New NEMJ Study on Door-to-Balloon and Mortality in STEMI

A day after I posted about the benefits of reducing door-to-balloon times to improve mortality, Door-to-Balloon Time and Mortality the New England Medical Journal released a major new observational study that showed that despite a significant reduction in door to balloon time in recent years, there has not been a corresponding decline in 30 –day in-hospital mortality.

Door-to-balloon time and mortality among patients undergoing primary PCI.

It is a provacative article and I look forward to the medical community discussion about the findings.

Here is my initial reaction:

In-hospital mortality is not the best measure of effectiveness of door-to-balloon time. Very few STEMI patients actually die in the hospital. 4.7% in this study. Most of the ones who die are very sick (patients in cardiogenic shock etc, some post ROSC) when they come in. Most STEMI patients, particularly those who come in shortly after presentation, do quite well and are out of the hospital within 2 or 3 days. Many leave with no loss of heart function at all! Perhaps a better measure of the effectiveness of door-to-balloon time would be heart function and post MI quality of life.

The authors suggest, and I agree, that we need to look less at door-to-balloon time and more at symptom onset-to-balloon time.

Here again is an article I mentioned in my previous post, Association of onset to balloon and door to balloon time with long term clinical outcome in patients with ST elevation acute myocardial infarction having primary percutaneous coronary intervention: observational study, which concluded:

“Short onset to balloon time was associated with better 3 year clinical outcome in patients with STEMI having primary percutaneous coronary intervention, whereas the benefit of short door to balloon time was limited to patients who presented early.”

The NEMJ article suggests that hospitals have gotten their in-house systems down so well that perhaps, they are saving everyone that can be saved and there is no more room for improvement.

An accompanying editorial to yesterday’s New England Medical Journal article, Time to treatment in patients with STEMI suggests that:

“It’s unlikely that reducing in-hospital delays by another few minutes will affect clinical outcomes, given the small portion of total ischemic time those minutes would represent and the success that’s been achieved in the system of in-hospital STEMI care. The primary opportunity for reducing total ischemic time and time to treatment, and for improving outcomes, now lies in the prehospital STEMI system of care.”

The time period for this study was 2005 to 2009, before most EMS systems started transmitting ECGs, using STEMI alerts, and field activations.


  • William Dillon says:

    Great Comments. The D to B metric is great to make sure the PCI centers process is good. It is not the entire story. Most of the gains in the future will be with Pre-hospital care. We have to figure out a way to improve patients activating the system earlier.

  • Jarrod Moody says:

    This was a good article. I have to agree that we often place a lot of focus on D2B times. We also factor in Door to EKG, EKG to Activation, etc. but these are all in-house. I do agree with the statement that “The primary opportunity for reducing total ischemic time and time to treatment, and for improving outcomes, now lies in the prehospital STEMI system of care.” With cardiac hospitals such as ours that are conducting an average of 90 PCIs a quarter with an average of 47 minute door to balloon times (consider that half of those are nights/weekends when cath lab staff is not in house) for patients presenting directly to the ER via POV or EMS, we can only make small adjustments to improve the process. I believe that the key lies in early recognition via EMS, transmission, and activation from the field. Yes, the AHA has put out a voluntary “EMS Recognition Program” that is supposed to hold pre-hospital providers more accountable and provide core measures to track but there has been no system integration for the purposes of data sharing and feedback between the hospitals and the pre-hospital setting. I agree that D2B is important but other factors should be taken into place. The national average for DIDO times at non-PCI centers is 56 minutes. Most transferring hospitals aren’t coming close to this number, and it takes a tremendous amount of effort and continual feedback to make headway in improving their times. Unless that organization decides to become a chest pain accredited facility (which is an entire entity of its own), there is no accountability. We have put all of the focus on the PCI centers when we are really just a small part of a very large process. The EMS agencies and the non-PCI centers in our area are doing a great job and want to improve the quality of care for our patients but the measures need to be looked at and adjustments need to be made for everyone.

    STEMI/EMS Outreach Coordinator

  • Brooks Walsh says:

    Yeah, in the majority of STEMI patients the mortality isn’t going to be too different ± a few minutes. For that matter, the mortality would likely be about the same with fibrinolysis vs primary PCI. (Reference somewhere on my computer…)

    That being said, the authors of the NEJM also looked at the subset of patients presenting with cardiogenic shock, where the mortality was about 27%. They found the same flat mortality curve, despite the similar decrease to DTB. I have the graph up at my Twitter account (

    This is a little harder to explain, given the central role that PCI has played in STEMI complicated by cardiogenic shock.

  • Jon Kavanagh says:

    Resurrecting the thread…

    I also add to the mix.

    Stroke has a “go/no go” window of opportunity. Should STEMI? We develop systems of care that include prehospital cath lab activation, but does it matter? Do those few extra minutes matter for the patient when the interventionalist teams are as fast as they are? Should we look at a 5 hr window, that if the patient is inside it, minutes matter, but if he’s outside of it, slow things down a smidge?

  • Jarrod Moody says:


    That is a very good question, and yes, we should be doing that. I don’t have the guidelines in front of me, but I believe that the American College of Cardiology recommends PCI for STEMI patients within 4 to 6 hours of onset of chest pain. I believe that the method of treatment decision depends on several different variables. For example, we have a transferring, non-PCI, rural hospital 65 miles away from our PCI Center that transports patients to us regularly. They have issues in obtaining ALS service (most services in that area are providing BLS only), issues staffing their hospital due to the location, and when patients are brought to us via ground, there really is no easy way to get to our campus due to the geographic layout. After taking that into account along with current transport times, I use the American College of Cardiology’s Best Practices to develop thrombolytic protocols for that specific facility and seek approval from our cardiologists. If we see progression upon re-evaluation, then we can make adjustments as needed. Realistically, how many patients are calling EMS within minutes of onset of chest pain? Some wait for hours. Others wait for days.

Leave a Reply

Your email address will not be published. Required fields are marked *