An evaluation of the clinical and cost-effectiveness of pulmonary artery catheters in patient management in intensive care: a systematic review and a randomised controlled trial.
Harvey S., Stevens K., Harrison D., Young D., Brampton W., McCabe C., Singer M., Rowan K.
OBJECTIVES: To evaluate the clinical and cost-effectiveness of managing critically ill patients in adult, general intensive care with or without pulmonary artery catheters (PACs). DESIGN: An open, multi-centre, randomised controlled trial with economic evaluation (cost-utility and cost-effectiveness analysis). SETTING: The setting was general (mixed medical/surgical) intensive care units (ICUs) in the UK admitting adults. PARTICIPANTS: Adult patients in participating ICUs deemed by the responsible treating clinician to require management with a PAC. INTERVENTIONS: These were insertion of a PAC and subsequent clinical management, at the discretion of the responsible treating clinicians, using data derived from the PAC. The control group were managed without a PAC but with the option of using alternative cardiac output monitoring devices. MAIN OUTCOME MEASURES: The main outcome measure was hospital mortality. Secondary outcome measures were length of stay in the ICU, length of stay in an acute hospital and organ-days of support in the ICU. For the economic evaluation, the main outcome measure was quality-adjusted life-years (QALYs) and the secondary outcome measure was hospital mortality. RESULTS: Sixty-five ICUs in the UK participated. Of these, 43 (66%) used alternative cardiac output monitoring devices in control group patients. A total of 1263 patients were identified as being eligible for the trial. Of these, 1041 (82.4%) were randomised and allocated to management with (n = 519) or without (n = 522) a PAC. There were no losses to follow-up. However, 27 patients (13 in the PAC group and 14 in the control group) were withdrawn from the trial because either the patient withdrew consent on recovering mental competency or the relatives withdrew agreement following randomisation. Data on 1014 patients were included in the analysis. Participants in the two groups had similar baseline characteristics. There was no difference in hospital mortality for patients managed with (68.4%) or without (65.7%) a PAC. The adjusted hazard ratio (PAC versus no PAC) was 1.09 [95% confidence interval (CI) 0.94 to 1.27]. There was no difference in the median length of stay in ICU, the median length of stay in an acute hospital or mean organ-days of support in ICU between the two groups. The economic evaluation found that the expected cost per QALY gained from the withdrawal of PAC was 2985 pounds. The expected cost per life gained from the withdrawal of PAC was 22,038 pounds. CONCLUSIONS: Clinical management of critically ill patients with a PAC, as currently practised in the UK, neither improves hospital survival for adult, general intensive care patients nor reduces length of stay in hospital. The lack of demonstrable benefit from a device previously believed to be beneficial could be explained by statistical chance, by misinterpretation of PAC-derived data, by ineffective treatment strategies based on data correctly interpreted using the current paradigm or by subsequent inaction following insertion of the device. It is also possible that detailed data on haemodynamics, however used, cannot modify the disease process sufficiently to influence disease outcome. The economic evaluation, using decision analysis techniques rather than conventional hypothesis testing, suggests that the withdrawal of the PAC from routine clinical practice in the NHS would be considered cost-effective in the current decision-making climate, and might result in lives or life-years being saved at modest cost. With the declining use of PACs in the UK and the findings of this report indicating no overall benefit from management with a PAC, it should now be possible to examine protocolised management with a PAC in selected groups of critically ill patients against appropriate controls, something that was difficult while PACs were the considered standard of care.