Optimized AED Placement May Improve Cardiac Arrest Outcomes

Debra L Beck

September 16, 2019

Determining the optimal location for automated external defibrillator (AEDs) may lead to increased defibrillation by bystanders and, possibly, increased survival among those experiencing an out-of-hospital cardiac arrest (OHCA), a new study suggests.

Predicted outcomes using an in silico — or computer simulated — model that mimicked a randomized clinical trial and simulated optimal placement of new AEDs improved relative coverage of OHCA by 52.0% to 95.6%, compared with historic cardiac arrest data.

"There's a lot of AED placement that's currently being done on an ad hoc basis, maybe by a business owner deciding to put an AED in their office, or by foundations that pool some money and purchase some new AEDs," said Timothy C.Y. Chan, PhD, from the University of Toronto.

"But when we looked at where cardiac arrests occurred and used simulations to determine optimal locations for AEDs, we saw that choosing the combinations of those locations for placement of AEDs could maximize coverage of cardiac arrest," Chan said.

Findings from the study were published in the September 24 issue of the Journal of the American College of Cardiology.

For this analysis, the researchers simulated two AED placement policies as interventions. In the first intervention, AED placement was chosen on the basis of location and hours of operation; in the second intervention, AED lacement was chosen on the basis of location and assumed 24/7 accessibility. The control group consisted of all real AED placements in Copenhagen during the study period.

The team used OHCAs that occurred from January 1994 to September 2007 to train the models and used OHCAs that occurred the study period of October 2007 to December 2016 for the analysis. All study OHCAs occurred in public locations.

The primary outcome was the total number of OHCAs that occurred within a 100-meter straight-line, equal to a 150-meter pedestrian route, from an available AED.

Although the actual Copenhagen AEDs covered 22% of all OHCAs during the study period, those placed using simulation — either with or without 24/7 availability — would have improved bystander intervention to 33.4% and 43.1% of arrests, respectively. These increases amounted to relative coverage gains of 52.0% and 95.9%, respectively (< .001).

Thirty-day survival rates increased from 31.3% in the control group to 34.7% to 35.4% in the two intervention groups, which translated to relative improvements in survival of 11.0% to 13.3% (P < .001).

Really, what's needed are head-to-head clinical trials that compare optimized and real AED deployment in major cities, Chan said in an interview with theheart.org | Medscape Cardiology.

"My hope is that people find these results compelling such that now that we've identified what promises to be a promising intervention, people might want to actually go and trial it in the real world," said Chan. Already he has been called upon to optimize placement of AEDs in smaller geographic areas, like the University of Toronto campus.

Chan is the Canada Research Chair in Novel Optimization and Analytics in Health and director of the Centre for Healthcare Engineering at the University of Toronto.

To be clear, Chan isn't suggesting that currently placed AEDs be removed. For example, if the historic data indicate no OHCAs in a sporting arena that already has one or more AEDs, he doesn't suggest that those devices be moved elsewhere.

"A sports arena that 200 days a year is filled with maybe 20,000 people is not a bad place to put an AED, of course, but when we do these analyses, as we've done previously in Toronto, there are the usual places that pop up — like Union Station and other high-traffic spots — but then there are other parts of the city where maybe it's a more depressed area and there is not currently AED coverage, where the data will show that we might want to place devices."

Chan acknowledged that the model does not account for issues like theft or cold temperatures that might render the AED nonfunctional. "I personally think that ATM vestibules are a great place to put these devices because they are generally open 24/7 and accessible not just by the customers of that bank," he said.

Indeed, Chan's group has previously shown that coffee shops and ATM lobbies are among the top spots for AED placement. They've also studied how drones might be used to deliver AEDs to an OHCA event.

"ATM lobbies are usually heated and they also have security cameras that might dissuade would-be thieves, or at least capture somebody taking the AED," he added. Another advantage to placing AEDs in ATM lobbies and coffee shops is that it would help people know where to look for the devices when they're needed.

They Only Work When They're Used

"AEDs can be very effective, but they are only effective when they are available," write Eric C. Stecker, MD, MPH, Oregon Health and Science University, Portland, and colleagues in an accompanying editorial.

"Unfortunately, AED placement has been haphazard in many communities," he added, either distant from where OHCAs occur or nearby but inaccessible or undiscovered at the time of the arrest.

Research has shown that public-access defibrillator use is associated with at least a doubling of neurologically intact survival.

"What I like about this paper is that it emphasizes that AEDs, or an emergency action plan, like we recommend for sporting events, isn't a 'set it and forget it' type of program. They are a work in progress and forever will be a work in progress, and this paper is a [quality assurance] reminder of this," said Mathew Martinez, MD, Lehigh Valley Health Network, Allentown, Pennsylvania.

As a next step, he'd like to see better collection and pooling of OHCA data. "These events are so rare that it's hard to tell if the events are really related to time course and other things or if it's simply chance."

"So, part of doing this type of QA improvement is starting with the data we have in hand, but continuing to collect more and better data to see if the signals change," he said in an interview.

Many organizations, including the American Heart Association and the European Resuscitation Council, offer guidelines for public-access defibrillators, but none are informed by means as sophisticated as this mathematical model, nor have they been tested comprehensively against historic data. This is the first in silico trial to examine AED intervention.

The optimization models in this study were originally developed for use in Toronto, a city with different OHCA incidence, city structure, and population characteristics than Copenhagen. These findings therefore provide further validation for the generalizability of optimization models for AED placement, Chan and colleagues note.

This study was supported by the Danish foundation TrygFonden, which has no commercial interest in the field of cardiac arrest. Chan, Stecker, and Martinez reported that they have no relationships relevant to the contents of this study to disclose.

J Am Coll Cardiol. 2019;74:1557-1567, 1568-1569. Abstract, Editorial


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: