Keith Lurie, M.D.
Co-Director, Cardiac Arrhythmia Center, Central Minnesota Heart Center, St. Cloud, MN
Staff Cardiologist, St. Cloud Hospital
Professor of Internal and Emergency Medicine, University of Minnesota
Chief Medical Officer, Advanced Circulatory Systems


Dr. Lurie – thank you for this opportunity to discuss impedance threshold device technology and the ROC PRIMED study.

I see that you attended Stanford School of Medicine. Where did you complete your undergraduate degree?

Yale University

Did you enjoy medical school?

Very much so. I could not afford to go without working and I worked 20 hours/week in a lab to pay my way. I was exposed to some fantastic teachers and became a researcher in the process.

Why did you choose Cardiology?

I worked in med school in a lab that did heart transplant research. I became interested in the biochemistry and pharmacology of the heart, did lots of heart transplants in animals, and became fascinated by the complexity and beauty of the cardiovascular system. I also like to do procedures so it was a natural.

You did your residency at the University of Pennsylvania and a fellowship at the University of California – San Francisco. Could you briefly explain to my readers the difference between a residency and a fellowship?

During residency you learn how to become a doctor and during a fellowship you narrow down to a specialty or subspecialty area. So during residency I became an internal medicine doctor (instead of, for example, a surgeon). During my fellowship I became a cardiologist and then later a cardiac electrophysiologist.

How did you end up at the University of Minnesota?

I wanted to stay in academic medicine, the U of Minnesota had an excellent program, and my sister and brother-in-law both were academic docs at the U of Minnesota. My wife and I loved San Francisco but it was a difficult place to raise a family and have an academic career. We have been very happy in Minnesota.

You’re also an inventor, the holder of several patents, and the founder of a company. Did you always see yourself as an inventor and an entrepreneur?

In college I never thought about becoming an inventor and I never wanted to become a businessman. I studied architecture and discovered I had some creativity but decided to go to medical school as I also loved biology. In 1990 I reported in JAMA on the use of a toilet plunger by family members who performed CPR on their dad with a household plunger (CPR: the P Stands for Plumbers Helper). A colleague told me I should patent the idea. I told him I knew nothing about patents but if he wanted to enlighten me I was game. Twenty years later I know a lot more about this process.

I started Advanced Circulatory Systems Inc in 1997 against my better judgment as I really did not want to become a businessman. However, while studying the plunger idea we discovered the idea of an impedance threshold device (ITD) and no other medical device company was interested in the ITD concept we discovered at the U of Minnesota. Every time we tested the ITD in pigs during CPR is was pretty amazing. Blood flow to the heart and brain was 4x higher with the ITD and active compression decompression (ACD) CPR (plunger CPR) compared with manual standard CPR and 2x higher with the ITD and standard CPR vs standard CPR alone.

Ultimately I realized I could write lots of papers but not necessarily impact patient care through research. Alternatively I could potentially start a company to get this technology out into clinical use and maybe help make a bigger difference. It has been a long journey, with some very high points and a few low points. One very high point was when the plunger device, (the CardioPump) and the ITD were used together on our 2007 U of Minnesota cardiac electrophysiology fellow after he had an out-of-hospital cardiac arrest while jogging in Minneapolis 1 month after he finished his EP training. He survived, was also treated with therapeutic hypothermia, and practices EP today! That was clearly a high point.

How did you come up with the idea for the ResQPOD (Impedance Threshold Device or ITD)?

I was bothered by the fact that there were disparities in the benefits associated with active compression decompression (ACD) CPR (CardioPump) and I was studying it in an acute model of VF. Patients in need of an implantable defibrillator (ICD) in 1991 had a new option to get a device with a pacer-like lead placed in their heart. Prior to that time when you needed an ICD the chest was opened with a thoracotomy and electrode patches were placed on the heart. With the new transvenous approach, the one we use today where the leads goes through a vein into the heart, the FDA made us induce VF successive times until we found the defibrillation threshold. This means that at least one time the ICD would always fail and then we needed it to recharge and defibrillate the patients with a higher amount of energy. Sometimes we had to do manual CPR and after that happened a couple of times I decided to compare the acute physiology of VF during manual standard CPR versus ACD CPR.

We obtained IRB approval and conducted a little study. With ACD CPR minute ventilation was 16 L/min vs 4L/min with standard CPR. Coronary perfusion, ETCO2, and systemic blood pressures were about 25% higher with the ACD CPR. However, we were also measuring the pressure inside the thorax with an esophageal manometer and there was no difference between standard CPR and ACD CPR. That bothered me. I was talking to a colleagues, Mike Sweeney, an anesthesiologist, about this and told him I did not know why the intrathoracic pressure was not going down each time we pulled the chest upward with the ACD CPR device. He suggested that we pre-oxygenate the next patient we operated on and when I did ACD CPR he would occlude the ET tube with his thumb. The first time he did this we watched the esophageal manometer pressure plummet to -10 mmHg when I pulled up with the ACD CPR device and I knew immediately that we had discovered something. It was a Eureka moment. By preventing air from rushing into the lungs, we created a greater negative vacuum and that sucked more blood back to the heart.

Within 6 months we published an article in Circulation showing that when performing ACD CPR in pigs, transiently preventing air from rushing into the lungs during the decompression phase (chest wall recoil) of CPR with a one way valve would result in a lower intrathoracic pressure. The lower intrathoracic pressure causes more blood to move back into the heart, refilling it better for the next time the chest is compressed. We showed that you could normalize blood flow to the brain with this approach (compared with 20% of normal blood flow with manual CPR) and provide the heart with 70% of its normal blood flow. We later modified the ITD so that if a patient woke up they could breathe through it and we added a timing light to guide the ventilation rate (1 quick breath per flash) and the compression rate (10 compressions between flashes). It was ~ 10 years later when we rediscovered that lower intrathoracic pressures cause lower intracranial pressures immediately, and thus one of the additional reasons brain flow was normal in these pigs getting ACD CPR and the ITD.

How exactly does the device work?

The device is mechanically relatively straightforward but physiologically more complex.

When you compress the chest air leave the lungs through the ITD with minimal or no impedance. When you ventilate a patient during CPR air goes into the patient through the ITD with minimal or no impedance. However, when the chest is recoiling (passively or actively) after being compressed, the ITD blocks air from entering the lungs, thereby creating a greater decompression phase vacuum inside the thorax. This pulls more blood back into the heart and lowers intracranial pressure. In this way the heart is more efficiently refilled after it is emptied during the compression phase and the brain gets more blood by two different mechanisms. If the patient starts to breathe after they are resuscitated, they can breathe through the ITD but it can be difficult.

The ResQPOD that is being sold commercially in the US has -10 cm H2O resistance when a patient inspires through it. The ITD used in the ROC study had a -16 cm H2O resistance. The higher resistance is needed to optimize ACD CPR (which we thought would eventually be used in the US at it is used now in some other countries). However, with standard CPR we never achieve a very large vacuum in the thorax with passive chest wall recoil so -10 cm H2O resistance is adequate. That is, with passive recoil of the chest, the vacuum created with an ITD is typically only -2 to -6 cm H2O.

So, in short, the ITD turbo-charges CPR. It doubles blood flow back to the heart when used with manual standard CPR and it lowers intracranial pressures (ICP) (part of the normal physiology we rediscovered about 8 years ago). This results in more blood flow to the brain as there is less resistance to forward flow.

You have to take the ResQPOD or ITD off of the patient when not doing CPR or it can cause potential harm after the patient has been resuscitated. On the other hand if a patient is gasping but in cardiac arrest then they are trying to generate a negative intrathoracic pressure on their own and you should leave the ResQPOD on the patient as long as CPR is needed and being performed.

Most people do not know that each time we take a breath and inspire spontaneously, or when we create a negative intrathoracic pressure within the chest during the decompression phase of CPR with the ITD, this lowers intrathoracic pressure and the lower pressure in the thorax is immediately transferred to the brain via the rich venous plexus surrounding the spinal cord. So with greater negative intrathoracic pressure, the ICP goes down.

Thus, we can lower ICP by lowering intrathoracic pressure. One way to do this is with the ITD or ResQPOD. Needless to say we did not rediscover this overnight but we believe this is contributes to why more patients wake up after cardiac arrest when they are treated with the ITD. The lower the ICP or resistance to forward blood flow, the more blood will go to the brain. This principle has not been well exploited in medicine. The converse is also true. One of the reasons that pressure bombs cause head injury is that pressure waves are transmitted through the abdomen and thorax via the spinal cord and venous plexus around the cords to the brain and this can cause a sharp and deadly rise in ICP.

The device was given a Class IIa rating in the 2005 AHA ECC guidelines, which is higher than any medication. Were you excited about that?

A double-blinded randomized study by Drs Parrillo and Aufderheide in 2004 made 3 fundamental new observations:

1. Hyperventilation during CPR was common and later shown in pigs to be deadly
2. Incomplete chest wall recoil during CPR because people lean on the chest is common and was shown later in pigs to be very harmful
3. The ITD doubled blood pressure during CPR.

That study along with other studies in animals and people convinced us to launch the product in the US. The study results were exciting. The AHA recommendation of Class IIa based upon increased circulation has helped get this technology out onto the streets and helped to save lives. Some of the largest EMS systems in the country are have implemented all of the Class I and IIa recommendations including the ResQPOD. Fewer people are dying in those cities after cardiac arrest now and that is very gratifying.

I think many of us who have been following the ResQPOD were surprised by the recent announcement by the National Institute of Health that the ROC PRIMED trial was stopping enrollment.

Tom Aufderheide, MD was quoted as saying, “While the ITD is based on sound physiologic principle, in this study it did not appear to improve survival rates for adults in cardiac arrest outside the hospital.”

Considering that the ROC PRIMED trial was a prospective, multi-centered, randomized clinical trial with large enrollment, are you concerned about these results?

I think it is important to begin my response by saying that we do not know all of the results of this trial; all we know is that if they had continued it that the results would not have been conclusive. I also know that if you ask Dr. Tom Aufderheide “would you want the ITD used on your mother if she went into cardiac arrest” he would say something like, “yes, as long as she was also receiving high quality CPR with good chest compressions.”

We know that complete study results will not be disclosed for 6 - 12 months. What we know is from discussions with key personnel involved with the study and from press statements. So to directly answer your question, I am not concerned with the results, nor am I surprised.

I feel it is important to focus on the lessons learned already from the ROC study so that we can improve outcomes uniformly when rescue personnel use the ITD. As the founder of the company that makes the ITD, I continue to believe in the fundamental physiology underlying the ITD, its potential to increase circulation in patients in cardiac arrest, and in the potential for the ITD to increase neurologically-intact survival rates, especially when the device is used in emergency medical services (EMS) systems that are already performing well.

Let me explain a bit about the study results. There were four key findings that we know so far:

1. There were no safety issues with the ITD.
2. Some sites saw a benefit with the active ITD and others did not; on average no increase in neurologically-intact survival to hospital discharge benefit was observed.
3. Overall survival rates increased during the study compared to baseline.
4. The study arms that concurrently evaluated 30 seconds of CPR versus 3 minutes of CPR followed by analyze and shock found that both groups had similar outcomes.

I also know from confidential discussions that the cities that had the highest survival rates before the ROC PRIMED study had an increase in survival rate with the active ITD compared with the sham during the this study.

Do you expect that the ResQPOD (Impedance Threshold Device or ITD) will be downgraded to a Class IIb rating in 2010?

While I am not involved in that process, and was not previously, the AHA committees that make that determination will look at all of the data available this year. The ROC ITD study was neutral and I believe that the ITD will continue to have a Class 2a recommendation. One study with over 1000 patients published by Aufderheide et al in 2009 showed a significant increase in survival rates compared with historical controls in 6 large EMS systems (Crit Care Med 2008; 36[Suppl.]:S397–S404). There was a >50% increase in survival for patients presenting with ventricular fibrillation. Another recent study comparing in-hospital ITD use with historical controls showed a 65% increase in neurologically-intact survival to hospital discharge for patients treated with the ITD: the historical controls had a 17% survival to discharge versus 28% in the ITD-treated group. (Thigpen et al, J. of Respiratory Care, in press, 2010). So far all of the studies have been positive and the ROC study overall was neutral. I believe there will be some subgroups even in the ROC study will show a strong trend towards a benefit.

In the study design, patients were randomized to one of two devices: the “active” ITD or the “sham” ITD. Did both devices look identical to the rescuers and were both equipped with a timing light for ventilations?

Yes, the devices were opaque and yellow and could not be distinguished when looking at them or feeling them. Both had timing lights that flashed 10x/min.

Do you think the simple fact that patients were not hyperventilated was an important factor in the relatively high survival rates seen in the study?

I am not sure I would agree with your conclusion that there was a relatively high survival rate. Before the study the survival rates to hospital discharge with good neurological function ranged from a low of 1.1% in Birmingham AL to 8.1% in Seattle. The mean neurologically-intact survival rate was 5% in all of the sites during the study: that is close to the national average and not very high at all. I am certain that performing CPR at the correct ventilation rate helped to some degree, but it is hard to know how much. I think both groups were likely to have had a very similar ventilation rate.

Editor's note: This question was based in part on Lisa' Bell's blog post Unexpected Increase in Survival Rates in ROC Study at JEMS Connect.

The study also randomized patients to an “early shock” or a “late shock”. Could you please explain the difference in these two approaches?

Half the patients were treated with 30 seconds of CPR followed by analysis and shock and the other half had 3 minutes of CPR followed by analysis and shock. If the patients still required CPR then it was continued with either an active or sham ITD. In some cases the ITDs were added in the 3 minutes of CPR, analyze, shock group, but usually the ITD was added after these 2 different CPR/analyze/shock interventions. This makes analysis of the ITD effect more challenging to say the least. We know that in many cases it takes longer to get an AED onto a patient so the 30 seconds was more often closer to 60 seconds. We know that the study did not show a difference in outcomes when comparing 30 seconds vs 3 minutes of CPR.

I noticed that even the “early shock” patients received some chest compressions prior to the first defibrillation attempt. Does this study demonstrate that boosting the heart’s perfusion pressure prior to a defibrillation attempt is a critical intervention?

We do not have a control group in this study to draw that conclusion. I can say that one site, Seattle, believed so strongly in CPR before shock for any patient in cardiac arrest >4 minutes before being able to deliver a shock that they refused to participate in the 30 sec vs 3 min of CPR analyze and shock portion of the study. This is also the site with the highest survival rate before starting the ROC PRIMED study.

Is there anything else you’d like to say to about the ROC PRIMED trial, cardiac arrest, or any of your inventions?

I do have more I would like to say and I apologize as it is quite lengthy. But let me premise my comments by saying there is no silver bullet in CPR and we have been foolish for decades to think there is or was. Survival from cardiac arrest requires multiple therapies, just like any other complex disease whether it is heart failure, leukemia, or AIDS. What we strongly recommend is a systems-based approach, where all of the AHA Class I and IIa recommendations are deployed sequentially. We have described this in a program called Take Heart America. With this approach we have doubled survival rates in multiple cities.

My additional comments about the ROC PRIMED study are below.

There were many reasons for the overall neutral results, recognizing that full details will not be available for months. They include:

1. Multiple study protocols were executed simultaneously:
1. 30 secs vs 3 mins of CPR analyze shock during BLS for all patients
2. Three fundamentally different BLS protocols, which included a) following the 2005 American Heart Association (AHA) Guidelines 30:2 protocol, b) continuous chest compressions without interruption for ventilation with a breath every 10 compressions with the 30 seconds vs 3 minutes of CPR, analyze, shock protocol, and c) continuous chest compressions without interruption for ventilation with a breath every 10 compressions without the 30 seconds vs 3 minutes of CPR, analyze, shock protocol.
3. CPR devices to assess depth of compression and compression rate were tested in some sites
4. Prehospital cooling was used in some sites
5. Post-resuscitation care varied greatly
2. Quality of CPR – variable from site to site, reflected in part by the baseline differences in neurologically-intact survival from 1.1% to 8.1%.
3. ITD put on late due to the analyze early vs analyze late protocol; on average the ITD was placed ~4 minutes after the start of CPR, often much later.
4. Potential interactions between analyze early vs analyze late protocol, as CPR was started and stopped prior to ITD placement are unknown.
5. The ITD was not always removed post ROSC, which can be harmful with the -16 cmH₂O cracking pressure in the ROC ITD.
6. There were very poor survival rates, as low as 1.1% at baseline in some sites, so an effect can be hard to detect even with tens of thousands of patients.
7. Overall survival rates increased during the study compared with historical, likely due, in part, to a Hawthorne effect that included increased focus on CPR quality. This further limits the statistical ability to show a difference between the active and sham group
8. Chest compression depth varied from patient to patient and some sites did not emphasize the basics of CPR – that it compress the chest ASAP 1.5-2 inches with minimal compressions. Without good compressions the ITD, which functions like a turbocharger during CPR, is ineffective. There was no mandatory training program or retraining program so that too varied from site to site.

