September 5th, 2013
Should We Be Looking Beyond Door-to-Balloon Time?
CardioExchange contributor Brahmajee Nallamothu interviews Daniel Menees and Hitinder Gurm, co-investigators of a New England Journal of Medicine study that calls into question the emphasis on door-to-balloon time for acute MI patients undergoing PCI.
A door-to-balloon (D2B) time of 90 minutes or less for STEMI patients undergoing primary PCI is a widely used performance measure for quality initiatives. This study found large improvements in D2B times but no significant overall change in unadjusted in-hospital mortality or risk-adjusted in-hospital mortality. The authors conclude that “additional strategies are needed to reduce in-hospital mortality in this population.”
Nallamothu: Although time to reperfusion has been linked to myocardial salvage in experimental models and mortality in numerous observational studies, you found that improvements in D2B time did not correlate with reductions in mortality. Why? Is it something wrong with the measure as we are applying it, or is it that more rapid treatment is unlikely to be beneficial?
Dr. Menees and Dr. Gurm: That is an excellent point. Time to reperfusion is clearly linked with myocardial salvage in animal models, and it makes intuitive sense that more rapid reperfusion should impact outcome. However, D2B time is not “time to reperfusion” — it only measures the delay to reperfusion once the patient reaches a hospital. Myocardial cell death starts soon after arterial occlusion, and the expected myocardial salvage one can expect from reperfusion drops sharply between 2-3 hours; after that time period, the anticipated benefit with respect to myocardial salvage is of limited magnitude. The total ischemic time has come down, but we are still not under that 2-3-hour threshold where we would see the most benefit. With D2B time, we are fixing the part of the process that is easier to fix but constitutes a smaller fraction of the total ischemic time.
There are other benefits of PCI at that point, such as electrical stability, but that is probably less affected by rapid reperfusion. A patient who has VF in the hospital will get defibrillated, and a 10- or even 30-minute difference in getting that artery open probably does not matter that much.
We also would like to address the issue you raised with how we are applying D2B time as a quality measure. D2B time is an attractive metric largely because it is something easily measured and can be applied across all hospital systems. However, in light our data, it is reasonable to question whether D2B time is the best measure of quality of care. Rather, what we are truly measuring is the quality of health systems and their ability to deliver care. We believe there is a subtlety to that distinction that is very important to recognize. Certainly there is a benefit to more efficient healthcare delivery, but we shouldn’t lose sight of the fact that this isn’t the same as improving patient outcomes. So while D2B time may have value, we’re not sure it’s affecting patient care the way we previously expected it to. That said, we’re not sure it’s time to abandon D2B time quite yet. Perhaps the benefit of shorter D2B times will be seen further downstream, in reduced long-term mortality or a decline in heart failure. These aspects still need to be studied.
Nallamothu: How do you think your findings should affect quality initiatives and use of D2B time as a performance measure?
Dr. Menees and Dr. Gurm: We are of the opinion that quality measures should be held to the same standard as new devices or drugs. We cannot perform randomized controlled trials to assess all quality measures, but studies like ours are needed to make sure that we do not become complacent and lose sight of the real goal – improving patient mortality, morbidity, and quality of life.
As far as D2B time is concerned, we probably need to stop focusing on it so much. A goal that almost everyone is meeting should not be a performance measure. We need to focus on total ischemic time, and this will require some innovation on the part of everyone — the general public, the media, community leaders, researchers, and the medical community. Perhaps communities (e.g., by zip codes, counties, etc.) may need to be publicly reporting their total ischemic times; that may possibly effect meaningful change. As everyone knows, prior randomized trials of media campaigns (e.g., the REACT trial) have been negative, and we need to explore alternate approaches.
It may be time to move away from D2B time, or at least recognize its limitations and consider its use in the context of other measures. We need to start trying to track total ischemic time (i.e., onset of symptoms) better, as well as take a closer look at the false activation rates across different regions and hospital systems. We think this is the real message behind our study — it’s time to look beyond D2B times, and we need start looking at other ways we can favorably impact patient outcomes.
Nallamothu: But beyond performance measures, how will these findings influence your own practice when it comes to evaluating and treating individual STEMI patients? Should interventionalists and hospitals still treat these patients as medical emergencies?
Dr. Menees and Dr. Gurm: STEMI is still a medical emergency. We would hate to lose the efficiency and collaboration that the D2B initiative has created. At the same time, physicians should be willing to take a pause when the clinical scenario does not make sense. We have seen an increase in false activations, and patients who have other diagnoses end up in the cath lab before undergoing a thorough evaluation because everyone is focused on the D2B time. Just last year, you diagnosed a perforated bowel in the cath lab in a patient who happened to have a left bundle branch block.
It won’t affect our practice at all. STEMI remains a medical emergency and should be treated as such. The message of our study shouldn’t be that recognizing and treating a STEMI in a timely and efficient matter is no longer of importance. However, we believe it is okay with taking the time to sort out the complex patient, whose diagnosis may be in question, prior to taking that patient emergently to the cath lab, even if it means sacrificing the 90-minute goal.
Nallamothu: Thanks for reminding me of one of my least favorite cases in the cath lab! You are careful to state that “further” efforts to reduce D2B times may not reduce mortality. Do you think we are moving too rapidly at this point?
Dr. Menees and Dr. Gurm: We remain concerned that one possible interpretation one can take from our study is that 90 minutes is not enough, that we need to have a D2B of 60 or 45 minutes. Faster reperfusion is probably always a good idea, but we doubt shaving more time off the D2B time will make much of an impact. We think the focus has to be on the pre-hospital delay.
No one should say after this study that it’s okay to treat a STEMI in 120 minutes. The natural question, though, is should we go even faster? There will clearly never be a randomized controlled trial to assess this issue, which is why we think studies like this are so important. We are worried that we have become too consumed with the “race against the clock,” and findings from our study should give us all pause when thinking about pushing even harder. There definitely comes a point where the pressure to move even faster comes with a consequence, which is namely misdiagnosis and/or medical error. If we are not effecting positive change, then these consequences are amplified even more.
Nallamothu: What are some future directions for measuring time to reperfusion that we should be considering?
Dr. Menees and Dr. Gurm: We need to embrace the concept of total ischemic time, which for most patients is the time from symptom onset to hospital presentation, and believe it should be a quality measure. This is an area on which we need to focus more. Trials of facilitated PCI have largely been disappointing, but perhaps we need to begin exploring reperfusion at the point of health provider contact (pre-hospital) if we are to truly to affect total ischemic times. Hopefully, once we start measuring total ischemic times, we will figure out how to reduce it.
Do you think that less emphasis should be placed on D2B time? Why or why not?