On All Fools Day, 1949, the Journal of Applied Physiology received a manuscript from Hellems and colleagues describing the indirect measurement of left atrial pressure in man using occlusion of a small branch of the pulmonary artery.1 In the half century since this publication, the pulmonary artery wedge pressure has been used in the cardiac catheter laboratory by generations of cardiologists to diagnose and assess the severity of congenital and acquired cardiac disease and to appropriately select patients for surgical correction. Twenty-one years later, in 1970, Swan and colleagues described similar measurements made using a balloon-tipped flow-directed catheter.2 This development moved the right heart catheter out of the cardiac catheterization laboratory and up to the bedside. Continuous bedside monitoring of haemodynamic variables extended the scope of right heart catheterization beyond simple diagnosis, as manipulation of the various haemodynamic variables for presumed therapeutic benefit became a practical proposition. The best example of this haemodynamic variable-driven therapy is the goal-directed therapy developed by Shoemaker in 1982.3 For the past 20 years, the package of right heart catheter insertion, and measurement of haemodynamic variables and their subsequent manipulation has been taught to every resident in critical care.
The development and subsequent acceptance of right heart catheterization by the critical care community paralleled another far more important and globally applicable development in medical science. The publication of the 1948 Medical Research Council report on the treatment of pulmonary tuberculosis with streptomycin,4 is generally considered to mark the birth of the modern randomized controlled trial (RCT). Like the right heart catheter, the use of the controlled trial as a clinical research tool has increased exponentially over the last 50 years, becoming totally accepted by the 1980s. If bedside right heart catheterization were developed today it would probably not be accepted without evidence from an RCT indicating that it reduced mortality, morbidity, length of stay or the cost of health care. Yet, because of its coincidence with the birth of the RCT, it escaped this formal evaluation and as a result unequivocal evidence for its benefit does not exist.
In 1996, Cooper and colleagues systematically evaluated all the available evidence for benefit from pulmonary artery catheterization.5 Of the 543 studies they identified, only one was considered level I evidence (large randomized trial with low risk of false positive or false negative errors). This was a trial of goal-directed therapy conducted in Italy by Gattinoni and colleagues,6 which showed no difference between treatment and control groups. There was weak level II evidence (small trials, high risk of errors) outside intensive care showing a benefit for right heart catheterization in the pre- and intra-operative care of some high risk surgical patients. They concluded that, for a mixed ICU population, there was currently no proof of benefit for the use of right heart catheterization but also no proof that its use was harmful.
The concept that there might be mortality attributable to or associated with the use of right heart catheterization had been proposed for patients with myocardial infarction, where those patients receiving right heart catheters had a higher mortality (see for instance Gore and colleagues7). However, the studies were simple case series, so it may be that the more severely afflicted patients received the right heart catheters. The first major attempt to deal with the confounding effects of case mix was published by Connors and colleagues in 1996.8 They performed a retrospective analysis of prospectively gathered data from a multi-centre study of decision making in critically ill adults (the SUPPORT project). A total of 2016 patients were included, of which 80% had acute respiratory failure or multi-system organ failure (MSOF); the remainder had chronic obstructive pulmonary disease, chronic heart failure, cirrhosis, non-traumatic coma, metastatic colonic cancer or lung cancer. A propensity score was developed to determine the probability of right heart catheterization and patients who had received a right heart catheter in the ICU within 24 h of admission were matched with another patient with the same diagnosis and propensity score who had not been catheterized. The results showed an increased mortality (odds ratio 1.24) in the patients receiving right heart catheterization.
The study was widely debated and received the expected barrage of criticism associated with any publication that challenges the accepted wisdom. Concerns were expressed in the UK that the study was not relevant to the UK ICU patient population, as patients with lung cancer and metastatic colon cancer are not often treated on UK ICUs. However, 80% of patients studied by Connors and colleagues had acute respiratory failure or MSOF, the bread and butter of UK ICU practice. More importantly, an analysis of the Scottish Intensive Care Society database gave a similar result to the Connors study.9 Concerns were also expressed about the validity of the case matching whether the patients were truly matched or whether there was an unrecorded factor that was influencing both the physicians decision to place a right heart catheter and the patients chance of survival. This is a methodological limitation of case-matching studies and can never be adequately answered with this study design. The only solution is to perform another study using the best available design to remove confounding factors, which is the RCT.
