Introduction
Hello, this is Ileana Piña, Professor of Medicine, Epidemiology, and Biostatistics at Case Western Reserve University in Cleveland, Ohio.
STICH: The Background
As many of you know, the American College of Cardiology (ACC) will be meeting in New Orleans very soon. As usual, we look forward with great expectations to the late-breaking clinical trials. I want to call your attention to a very important clinical trial that has been presented in multiple parts.[1]
Perhaps this is, , the most challenging part of this trial and even the most difficult. The Surgical Treatment for Ischemic Heart Failure (STICH) trial was a very large trial that encompassed 127 clinical sites in 26 countries. It has a very complex clinical trial design meant to answer the question that if we offered [coronary artery] bypass surgery to our patients who had low ejection fractions (heart failure patients who we know have ischemic heart disease), could we ultimately change the outcome?
Two hypotheses have been studied in this trial. One of the hypotheses (hypothesis number 2) has already been presented. It evaluated surgical ventricular reconstruction (removing the part of the myocardium that is fibrosed, scarred, and nonfunctional, and closing the myocardium back) in addition to bypass surgery.
The trial was powered to detect differences in endpoints. This hypothesis was presented and has been published. It included more than 900 patients with a mean age in the early 60s, of which 15% were women, mostly white, and approximately 30% had diabetes. Creatinine levels were elevated in only 8%-9% of the patients.
The patients needed to have good medical therapy in their background, and they were examined to determine whether mitral regurgitation was present. They were catalogued according to mitral valve regurgitation parameters, whether it was mild, moderate, or severe. Stenosis was also categorized as single vessel, 2-vessel, or 3-vessel disease.
The background medical therapy was excellent with more than 85% of the patient on beta-blockers, and more than 87% of the patients on either an angiotensin-converting enzyme (ACE) inhibitor or an ACE blocker. Surgery was performed by the surgeon's usual technique, with removal of this ventricular tissue.
In spite of the fact that the ventricles in the patients who had resection of the ventricle were smaller, there was absolutely not difference in the 30-day mortality. The curves that tracked the event rates of death or cardiac hospitalization did not show any significant benefit of ventricular reconstruction in addition to bypass surgery.
The Most Challenging STICH Hypothesis: Coming to ACC 2011
The strata that will be presented at ACC are from hypothesis 1. Hypothesis 1 is perhaps the most challenging in this trial, which compares optimal medical therapy with surgery. By optimal medical therapy, I mean aggressive use of medical therapy including statins, aspirin, dietary recommendations, and activity recommendations -- everything that you would expect to be included in evidence-based care.
The group with the intervention will have had the same optimization of medical therapy but will also be recommended for bypass surgery. All of these patients will have low ejection fractions and coronary disease. This is the group of patients who are not eligible for surgical ventricular reconstruction.
The planned enrollment was approximately 1200 patients. The endpoints have been adjudicated; the database has been locked. This trial could, in fact, change practice if the bypass group fails to show any additional evidence of survival or event rate above that of the optimal medical therapy.
PROTECT: BNP-Guided Therapy
An extension of this trial that will be presented at ACC is the Pro-BNP Outpatient Tailored Chronic Heart Failure Trial (PROTECT). PROTECT, about which I have spoken in one of my earlier blogs, was Jim Januzzi's trial from Boston examining the use of B-type natriuretic peptide (BNP)-guided therapy vs optimization of therapy without knowledge of BNP.
As you may recall, the patient number was approximately 150 and evenly divided. The trial was stopped early because the patients who were tracked by BNP and optimization of therapy did better. The event rate was significantly lower, and the most event reduction was, in fact, in heart failure hospitalization, which will be the next topic that I will talk about today.
This trial has also reported echocardiogram data. It is my understanding that Dr. Januzzi will be presenting additional information. Contrary to the findings in the European trials, PROTECT showed the same benefit to patients whether they were below the age of 75 or older than 75.
Why is this important? Because BNP-guided therapy may help clinicians continue to adjust medical therapy appropriately with a target to follow. In this case, it was a pro-BNP target of below 1000 pg/mL. Of interest, the greatest increase in the patients where BNP was known was in spironolactone therapy.
Comparative Effective Research: Let's Test the Delivery System, Not the Device
I want to turn my attention now to a commentary that I recently wrote for a journal concerning the comparative effectiveness research of the 30-day mortality and the 30-day hospitalization rate of patients with heart failure. As many of you know from my blogs, this is an area that is of great interest to me, because let's face it, coming back into the hospital 30 days after discharge because of heart failure is not evidence of quality of life, and we must do better.
The numbers in many areas of the country are actually going up instead of getting better. There is great enthusiasm around the country about the Hospital to Home (H2H) initiative fostered by ACC and the Institute for Healthcare Improvement.
The recommendations of that group are that 3 things be accomplished prior to discharge: (1) patients understand their medications and patients can obtain their medications; (2) they have a follow-up appointment very quickly (within 7-10 days) after discharge,; and (3) they know what symptoms to tell their follow-up clinicians, and they know who to call when they feel these symptoms. Many programs are trying to do this, even though there is no perfect answer and no single ideal solution that will fit everything.
So, why don't we use comparative effectiveness research to test, not necessarily the telemonitor that gets installed in the home, but the system in which the telemonitoring resides? A recent New England Journal of Medicine article about a telemonitoring system failed to show superior event rates with a nurse-driven disease management program. In fact, most disease management programs for heart failure have been fairly successful. So, rather than testing the device, we should be testing the delivery system of the device. Perhaps in a nurse-driven, well-coordinated system, most telemonitors would work.
Our patients really deserve this. Our patients deserve this before payment comes into play, and we are in an era of changing practice patterns. We know that by next year we certainly can't do business the way we have been doing it. Our patients deserve better than this. Isn't it all about the patients?
I thank you for joining me today and look forward to talking to you again when I will report to you the results of the STICH trial. This is Ileana Piña signing off. Good day.
© 2011 WebMD, LLC
Cite this: Ileana L. Piña. Optimal Practices for Heart Failure: What's the Latest? - Medscape - Apr 04, 2011.
Comments