Why do we do what we do? Because people like Zvi Frankel have to work so hard, just to find safe, effective medical treatments. His account of his quest — to help his grandfather find the right heart-valve operation — first was published in JAMA Internal Medicine.  

It reminds us of the value in making data about medical outcomes clear and available.

Here is his unabridged report.

 

Surgical Aortic Valve Replacement versus Transcatheter Aortic Valve Replacement:

A Consumer’s Perspective Regarding Data Education and Transparency of Hospitals

By: Naftali Zvi Frankel

DIAGNOSIS OF SEVERE AORTIC STENOSIS

When my grandfather was diagnosed with severe aortic valve stenosis (AS), meaning his aortic valve would need to be replaced, I decided to thoroughly research the current treatment options available. As my grandfather was relying on my advice and recommendations, I made a conscious decision to leave no stone unturned in what, retrospectively, would become a tedious and arduous project distilling medical fact from fiction.  My primary goal was to research and determine the safest available procedure, hospital, and surgeon.

Some of the earliest Surgical Aortic Valve Replacement (SAVR) procedures were performed by Dr. William Muller Jr. in the 1960s[1]. With the passage of time and advancements in medicine, the surgery has become relatively safe.  In an article written by Dr. Lars Svennson, a respected surgeon at the Cleveland Clinic, he noted that the mortality rate for SAVR in the 1960s was 25 percent to 50 percent, but currently in his hospital, the survival rate is 99.2 percent.  Dr. Svennson noted that even in elderly patients the mortality rate is 1.4 percent for patients over age 80[2].

However, there are patients who develop severe aortic stenosis, necessitating Aortic Valve Replacement (AVR), yet are so ill with serious comorbidities and/or frailty that they are deemed “inoperable.”  In truth, there is no such thing as being “inoperable”; it is a term that has been used to describe patients who have a very high risk of not surviving the rigors of heart surgery.

Once the symptoms indicative of severe aortic stenosis begin, the average reported survival rate for a patient who doesn’t undergo AVR is as low as 50 percent at two years, and 20 percent at five years[3].

Transcatheter Aortic Valve Replacement (TAVR) is a relatively new procedure, first performed by the pioneering Dr. Alain Cribier on April 16^th 2002[4]. TAVR is a life-saving alternative procedure for those who need AVR, but are considered “inoperable” for SAVR.

In recent years and months, there has been a continuous stream of compelling media stories regarding bright and promising success stories of patients considered “inoperable,” “serious risk,” and now even “intermediate risk,” undergoing TAVR.

The hospital valve team informed my grandfather that since they evaluated him to be in an “intermediate/serious risk” category, he would be eligible to join a 50/50 randomized trial. As part of the trial, he would not know whether he underwent SAVR or TAVR until after waking from the procedure. Both procedures were presented by the hospital as equally acceptable options.  They noted the tremendous advantage of not having to undergo the rigors of open-heart surgery if he would receive TAVR.  However, if my grandfather preferred, he was advised that he was free to decline joining the TAVR trial and simply set a surgery date to undergo traditional SAVR.

RESEARCHING THE FACTS

My investigative efforts in determining the safer procedure began with the simple and practical step of searching for information on Google. The next chapter in my trek was reading and analyzing the latest medical journal articles available on the subject matter.  In addition, I became versed with the complex world of medical outcomes statistics and hospital rankings. Finally, the culmination of my efforts was to compare and contrast the data I discovered, and, utilizing the remarkable power of email, send queries in an attempt to solidify my conclusions with the opinions and recommendations of respected surgeons and cardiologists around the world.

Upon searching Google, one of the more prominent results which I immediately came across was the website of a world-renowned hospital, which provided professionally filmed information videos regarding TAVR.

I clicked on the TAVR video link[5], and the presentation began:

“…The results of the [TAVR] trial are for us, as heart surgeons and cardiologists…the same as landing a man on the moon…”

“…One of the groups who have benefited the most are the older weaker members of our community, because when they come in for open-heart surgery they are already frail. They don’t have the ability to weather the storm as younger patients might. So when we could do less-invasive or non-invasive procedures to fix heart valves we could dramatically change the chance for their survival….”

In another video posted on the same website, any viewer would be bound to be moved by the convincing testimonial of an actual patient who underwent TAVR[6]:

“…No stitches, no pain, and in three days I was home. Unbelievable…”

Perusing another result on the search engine, an informative website published by a hospital presents the following description of TAVR[7]:

“…“I had no pain” is often what we hear from our TAVR patients.  While open heart surgery can be very painful with a long recovery time, TAVR patients experience very little discomfort as a result of the procedure.  In fact, our patients are back to doing the things they love, like going to church, shopping, even a little fishing within a week of the procedure.  By having this new valve in place, their heart is able to its job better and patients immediately feel an improvement in energy levels and breathing…”

I ultimately learned that TAVR is a potentially phenomenal life-saving procedure, which also carries potentially severe or even deadly risks, some of which are never even mentioned or spoken about to the potential TAVR candidate.

The process in deciding who is considered a low, intermediate, or high risk candidate for surgery, which determines whether the patient is a candidate for TAVR, relies heavily on a risk scoring system.  There are two scoring systems, the European System for Cardiac Operative Risk Evaluation (“EuroSCORE”), and the Society of Thoracic Surgeons (“STS”) risk of mortality scoring system.

Speaking at the CardioVascular Research Foundation TCTAP 2013 Summit, Dr. Alain Cribier pointed out the fallacy of determining whether a patient is a candidate for TAVR based on a risk score index[8]. Dr. Cribier explained, “Today, I don’t know what exactly is an intermediate risk patient. For example, you take a patient who has a EuroSCORE of 20 percent, he is a high risk patient. If you have a EuroSCORE of 19 percent, he is an intermediate risk patient. This is for me totally ridiculous…”

It was clear to me that the decision whether my grandfather should undergo SAVR or TAVR would not be based on a computer generated STS risk score cited by a doctor, nor a utopian sounding video published on the internet.  The only way I would be comfortable in recommending TAVR rather than SAVR, would be if I knew it was truly the safer, or equally safe, alternative. That decision would have to be firmly based on established data and recorded outcomes.

The following were some of the serious increased risks/concerns from TAVR , which I discovered by researching the subject but that I did not observe were publicized or explained to the patient body:

1)      Increased long term mortality due to aortic regurgitation (AR) after TAVR

2)      Increased need for pacemaker implantation after TAVR

3)      Unstudied potential damage caused by cerebral microemboli after TAVR

4)      Unknown durability of valve after TAVR

I observed that although commercials, websites, and patient education pamphlets were excellent in conveying the potential benefits of TAVR, they simply omitted some of the very serious risks and concerns involved.  However, as Aldous Huxley wrote, “Facts do not cease to exist because they are ignored.”

AORTIC REGURGITATION

Despite accompanying my grandfather to the doctor for a pre-surgical evaluation as well as reading some informative websites, the first time I came across the medical condition called “aortic regurgitation” (“AR”) was when I began reading the medical journal articles.  Slowly, I began to realize that not only is AR a relevant factor in deciding whether to have SAVR or TAVR, it may be the most important factor.

AR is a complication that occurs fairly regularly after TAVR.  On the other hand, AR is extremely rare following traditional SAVR. The explanation why AR is relatively common after TAVR as compared to after SAVR, rests primarily in the fundamental differences between the two kinds of surgery.

During SAVR, there is a chest incision that provides the surgeon the ability to remove the old non-functioning valve and to debride any potentially dangerous calcification from the valve annulus.  In addition, the surgeon physically measures the valve annulus during the surgery, which enables the new valve to be cleanly sutured in place with an absolute measured sealed fit.

On the other hand, during TAVR there is no chest incision, so the attempt in sizing the new valve relies on the medical team’s ability to accurately read pre-surgical images of the patient, such as a CT scan of the heart. Further, during TAVR, the old valve is crushed and remains beneath the newly deployed transcatheter valve, and any annular calcification remains in place. The possibility of misestimating the size of the valve, or failing to properly deploy the transcatheter valve, and the fact that the calcification deposits remain in place, result in an increased likelihood of not having a proper valve seating and seal. The lack of proper valve seating and seal causes AR to occur.

I reviewed an article  based on a collection of TAVR research studies completed between 2002 and 2012[9]. The intent of the study was to “establish the incidence, impact, and predictors of post-transcatheter aortic valve replacement (TAVR) aortic regurgitation (AR).”

The article presented the pooled estimate of moderate or severe aortic regurgitation post-TAVR, which was 11.7%, and the pooled estimate of mild AR post-TAVR, which was 45.9%.

The conclusions of the study revealed that moderate or severe aortic regurgitation is common after TAVR and is an adverse prognostic indicator of short- and long-term survival.  In addition, the study showed that even mild AR may be associated with increased long-term mortality.

In the published PARTNER trial[10], patients who had TAVR with moderate or severe regurgitation were found to have a 51 percent 2-year mortality, and those who had mild aortic regurgitation had a 33 percent 2-year mortality.  In contrast, patients without any regurgitation or a trace leak had a 26 percent 2-year mortality rate.

Commenting in an article regarding the PARTNER trial,  Dr. Robert Bonow of Northwestern University stated[11] “Here you have 40 percent of TAVR-treated patients with mild aortic regurgitation, and another 10 percent with moderate to severe aortic regurgitation, so 50 percent of patients have mild regurgitation or greater—that’s a potential Achilles’ heel.”

In another article, Dr. Peter Block of Emory University School of Medicine noted that “Aortic regurgitation after TAVR is serious business…” “…I agree that everything should be done before the procedure to ensure appropriate valve sizing and then to deploy the valve carefully so that any leak is minimal. But in many cases that simply cannot be achieved…” “…We have to rely on the next-generation valves to solve this problem with better design and better engineering. If they cannot, then TAVR has a serious Achilles’ heel[12]…”

Dr. Vinod Thourani of Emory University expressed the current medical world’s lack of understanding regarding the mortality caused by TAVR aortic regurgitation “This has surprised all of us…Other people walk around with mild aortic regurgitation all day, every day, and they don’t have a 40 percent mortality at two to three years. . . I’m not sure anyone in the room knows the answer to that, and that’s something we need to work on[13].”

PACEMAKER IMPLANTATION

I came across studies comparing SAVR to TAVR with respect to the post-surgical complication of needing a pacemaker implanted.  One such study compared a group of 411 patients with severe aortic stenosis and no prior pacemakers who underwent TAVR, with another group of 411 elderly patients with severe aortic stenosis who underwent SAVR[14]. The study found that 7.3% of the patients who underwent TAVR needed a new pacemaker, compared to 3.4% of the patients who underwent SAVR.

SILENT CEREBRAL ISCHEMIA

Studies have been performed comparing SAVR to TAVR regarding the number of patients found to have new clinically silent cerebral ischemia as a result of the procedure.  One such study showed that by using cerebral magnetic resonance imaging, new clinically silent foci, or lesions, were detected in the brain of 84 percent of the patients who underwent TAVR, as opposed to 48 percent of those who underwent SAVR[15].

In another study, researchers using transcranial Doppler monitored 83 patients who underwent TAVR. It was discovered that 100 percent of the patients were found to have HITS (high intensity transient signals) representing cerebral microembolization, throughout the TAVR procedure[16].

At this time it is still unknown whether there are any long term clinical effects due to silent cerebral ischemia.  However, embolic protection devices are being developed and tested in an attempt to capture emboli released during TAVR before it reaches the brain.

DURABILITY

It has been established that SAVR bioprosthetic valves could last for 15-20 years, generally depending on the age of the patient, while mechanical valves have been found to last indefinitely. On the other hand, the long term durability of TAVR valves is simply unknown.

In an article published in 2012, Dr. John Webb of St. Paul’s Hospital, Vancouver, BC, a senior author of a study reporting on the durability of the TAVR valve commented: “We did not know if these valves would last five years. We now do[17].”

In another article published in 2013, Dr. Christopher Cannon of Brigham and Women’s Hospital, observed that the TAVR valves “aren’t falling apart after three years  . . so that’s very encouraging[18].”

 

DATA REQUESTS/ FOIL

With the recognition that surgery is very much included in the cliché, “practice makes perfect,” I decided to search for data that would establish the top hospitals, and specifically, the top surgeons performing SAVR and TAVR.

I noted that there were studies averaging the mortality risks of SAVR and TAVR based on outcomes gathered from dozens of different hospitals.   However, with the recognition that those studies average together superior and inferior hospitals and surgeons, I decided to focus on the actual outcome statistics data of specific hospitals and individual surgeons.  By doing so, I would be able to acquire a tangible understanding regarding the outcomes of SAVR versus TAVR not from averaged statistics, but based on the actual outcomes numbers of specific hospitals and surgeons who might treat my grandfather.

I discovered that the New York State Department of Health publishes reports with detailed comparison data of cardiac surgery outcomes from individual hospitals and surgeons in the state[19].  Such transparent data is extraordinarily helpful for anyone wishing to make an informed determination as to which are the best hospitals and who are the best surgeons.

I decided to review the hospital ratings published by the Society of Thoracic Surgeons (STS). First of all, I noticed that the STS ratings didn’t include many high volume hospitals, including New York City hospitals and the Mayo Clinic, eliminating the possibility of accurately comparing and contrasting top hospitals. In addition, the STS didn’t disclose any actual outcomes numbers of procedures and surgeons. Thus, the STS data was not of assistance in my efforts to find the best possible hospital and surgeon available for SAVR and TAVR.