There is obviously not a consensus view on the benefit of right heart catheterization in intensive care in the UK. The correspondence in the British Medical Journal that followed publication of the Connors paper,1012 revealed a spread of opinion which is reflected in the wide range of right heart catheter use between units, varying from 3 to 76% of admissions in a recent survey of 69 UK ICUs conducted by the Intensive Care National Audit and Research centre (ICNARC). However, intensive care clinicians in the UK are clearly united in their view that answering the question is important, ranking evaluation of right heart catheterization amongst the three most important research priorities in a recent study.13 Over 80% of Scottish intensive care clinicians surveyed indicated that they would welcome a clinical trial.14 The NHS Research and Development Health Technology Assessment (NHS R&D HTA) Programme have responded to the specialitys concern by funding the first randomized controlled trial of the use of right heart catheterization in intensive care in the UK, the PAC-Man study (Pulmonary Artery Catheters in patient Management).
This trial presents a unique set of challenges. Unlike most trials, the intervention under investigation is already in widespread use rather than being introduced with the trial. Thus, in participating units half the trial patients will be randomized not to receive a monitoring/treatment package that they would have otherwise received. The only way this can be ethically justified is if the clinicians involved have equipoise; that is, they believe that on current evidence it is impossible to determine if use of the right heart catheter increases, decreases or does not alter risk of mortality for intensive care patients. This is the only rational interpretation that can be placed on the available published data, but it will require courage, clear thinking and intellectual honesty to apply this to individual patients.
What is to happen to patients who refuse consent or whose relatives refuse assent for the study? A physician with true equipoise could take the view that there is no evidence for harm or benefit from the catheter, and could remove it from use for the duration of the trial, so the best available treatment offered to patients outside the trial did not include right heart catheterization. Ethically this is perfectly acceptable as no physician has a duty to provide a therapy of unproven efficacy to any patient. Alternatively, patients who refuse consent or whose relatives refuse assent could all receive a right heart catheter on the basis that even in the absence of objective evidence of benefit, it comprises standard treatment in many units and so is the best available treatment. Ethically, this is again an acceptable view, but it requires investigators to be disciplined and even-handed when recruiting for the trial and to present the arguments clearly, otherwise the natural tendency for relatives to favour the most active treatment will make them refuse assent, leading to poor recruitment and bias.
Clinical trials fall broadly into two categories: efficacy trials and effectiveness trials. Efficacy trials address whether an intervention can produce the desired effect in ideal circumstances. In the context of right heart catheters this translates to a study where the patient group studied, the indications for the catheter, the technique for placement and all subsequent management are rigorously controlled by protocol. This is the approach the American intensive care community has taken with the ARDSNet (http://hedwig.mgh.harvard.edu/ardsnet/ards05.html) study of right heart catheters in acute lung injury. The limitation of this approach is that the results are really only applicable to the patient group studied and the protocol used. In the UK, the NHS R&D HTA who are supporting the PAC-Man study have agreed to fund a pragmatic or effectiveness approach, where the use or otherwise of the catheter is determined by randomization but little more is controlled. The advantage with this type of trial is that the results can be immediately used to inform changes in the healthcare process to the advantage of a large number of patients. The disadvantage is that potential benefits to small well-defined subgroups of patients are masked and the number of patients required is larger. Both approaches are vulnerable to the naysayers, who would claim that either design did not address the particular group of patients they treat.
For the PAC-Man study to work as an effectiveness trial, all patients suitable for right heart catheterization on an ICU will have to be entered into the trial. The idea that some patients are too ill to be denied the benefits of right heart catheterization immediately introduces a major bias in the study, removing the very patients likely to show a benefit. This problem has already stopped one randomized controlled trial of right heart catheterization undertaken in 1991 in Canada, where only 33 of 148 (22%) of eligible patients entering the ICUs were recruited.15 The physicians involved thought it was unethical not to catheterize the rest. They clearly did not have equipoise. We hope that with passage of time and more informed debate will ensure that the same situation does not occur in the UK.
The PAC-Man study is a unique opportunity for the UK intensive care community. Years of work by individuals and professional bodies have raised the profile of the specialty to the point where we have been entrusted with a large portion of the NHS R&D HTA budget to run a major study that will involve 6000 patients and a third of UK ICUs. The study will answer a question that no one in the worldwide ICU community has been able to address properly for 30 years. It is perhaps unfortunate that we could not make the first major, non-commercial, multi-centre clinical trial in UK ICUs a study of a new and exciting therapy, but the questions about right heart catheterization are important both to individuals and for the credibility of the specialty. The study will set up the infrastructure for many more trials and, with the involvement of such a large proportion of the UK intensive care community, allow important questions to be efficiently and quickly addressed.
Participation in the PAC-Man study will require intensive care physicians to take a clear and objective assessment of the evidence on which they base one of the cornerstones of their practice, and suspend their prejudices for a year whilst the study provides the definitive answer. The rewards to the speciality for this intellectual courage are enormous. The penalty for nihilism may be the attentions of the National Institute for Clinical Excellence (NICE), which came into existence on All Fools Day 1999, exactly half a century after Hellems paper was submitted.
J. D. Young
Radcliffe Infirmary
Oxford OX2 6HE
UK
References
1 Hellems HK, Haynes FW, Dexter L. Pulmonary capillary pressure in man. J Appl Physiol 1949; 2: 249
2 Swan HJC, Ganz W, Forrester J, Marcus H, Diamond G, Chonette D. Catheterization of the heart in man with the use of a flow-directed balloon-tipped catheter. JAMA 1970; 283: 44751
3 Shoemaker WC, Appel PL, Waxman K, Schwartz S, Chang P. Clinical trial of survivors cardiorespiratory patterns as therapeutic goals in critically ill postoperative patients. Crit Care Med 1982; 10: 398403[ISI][Medline]
4 Medical Research Council. Streptomycin treatment of pulmonary tuberculosis. BMJ 1948; ii: 76982
5 Cooper AB, Doig GS, Sibbald WJ. Pulmonary artery catheters in the critically ill. An overview using the methodology of evidence-based medicine. Crit Care Clin 1996; 12: 77794[ISI][Medline]
6 Gattinoni L, Brazzi L, Pelosi P, et al. A trial of goal-oriented hemodynamic therapy in critically ill patients. SvO2 Collaborative Group. N Engl J Med 1995; 333: 102532
7 Gore JM, Goldberg RJ, Spodick DH, Alpert JS, Dalen JE. A community-wide assessment of the use of pulmonary artery catheters in patients with acute myocardial infarction. Chest 1987; 92: 7217[Abstract]
8 Connors AF Jr, Speroff T, Dawson NV, et al. The effectiveness of right heart catheterization in the initial care of critically ill patients. SUPPORT Investigators. JAMA 1996; 276: 88997[Abstract]
9 MacKirdy FN, Livingston BM, Howie JC. The relationship between the presence of pulmonary artery catheters and the case mix adjusted outcome of patients admitted to Scottish ICUs. (Abstract). Clin Intens Care 1997; 8: 9
10 Dexter TJ. Pulmonary artery catheterisation. Catheterisation is useful when used by experienced clinicians. BMJ 1996; 313: 1328
11 Lazarus R. Pulmonary artery catheterisation. Proposed moratorium on use of pulmonary artery catheters is premature. BMJ 1996; 313: 1328
12 Konarzewski W. Pulmonary artery catheterisation. Pulmonary artery catheters should be banned from intensive care units. BMJ 1996; 313: 1328
13 Vella K, Goldfrad C, Rowan K, Bion J, Black N. Use of consensus development to establish national research priorities in critical care. BMJ 2000; 320: 97680
14 MacKirdy FN. The value of the pulmonary artery catheter. Br J Intens Care 1999; May/June: 98
15 Guyatt G. A randomised controlled study of right heart catheterisation in critically ill patients. J Intens Care Med 1991; 6: 915[Medline]