailed description of the authors' previously published PACMAN trial,⁴ and an economic evaluation examining the cost effectiveness of withdrawing these catheters from UK intensive care units.³ Given the pre-existing evidence that the use of these devices does not benefit patients, the report's conclusion that withdrawing pulmonary artery catheters may reduce mortality at the cost of £2985 (€4446; $5622) per quality adjusted life year deserves closer scrutiny.

The authors examined cost effectiveness from a decision science perspective,¹⁰ an approach that many clinicians will be unfamiliar with. In essence, whereas traditional hypothesis testing seeks to detect differences that are not likely to have arisen by chance, decision analysis seeks to determine what is most likely to happen if a particular course of action is taken. In this case, the most likely result of withdrawing pulmonary artery catheters from UK intensive care units is a net health gain at moderate cost. We also note that the conclusions are based upon assumptions regarding the duration and quality of life of survivors of critical illness, as the authors could find no appropriate health related quality of life data for survivors of intensive care units. The assumptions on which the analysis is based are open to question, and the results may not hold true for other healthcare systems.

Is the lack of benefit specific to the pulmonary artery catheter or does it extend to other devices that measure and monitor central haemodynamics? At present that question cannot be answered with certainty, but in the absence of benefit from the pulmonary artery catheter, the onus should be on the marketers of other haemodynamic monitors to demonstrate their clinical usefulness and cost effectiveness, not simply show that they are as good as a device that has no proved benefit.

Future research

Is any further research needed? While it is tempting to say no, we should perhaps start with the construct that measurement of central haemodynamics does not benefit critically ill patients. This may be because manipulation of haemodynamics beyond that which can be achieved through simple clinical examination and parameters such as heart rate, blood pressure, peripheral perfusion, and urine output does not alter outcome. It may equally be that manipulation of central haemodynamics can improve outcome but as yet we just do not know how. Any future research should therefore evaluate specific treatment protocols in specific subgroups of patients. The HTA report highlights the lack of data on health related quality of life and life expectancy for patients surviving treatment in an intensive care unit; given that treatment in intensive care units is consuming an ever larger portion of healthcare budgets, systematic collection of these data should be a priority for the critical care community, healthcare funders, and policy makers.

What should clinicians do with all this information? Supporters of the pulmonary artery catheter have rightly pointed out that monitoring devices do not alter mortality, only treatment alters mortality.¹¹ However, at least five large randomised controlled trials tell us that this device does not benefit patients. The trials include those that allowed clinicians to direct therapy and trials in which treatment was protocolised.^4-6,⁸ No large trials challenge these results, and the results are consistent across patient groups and in different countries. Given that the use of pulmonary artery catheters increases the risk of important complications, continued use of these devices is difficult to defend. The onus is now on the proponents of the pulmonary artery catheter and related devices to limit their use to clinical trials and to show that protocols based on such devices do benefit patients.