I also reviewed the U.S. News and World Report “BestHospitals” rankings on their website[20]. However, before accepting their rankings as fact, I wanted to clarify with them how they formulated their rankings.

I emailed Dr. Murrey Olmsted, the project director of the “Best Hospitals Project,” to ask for his insight regarding his “US News and WorldReportBestHospitals” rankings.  I wrote that I had been reviewing the most recently published (2008-2010) hospital/surgeon outcomes statistics reported by the New York State Health Department for cardiac surgery. I noted that  New YorkState reported overall statistical mortality rate for one hospital  [isolated CABG, or Valve, Valve/CABG] was 3.53 percent compared to the reported corresponding [isolated CABG, or Valve, Valve/CABG] outcome mortality statistic for a second hospital, which was reported as 1.32 percent.

I asked Dr. Olmsted based on the New YorkState reported data, why would the first hospital be consistently rated by US News & World Report as a better hospital for cardiac surgery than the second one?

Dr. Olmsted responded to my email explaining that the New YorkState data is a complete census of the mortality associated with specific surgeries, with each procedure being evaluated separately. Their goal is to provide data to consumers about specific surgeries and surgeons.

On the other hand, Dr. Olmsted explained, “Best Hospitals” evaluates the numbers by specialty area and not by specific procedure. As a result, “Best Hospitals” may include a specific procedure such as Aortic Valve Replacement Surgery within the mix of conditions analyzed, but it wouldn’t have data on the success or failure rates for that particular procedure.

Dr. Olmsted further explained that Best Hospitals rankings are a list of hospitals that they feel perform at the highest level across the array of conditions and procedures listed in their methodology report. He noted that their ranking list is intended to provide some guidance as a starting point for hospitals to consider, but not as a definitive guide for any single person about where they should go for care.

As my only interest was to determine which hospital and surgeon would offer the safest and most successful results for AVR surgery, I realized that I would not be able to rely on the “Best Hospitals” rankings to determine the safest hospital for AVR surgery.

Now that I had the general [isolated CABG, or Valve, Valve/CABG] comparative outcome data for cardiac surgery in New YorkState, I decided to proceed to the next step and determine which hospital and surgeon was the best specifically for isolated SAVR and TAVR.

I contacted Ms. Kimberly Cozzens, the Cardiac Initiatives Research Manager of the New York State Department of Health Cardiac Services Program, to request the isolated data.  Ms. Cozzens responded that the Department of Health only publishes the combined outcomes data, not the isolated AVR data.

When I requested the AVR data for three specific surgeons operating in two specific hospitals, Ms. Cozzens sent me the volume of cases and observed mortality rate for Isolated Aortic Valve Replacement of the requested hospitals and surgeons.  Ms. Cozzens noted that the mortality rate found in the unpublished data was not risk adjusted to account for patient severity of illness or other factors (something that they are careful to do with their published data).

As I mentioned before, the most recently published data from the NYS Department of Health was from 2008 to 2010.  When I requested the more recent 2012 isolated SAVR outcome data of three specific surgeons who had very impressive outcomes in the 2008-2010 report, Ms. Cozzens told me that I would have to request that unpublished data with a Freedom of Information (FOIL) request. In addition, I requested the outcomes data of TAVR from the two hospitals which I had been told had the best reputation in New York for experience in performing TAVR, to which she once again responded that I would have to request the TAVR outcomes data with a FOIL request.

I contacted Mr. James O’Hare, the Records Access Officer in the New York State Department of Health, with a FOIL request for the outcomes records. His office got back to me promptly with the raw 2012 reported outcome statistics for the isolated SAVR surgeries performed by the surgeons I had requested; they also sent me the reported 2011 and 2012 outcome statistics for TAVR in the hospitals I had requested.

The outcomes data that I received included an explanation that the observed mortality rate reported were inclusive of all mortalities reported during or after the surgery until discharge.  On the other hand, the observed stroke rate reported only included strokes which “did not resolve within 24 hours.”  Thus, it would be impossible to determine how many neurological incidences occurred which were resolved within 24 hours.

In the 2012 isolated SAVR outcomes of the three requested surgeons (from two hospitals), I discovered the volume of surgeries performed ranged, depending on the surgeon, from 49 to 103 patients. The observed mortality rate was 0 percent for all three doctors.  The observed stroke rate (post-op strokes) ranged from 0 percent to 3.23 percent, depending on the surgeon.  Interestingly, the same surgeon of the three who was reported to have performed the most isolated SAVR procedures in 2012 (103 patients), was also the only one of the three surgeons who had both a zero observed mortality rate, and a zero observed stroke rate.