So what are the lessons we learned?

1. When using the ResQPOD you need to continue to perform high quality CPR; it is fundamental to success. Emphasis on correct compression depth and rate needs to be re-emphasized while at the same time the importance of full chest wall recoil is stressed. In addition, rescuers need to follow the directions for use related to the timing of ResQPOD placement and removal. Training and regular retraining with hands-on, I believe every 3-6 months, is important.
2. Best results are achieved with a systems-based approach which includes the practice of all of the key AHA 2005 Guidelines. The key guideline recommendations and their class level of recommendation are below.
3. Early application of the ResQPOD is essential. It should be placed as soon as possible, but after chest compressions are initiated. Chest compressions should always be started right away. Focusing on quality of CPR (it is hard to do and regular retraining is essential) and tracking outcomes are essential to deriving value from CPR in general and from the ResQPOD.

I believe we are on the threshold for a new era in CPR. The kinds of studies that ROC has performed are going to have to change so that systems-based approaches are used otherwise ROC will continue to have neutral or negative studies. Co-variables related to care outside and inside the hospital are too critical to go uncontrolled in the future. These uncontrolled variables impact the integrity of care and study outcomes.

Once we deploy what we know works correctly in systems that function well initially (systems with a 1% survival should never be ROC study sites), we will see even more progress. Ultimately we will also want to regularly deploy automated CPR devices and new techniques that optimize circulation during CPR, (including a device that generates a continuous negative intrathoracic pressure between each intermittent positive pressure breath), perhaps cooling during CPR for patients still in cardiac arrest, and post-resuscitation care that includes cooling ASAP.

I noticed during my research for this interview that NASA inducted the ResQPOD into the Space Foundation Technology Hall of Fame. How did that come about and why does NASA carry the ResQPOD on the Space Shuttle?

In 1998 we started working with NASA and the US Army using the ITD in spontaneously breathing volunteers. Breathing through an ITD with a resistance of only -7 cm H2O, through a device called the ResQGard, lowers intrathoracic pressures, enhances blood flow back to the heart, and lowers ICP. We tested this in many human volunteers, including hypotensive volunteers. This other version of the ITD called the ResQGard is part of the standard care NASA astronauts get if they are hypotensive after prolonged space flight. Upon return to earth it takes a while for their bodies to get used to gravity and they often have severe orthostatic hypotension. The ResQGard helps maintain higher blood pressure and cerebral perfusion. The ResQGard is now being used by EMS systems to treat spontaneously breathing hypotensive patients and it is being used by military personnel to treat hypotensive soldiers. Research on the ITD has been funded by the NIH, the Defense Department, and by NASA. The ITD was inducted into the Hall of Fame as a result of this research. I am not aware that the ResQPOD is being used on the Space Shuttle although it should work in space.

Thank you for this opportunity to share my thoughts with you and your readers.

Helpful links:

ResQPOD Product Animation

ResQPOD demonstration video - pig in cardiac arrest

Rebroadcast of 11/12/09 webinar: Improving Resuscitation with a Systems Based Approach: An Update on Six Clinical Investigations by Keith Lurie MD and Charles Dick MD

Cardiac Arrest Trial Cut Short

NHLBI Stops Enrollment in Study on Resuscitation Methods for Cardiac Arrest

Resuscitation Outcomes Consortium (ROC) PRIMED cardiac arrest trial methods part 1: rationale and methodology for the impedance threshold device (ITD) protocol.

Resuscitation Outcomes Consortium (ROC) PRIMED Cardiac Arrest Trial Methods Part 2: Rationale and Methodology for "Analyze Later" Protocol

Study Compares Passive Oxygen Insufflation and BVM