A Randomized, Controlled Trial of the Use of Pulmonary-Artery Catheters in High-Risk Surgical Patients
Some observational studies suggest that the use of pulmonary-artery catheters to guide therapy is associated with increased mortality.
The authors performed a randomized trial comparing goal-directed therapy guided by a pulmonary-artery catheter with standard care without the use of a pulmonary-artery catheter. The subjects were high-risk patients 60 years of age or older, with American Society of Anesthesiologists (ASA) class III or IV risk, who were scheduled for urgent or elective major surgery, followed by a stay in an intensive care unit. Outcomes were adjudicated by observers who were unaware of the treatment-group assignments. The primary outcome was in-hospital mortality from any cause.
Of 3803 eligible patients, 1994 (52.4 percent) underwent randomization. The base-line characteristics of the two treatment groups were similar. A total of 77 of 997 patients who underwent surgery without the use of a pulmonary-artery catheter (7.7 percent) died in the hospital, as compared with 78 of 997 patients in whom a pulmonary-artery catheter was used (7.8 percent) - a difference of 0.1 percentage point (95 percent confidence interval, -2.3 to 2.5). There was a higher rate of pulmonary embolism in the catheter group than in the standard-care group (8 events vs. 0 events, P=0.004). The survival rates at 6 months among patients in the standard-care and catheter groups were 88.1 and 87.4 percent, respectively (difference, -0.7 percentage point [95 percent confidence interval, -3.6 to 2.2]; negative survival differences favor standard care); at 12 months, the rates were 83.9 and 83.0 percent, respectively (difference, -0.9 percentage point [95 percent confidence interval, -4.3 to 2.4]). The median hospital stay was 10 days in each group.
Conclusions We found no benefit to therapy directed by pulmonary-artery catheter over standard care in elderly, high-risk surgical patients requiring intensive care.
This is an interesting and important article. The array of physiological values laid bare by the use of pulmonary artery catheters is well known to all intensivists. According to current estimates, there are more than 1.2 million pulmonary-artery catheters placed annually in the United States, with associated costs of over $2 billion. The uses of the pulmonary-artery catheter expanded from diagnosis alone to include the direction of therapy. In the initial enthusiastic use of this procedure, no clinical trials were conducted to determine whether patient outcomes were altered by the data derived from insertion of these catheters or the associated therapeutic interventions. Benefit was simply assumed. Debates continued to rage about the appropriateness of its use in given clinical situations. Its cost and the need to develop expertise to interpret the obtained values were the limitations for its routine use in Indian ICUs. By the time our ICUs were evolving from these limitations, recent articles started raising the questions of the utility and mortality.
In 1996, Connors et al  published the results of an observational study involving 5735 critically ill medical and surgical patients concluding that pulmonary-artery catheters were associated with an increase in mortality. The response to that trial in the medical literature and the lay press clearly illustrated the state of chaos regarding the use of pulmonary-artery catheters. Recent nonrandomized clinical studies involving patients during the peri operative period, including a meta-analysis of the literature and a prospective, observational study also showed that pulmonary-artery catheters are not beneficial and that their use may be associated with increased morbidity.
It is thus important and interesting to see this study which raised the question of utility and risks of PA catheters. Its importance also lies in the fact that progress is being made toward a definition of the appropriate role of the pulmonary-artery catheter in clinical practice. This study is a prospective, randomized, controlled clinical trial comparing therapy directed by a pulmonary-artery catheter with standard care in 1994 high-risk surgical patients. There was no significant difference in the in-hospital mortality between the groups (7.8 percent in the group with pulmonary-artery catheters vs. 7.7 percent in the standard-care group), but there was a significant increase in the incidence of pulmonary embolism in the catheter group. In the catheter group more patients received inotropic agents, vasodilators, anti hypertensive medication, packed red cells and colloid, than in the standard-care group, indicating that the study protocol resulted in differential treatment in the two groups.
Sandham and colleagues have substantially furthered the progress in research in critical care with their current study. The benefits of this study are clear. It shows that:
A. Physicians will allow their patients to be randomly assigned to either treatment group in a clinical trial that involves not only the insertion of a pulmonary-artery catheter, but also the implementation of therapy directed by that catheter
B. It is feasible to conduct large, adequately powered, multi center, controlled trials of pulmonary-artery catheterization.
Any study that questions the utility of PA catheters needs to be interpreted in the light of the expertise in interpretation of the information gathered from PA catheter and the therapeutic decisions taken on the basis of this interpretation. It is obvious that you cannot blame a tool if the workman is bad.
As pointed out by Vincent recently, there are a number of limitations of this NEJM study by Sandham et al. These include:
A. The overall mortality rate was as low as 7.7%, indicating that the study population was not acutely ill. Studies in the past have shown that low risk patients may not benefit from PA catheter insertion.
B. The timing of the insertion of PA catheter has not been mentioned. It is presumed that the insertion was done at the time of anaesthesia. An earlier insertion with prior optimization has been shown to improve survival in some studies.
C. The therapeutic protocol seemed arbitrary.
D. The results of this specific peri- operative population cannot be applied to all the critically ill patients.
Thus it is clear that PA catheters can still help in making therapeutic decisions in certain groups of patients. It remains to be seen whether the results of this trial extend to groups of patients other than the high-risk surgical patients who were studied.