On the other hand, in the 2011 TAVR outcomes data from the same two requested hospitals, I discovered that the volume of TAVR surgeries performed was 56 versus 163 patients.  The observed mortality rate was 1.79 percent by the lower volume hospital, and 2.45 percent by the higher volume hospital.  The observed stroke rate ranged from 0 percent by the higher volume hospital to 1.79 percent by the lower volume hospital.  Interestingly, the hospital with the lower observed mortality rate of 1.79 percent, had the higher stroke rate of 1.79 percent.

In the 2012 TAVR outcomes data from the same two requested hospitals, the volume of TAVR surgeries performed was 88 versus 272 patients.  The observed mortality rate was 5.68 percent by the lower volume hospital, and 5.51 percent by the higher volume hospital. The observed stroke rate ranged from 2.21 percent by the higher volume hospital to 2.27 percent by the lower volume hospital.

What was most noticeable was that although the hospitals had another year of experience performing TAVR, their mortality rates increased by approximately 3 to 4 percentage points and the stroke rate of one of the hospital increased by more than 2 percentage points.

Now, being mindful of the serious (and potentially fatal) risks involved with having aortic regurgitation (AR) post-TAVR, I sent another FOIL request asking for a copy of the observed AR rates reported from the same two hospitals.  The Records Office responded that the Department of Health doesn’t collect the AR outcomes data; therefore there weren’t any records responsive to my request.

In an attempt to compare the 2012 outcomes data of the two New York hospitals to other top hospitals currently performing TAVR in the United States, I sent FOIL requests for outcomes data to the local health department records offices in California, WashingtonD.C., and Ohio. None of those locales acknowledged that they had any such reported hospital/surgeon outcome data- for TAVR or SAVR.

CONCLUSION

Honest consumer education and true outcomes transparency would empower patients to make their medical care decisions based on solid facts, rather than blind trust.  As outlined in this article, the decision of undergoing TAVR rather than SAVR is a complex life and death decision. Patients deserve to be equipped with accurate and transparent data, so they can make a properly informed decision.

There is no question that TAVR is a lifesaving procedure for those who are diagnosed with severe AS and are deemed “inoperable.” For “inoperable” patients, TAVR is a wonderful beacon of hope and life for a condition that approximately a decade ago was tantamount to a death sentence.

However, for the majority of patients with AS who have the option of undergoing SAVR, it must be made manifestly clear that TAVR is an evolving procedure that is still riddled with serious risks and uncertainty.

Patients are the victims of a climate where there are exaggerated advertisements and misleading videos.  Instead of producing false utopian imagery by only presenting the benefits of TAVR, hospitals should accurately and honestly convey the risks involved, and the benefits of SAVR.

In addition, there is an obvious need for transparency regarding hospital outcomes statistics.  The revelation of outcomes data of individual surgeons in every hospital performing SAVR and TAVR will relieve patients of the heavy load of uncertainty.

Currently, patients aren’t informed if their local hospital and surgeon have a 1 percent mortality rate, or a 50 percent mortality rate.  The result is that patients must regularly make life and death decisions based not on incontrovertible facts and statistics, but rather on unsubstantiated guesswork and potentially misplaced trust.

TAVR has been performed in the United States for approximately five years without any published outcomes statistics from specific hospitals and surgeons.  Isolated SAVR has been performed in the United States for over 50 years, likewise, without published outcomes statistics per specific hospitals and surgeons.

It is my fervent hope that in the future, percutaneous valves will evolve and be perfected, one day becoming just as safe as percutaneous stents.  In the interim however, patients should be granted the ability to compare TAVR and SAVR, based on the risk adjusted outcomes statistics of the actual hospitals and particular surgeons who may perform their procedure—not based on videos and advertisements.

In addition, transparency of specific mortality and complication statistics will enhance the standards of medical care by stimulating an evolution of competition amongst hospitals and surgeons that is purely based on results and success, and not public relations and media reputation. Naturally, such competition will promote higher standards of medical care.  As President Herbert Hoover said, “Competition is not only the basis of protection to the consumer but it is the incentive to progress.”

I presented my grandfather with the risks and benefits of SAVR versus TAVR, and the surgeon outcomes data. Upon weighing the data, my grandfather chose to undergo SAVR by the surgeon with the best outcomes statistics in New York City. The operation was successful.