Therapeutic Hypothermia at a Large Academic Medical Center: Lessons for Success

The only Level 1 trauma center in Alabama, the 908-bed University of Alabama at Birmingham (UAB) Hospital initially implemented a therapeutic hypothermia program in the spring of 2009. Dr. Wang and colleagues published a practical and detailed reflection of the UAB the program in the September 2011 The Joint Commission Journal on Quality and Patient Safety.¹ We talk to Dr. Wang about both the program’s success and his thoughts about where efforts are best directed.

What is the current status of the therapeutic hypothermia program at UAB?

We have cooled at least 150 patients in about a two-year time span. The therapeutic hypothermia program has been wildly successful for us, which has been a pleasant surprise. The initial thought was that we would be chilling one patient a month, but it has turned out to be one per week. In December of this past year, we had a day where we were cooling three patients at the same time. For that reason, we have four Arctic Sun cooling systems (Medivance, Inc.). We have had plenty of experience to justify our need for multiple machines.

What were the mistakes your facility made when it first began the program?

As you can imagine, therapeutic hypothermia is an entire program and process — not just a set of orders on a piece of paper. As with any program in the hospital, you have to rally the troops and get everyone excited to achieve success.

When you start any program, it is important to have all the stakeholders and interested parties together at the table. However, our initial efforts included only physician-level discussions, without engaging nurses or other providers. Order sets were completely at odds with standard protocols in the medical center. Many logistical details were not fully thought out. Nurses and physicians received no training in therapeutic hypothermia techniques. Our first efforts went so poorly that nurses filed a patient safety complaint.

I stepped in shortly after that initial experience, using the safety concerns as fuel for quickly reorganizing the program. I quickly found enthusiastic key nursing partners from the emergency department, medical intensive care unit and the coronary care unit (CCU), the nurse educators, who showed great enthusiasm for the technique. Within 4 days, we had our first “super meeting” with 20-25 people at the table. The teamwork and enthusiasm shown by these individuals were incredible. Had I not fortuitously met these people, the program would not have been so successful and — in fact — may have died.

How was training done?

We quickly realized the center has over 500 nurses, physicians and other staff who might need hypothermia training. It became apparent that we could not teach hypothermia in large and boring class sessions. Instead, we went to a model of short, modular training. The ED and ICUs were accustomed to using “huddle sessions” for quick training points. We adopted this approach for therapeutic hypothermia, training small groups of nurses in 10-15 minute sessions. This allowed for training to be done in the middle of the day and it did not require any special scheduling. The only challenge was that it required multiple sessions to get everyone through. We did something similar with the physicians.

How many cardiac arrest patients go to the cath lab?

About 20% of our therapeutic hypothermia patients end up going to the cath lab. The cath lab at UAB is supremely efficient at cardiac catheterization. They have not had a 90-minute ST-elevation myocardial infarction (STEMI) “miss” in over a year. One of the side effects of having such a supremely efficient system is that it is difficult to insert other procedures or processes such as applying cooling pads or giving cold IV fluids.

Patients are not chilled in the emergency department first, before they go to the cath lab?

Hypothermia and reperfusion are equally important priorities in post-arrest patients, and both processes really have to happen at the same time. Post-cardiac arrest brain injury probably happens from the second that restoration of pulses is achieved. Return of spontaneous circulation (ROSC) triggers a cascade of self-destructive processes, including the production of brain edema and swelling. The patient must be made as cold as possible as soon as possible to limit this damage. Lab animals chilled during or before cardiac arrest respond extremely well and are neurologically functional post arrest.

Practically speaking, at our hospital, we apply the Arctic Sun system, give cold saline, and try to get the patient as cold as possible before mobilizing the patient to the cath lab. During the cath lab process, we use the Arctic Sun machines to continue cooling. We then make sure the CCU team is ready to quickly extract the patient out of the cath lab and to aggressively continue cooling.

Does the ED send anyone to the cath lab with the patient?

Because there is so much going on in the cath lab, the ED does try to send an extra person whose sole job is to connect and turn on the Arctic Sun machine. Continuity of care across different units in a hospital remains challenging, a phenomenon that I am sure exists nationally.

Are there other hypothermia pearls for the cath lab?

Cardiac arrest should really be renamed “whole-body arrest.” There is injury not only to the heart, but also to all vital organs including the brain, lungs and kidneys. Cath lab physicians and staff must have a stronger awareness of the multi-organ injury present in these patients. It is often difficult to manage these other organ injuries during cardiac catheterization. For example, over half of out-of-hospital cardiac arrests have evidence of pulmonary aspiration. The team might consider trying to solve ventilatory and hypoxia issues first before mobilizing to the cath lab. At our institution, we do an early head CT scan on post-arrest patients to ensure the absence of cerebral hemorrhage before going to the cath lab.

You instituted a physician volunteer on-call team for the program that became permanent. What impact did that have?

Although some of us had cooled patients before, we were starting a large program and there was a great deal of insecurity about what would happen. For the first few cases, we asked emergency department staff to call us for each case because we wanted to watch the process and identify problems. We discovered that having a physician at the bedside changed the dynamics of patient care. Nurses enjoyed having an expert physician at the bedside to answer questions and guide the process. After about 6 months, with the medical center’s blessing, we matured our observation system into a formal physician consult team. Invariably there is one of us working in the emergency department, so we usually have someone in the hospital available to see the patient

We see our role as the “glue” that holds the whole process together. Providing therapeutic hypothermia is like cooking a Thanksgiving turkey in tag-team fashion – there can be significant confusion without a single person in charge. By having a single physician response team, we found it improved the communication between our care teams.

UAB began without the Arctic Sun technology or any kind of cooling machines. Would you recommend that facilities start out as you did, with ice?

It is very possible to cool patients manually. We did our first 25-30 cases that way. The original trials that appeared in the New England Journal of Medicine were all done by manual cooling.² However, there are many challenges in cooling a patient manually; for example, temperature undershoot or overshoot. Technology is clearly an advantage in providing therapeutic hypothermia. Our trial efforts with the Arctic Sun were well received by nursing staff who felt that it simplified the process. When thinking about delivering a technique in a timely and safe fashion, this technology is the way to go.

How have staff and physicians been motivated to continue?

One of the challenges of implementing any hospital program is motivating nursing and physician staff. Our best motivator turned out to be reunions with the patients. Some of our first survivors asked to visit the ICU nurses and physicians who had cared for them. These sessions generated significant emotional energy. Many nurses who were frank skeptics of the technique became passionate supporters after meeting survivors.

Are there parallels as to how STEMI door-to-balloon programs were implemented and how therapeutic hypothermia programs can move forward?

As I mentioned before, hypothermia and reperfusion are equally important priorities in post-arrest patients, and both processes must happen at the same time. When we set out to design efficient systems of STEMI care, I don’t think that anyone considered that post-arrest patients would fall into the mix. We should reconceptualize the approach to cardiac arrest patients requiring cardiac catheterization.

Dr. Wang can be contacted at hwang@uabmc.edu.

References

1. Wang HE, Thomas JJ, James D, Barlotta K, et al. Post-cardiac arrest therapeutic hypothermia: overcoming the barrier of workplace culture and other implementation lessons. Jt Comm J Qual Patient Saf 2011 Sep;37(9):425-432.
2. The Hypothermia after Cardiac Arrest Study Group. Mild therapeutic hypothermia to improve the neurologic outcome after cardiac arrest. N Engl J Med 2002; 346:549-556.