Abstract & Commentary
Pulmonary artery catheters (PACS) are widely used in critically ill patients. Proponents of the catheter, introduced into the clinical arena more than 30 years ago, argue that physiologic data provided by the use of the PAC permit clinicians to target treatment and improve patient outcomes. Despite its widespread use, the risks, benefits, and appropriate use of the PAC in specific clinical situations have not been identified. The purpose of this study was to compare therapy guided by a PAC to therapy guided without a PAC among high-risk elderly patients undergoing surgery followed by a stay in the ICU.
The study, conducted over a 9-year period (1990-1999) by the Canadian Critical Care Clinical Trials Group, is the first randomized, controlled trial published that has evaluated the effect of the use of the PAC on mortality and morbidity in "high-risk" surgical patients. Nineteen centers in Canada participated. Patient eligibility criteria are summarized in Table 1. Patients randomized to the standard-care group did not receive a PAC (measurement of central venous pressure with a central line was allowed). Patients randomized to the PAC group had a PAC placed before surgery and treatment was directed to physiologic goals as summarized in Table 2. Clinical and outcome data were collected 24 hours postoperatively, weekly during the ICU stay and hospital stay, and 6 and 12 months postoperatively.
|
Table 1. |
|
Eligibility Criteria |
|
• Age > 60 years |
|
• ASA Class III or IV Risk |
|
• Scheduled for urgent/elective major surgery (abdominal, thoracic, vascular, hip) |
|
Table 2. |
|
|
Physiologic Goals |
|
|
Oxygen delivery (DO2) |
550-600 mL/min/m2 |
| Cardiac index (CI) | 3.5-4.5 L/min/m2 |
| Mean arterial pressure | 70 mm Hg |
| Pulmonary artery wedge pressure | 18 mm Hg |
|
Heart rate |
< 120 beats/min |
| Hematocrit | > 27% |
During the 9-year data collection period, 3808 patients were screened and 1994 (52.4%) underwent randomization, with 997 patients in each group (satisfying a priori power analysis criteria). In the standard group, 945 patients (94.8%) received planned therapy; 2.4% crossed over to the PAC group. More patients in the PAC group than the standard care group received inotropic agents, vasodilators, antihypertensives, RBC transfusions, and colloids. Goals for CI and DO2 index were met in 18.6% and 21% of patients at entry and in 79% and 62% of patients, respectively, after surgery.
The baseline characteristics of the patients in both groups were similar in terms of the Goldman Index, vital capacity, forced expiratory volume in the first second, and blood values for hemoglobin, bilirubin, and creatinine. There was no difference in median hospital length of stay between the 2 groups (10 days; P < 0.41) and no difference in in-hospital mortality (7.7%, 7.8%; P < 0.93). There were no differences in morbidity between the 2 groups for acute myocardial infarction, congestive heart failure, dysrhythmias, hepatic or renal insufficiency, catheter-related sepsis, wound infection, pneumonia, or adverse catheter related events. There was a higher incidence of pulmonary embolism in patients in the PAC group (0 events in the standard group vs 8 events [0.8%] in the PAC group; P < 0.004). However, thromboprophylaxis was used in 90.9% of the standard care group and in 88.1% in the PAC group (P < 0.05). (Sandham JD, et al. N Engl J Med. 2003;348:5-14).
Comment by Karen Johnson, PhD, RN
This study provides us with 2 important contributions to the highly debated issue of whether the use of PACs makes a difference in the care of critically ill patients. First, it demonstrated that patients with a PAC did not have a higher mortality than patients who did not have a PAC, as has been previously reported.1 Secondly, an randomized, clinical trial on the use of PACs can be done in terms of the feasibility of conducting a large multicentered investigation and in terms of physician cooperation. The Canadian Critical Care Clinical Trials Group must be congratulated for pulling this study off!
How should we change our care based on these results? The editorial accompanying this paper states that these findings should affect patient care and that the "routine insertion of PACs perioperatively in high risk surgical patients is not warranted."2 Time out! This is an inflammatory generalization! Before we change patient care, we need to address some additional questions as a result of this study. Here are just a few. 1) Was this an evaluation of the PAC or just another evaluation of goal-directed therapy? Would more aggressive intra-operative achievement of physiologic goals have made a difference in postoperative outcome? 2) Was everything done to ensure adequate data collection and interpretation of hemodynamic data? 3) Is the ASA Classification System an appropriate and accurate way to stratify surgical patients as "high-risk"? Why did 25% of the Class III patients not even receive a central line? 4) Why did more patients in the standard therapy group than the PAC group develop renal insufficiency? What are the cost implications of this? 5) Almost 1100 patients who met study criteria refused to participate in the study. Physicians did not refer 365 eligible patients. Did these exclusions create a bias in the sample?
Other studies have demonstrated the difficulty with using goal-directed therapy.3 The goal-directed therapies used in this study were directed at oxygen delivery indices only. There were no resuscitation end points, such as mixed venous oxygen saturation, oxygen consumption, oxygen extraction, lactate, base deficit, pH, etc. Was it the PAC or the goal-directed therapy that failed to improve patient outcome in the PAC group? It is difficult to separate the therapy from the tool used to direct the therapy.
The investigators did not indicate that any of the recommendations from the Pulmonary Artery Consensus Statement were incorporated into the data collection process of this study4 (however, this study was long underway when the recommendations were published in 1997). The reliability and validity of hemodynamic data obtained by the bedside nurse and interpreted by the critical care practitioner were not addressed. What were the training, credentialing, and continuing quality improvement activities that took place in these institutions? It has been recommended that "no study evaluating the use of the PAC can be performed without first controlling for the accuracy of data and skills of the bedside nurse."5 The usefulness of the PAC cannot be demonstrated without addressing the physiologic and technical variables that influence data collection and interpretation. The "Pulmonary Artery Catheter Continuing Education Program" was specifically developed for multidisciplinary use in addressing these issues (www.pacep.org). Future studies must use this program as credentialing criteria for data collection and interpretation processes. It is imperative that all clinicians who use the PAC receive this training.
Why did more patients in the standard therapy group develop renal insufficiency? There were 95 patients (9.8%) in the standard care group and 70 patients (7.4%) in the PAC group who developed this complication. Although this did not achieve statistical significance (P < 0.07), it is clinically significant. Renal insufficiency was defined as a "50% increase in creatinine concentration or the need for dialysis in a patient with pre-existing non-dialysis dependent renal failure." Twenty-five more patients in the standard group developed renal insufficiency. If these 25 patients required continuous renal replacement therapy, the cost of treating this complication would be significantly higher in the standard care group than in the PAC group.
This study represents significant progress toward identifying the role of PACs in the diagnosis and management of critically ill patients. More studies like this are needed to help determine the risks and benefits of the PAC and identify appropriate use of the PAC in specific clinical situations. The quest to identify the clinical value of data obtained from PACs continues.
Dr. Johnson is Assistant Professor University of Maryland School of Nursing, Baltimore.
References
1. Connors AF, et al. JAMA. 1996;276:889-897.
2. Parsons PE. N Engl J Med. 2003;348:66-68.
3. Boyd O, et al. JAMA. 1993;270:2699-2707.
4. Society of Critical Care Medicine. Pulmonary Artery Catheter Consensus Conference: Consensus Statement. New Horiz. 1997;5:175-192.
5. Ahrens TS. New Horiz. 1997;5:281-285.
Pulmonary artery catheters (PACS) are widely used in critically ill patients. Proponents of the catheter, introduced into the clinical arena more than 30 years ago, argue that physiologic data provided by the use of the PAC permit clinicians to target treatment and improve patient outcomes.You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content