Everyone knows it's important to quickly provide percutaneous coronary intervention (PCI) to patients who come to the hospital with a myocardial infarction (MI). Guidelines from the American College of Cardiology and American Heart Association call for PCI within 90 minutes of emergency-department arrival for a patient with ST-segment elevated MI, and since 2006, the D2B (Door-to-Balloon) Alliance has enrolled more than 1,000 U.S. hospitals in a collaboration to try to meet that time frame.
But what has all that effort accomplished? Researchers from the University of Michigan recently tried to answer this question by looking at changes in door-to-balloon times and mortality rates during the time that Michigan hospitals were focusing on getting patients to PCI more quickly. They found, in a study published in the Nov. 8, 2010 Archives of Internal Medicine, that the programs were having an effect—delays dropped significantly—but with no corresponding change in mortality. The apparent lack of benefit came as a surprise, even to the study's authors.
ACP Hospitalist recently spoke to study author Hitinder S. Gurm, MD, a cardiologist and assistant professor of medicine at the University of Michigan, about whether the results mean that common assumptions about the importance of rapid PCI are wrong, and where physicians and hospitals should go from here.
Q: What motivated you to do this study?
A: In Michigan, all hospitals that perform PCI participate in a quality improvement collaborative called the Blue Cross/Blue Shield Cardiovascular Consortium. We share data and processes and work on quality improvement across all the participating centers. A few years ago, the consortium developed a focused plan to improve door-to-balloon times across the state and a lot of people worked very hard towards this goal. We wanted to evaluate both what happened to door-to-balloon time and what would be the resulting change in patients' outcomes.
Q: How did the results compare to your expectations?
A: We were quite pleased overall. Door-to-balloon time went down from a median of 113 minutes in 2003 to about 76 minutes in 2008. That was quite a big improvement, and the number of patients that were getting door-to-balloon in less than 90 minutes: It was less than 30% in 2003 [and] it was a little over 67% in 2008. More recently, we have seen a further decline, albeit of a smaller magnitude. That was the good part of the story. A lot of people had been working hard for this and it was nice to see that they were able to achieve the goal of treating patients more rapidly.
The downside was we didn't see any change in in-hospital mortality. My first instinct was well, maybe the door-to-balloon has changed in healthier patients and the sickest ones are still taking longer and that's why that's happening. So we looked at risk-adjusted outcomes. The sickest patients had longer door-to-balloon time, which makes sense. If a patient's sicker when they hit the door, it takes longer to get them from the ER to the cath lab and get the artery open. [However], we saw a decline [in door-to-balloon time] that was parallel in every group. All groups did better with respect to door-to-balloon time but mortality did not change.
Q: What conclusions can you draw about the value of efforts to reduce door-to-balloon time?
A: There are two things that I carried away from this. One is that maybe we are not looking at the ideal metrics. What we need to focus on is the total ischemic time—the time from the patient's symptom onset to the time the artery gets opened. From a patient's perspective, it doesn't really matter if their myocytes are dying outside the hospital or inside the hospital. We need to do a better job of educating our patients, working with EMS services, and other measures so that patients get transferred to hospitals that do PCI and get transferred there quickly.
The second is that even strategies that appear so obviously beneficial as maternal love or clean air need more rigorous evaluation. Prior to evaluating these data, I honestly believed that reducing door-to-balloon time reduced mortality and yet our data did not provide any suggestion to support that belief. And maybe there are other benefits [to reducing door-to-balloon time]. Maybe long term these patients will do better. Maybe they'll have less heart failure. But I think we still need to evaluate therapies or strategies that we think make good sense, to make sure that they actually translate into patient-level benefits.
Q: Should such evaluations be completed sooner—for example, before the launch of a nationwide quality improvement initiative?
A: The way I look at it is there is very little downside to a short door-to-balloon time. The only downside I see is that some patients are rushed to the cath lab who are not having a myocardial infarction. It would make sense to wait and watch for a few minutes to see what actually is going on and not do unnecessary cardiac caths on these patients.
Q: What about the costs, both financial and opportunity, of this focus on door-to-balloon time?
A: [Door-to-balloon time] has become a publically reported scorecard and hospitals are advertising the fact that their door-to-balloon time is half of what the national mandate is. Given all that passion that surrounds door-to-balloon, we've sort of lost sight of what to me is more important, which is how are our patients doing. There's this whole movement towards public reporting of processes and I think that's a good thing, because we want to make sure patients are getting appropriate treatments, but at the same time we should evaluate whether all the involved expense actually translates into better health outcomes.
There are companies out there that are helping hospitals reduce door-to-balloon time, but before or as we go spending millions of dollars doing this, we need to assess if these reductions in door-to-balloon time translate into actual benefit for patients.
Q: If, as you suggest, the focus should be total ischemia time, what can physicians do to reduce the delay between symptom onset and arrival at the hospital?
A: The challenge is that a significant number of patients who come with ST-segment elevation MI have no prior heart disease. If we only focus on patients who already have had heart disease, we will miss a significant number of those who have not had heart disease prior to that, and the STEMI may be their first presentation. We need to do a better job of community education—making sure that patients know not to drive to the emergency room, [and instead] call EMS if they have symptoms that suggest they might be having an infarction. The challenge is I've had patients who are very educated—sometimes physicians—who when they have cardiac chest pain, their first response is denial. Achieving meaningful reduction in symptom-to-balloon time has been challenging. There have been some randomized trials that have focused on community education and they were generally negative. This definitely requires more research.