Tuesday, October 18, 2005

The saga of the pulmonary artery catheter

Medpundit posted about this the other day. Here’s my perspective.

Bedside pulmonary artery catheterization, more commonly known as Swan-Ganz catheterization, began life around 1970. As an internal medicine house officer in the late 70s I watched it come into vogue long before the rigors of evidence based medicine. Offered initially as a tool for management of myocardial infarction it soon enjoyed a broader range of use in critically ill medical patients. In the 1980s papers by Shoemaker and colleagues [1] [2] suggested that aggressive protocol driven therapy guided by the pulmonary artery catheter (PAC) improved outcomes in high risk surgical patients.

In the 1990s several papers addressed PAC guided therapy to maximize oxygen delivery in septic patients, with negative results [3] [4]. In 1996 the SUPPORT investigators published a prospective cohort study of outcomes associated with the use of the PAC in a variety of critically ill patients. This oft-quoted and controversial study suggested harmful effects of the PAC and was followed by editorial commentary suggesting that it was time to pull the catheter. Many writers called for prospective randomized trials of PAC and the American College of Cardiology and other professional societies published guidelines for use of the catheter based on the best evidence at the time.

Subsequently, higher level studies have failed to show benefits of the PAC. Finally, in the October 5 issue of JAMA are two studies and an editorial which suggest we may be approaching the final chapter of the PAC saga. The ESCAPE trial, looking at severely ill patients with heart failure, was a bust. In the same JAMA issue this meta-analysis showed no benefit of the PAC across a spectrum of critical illness. An accompanying editorial is linked here.

It’s been a long and interesting ride. If the ongoing FACTT study fails to show benefits of the PAC in patients with ARDS it may indeed be time to pull the catheter.

No comments: