Peri-operative mortality with/after TAVI for aortic stenosis

More on TAVI: A newly published analysis of the existing/ reported data for peri-operative mortality after transcatheter aortic valve implantation looking at 12 previous studies was recently (June) published in the journal of Interventional Cardiology.  (While the study looks at the causes of death – we here at Cartagena Surgery – are going to talk about the rate of death in this study.)

Article information:

Causes of Peri-Operative Mortality After Transcatheter Aortic Valve Implantation: A Pooled Analysis of 12 Studies and 1,223 PatientsThe Journal of Invasive Cardiology 2011;23(5):180-184.

Raul Moreno, MD; Luis Calvo, MD; Pablo Salinas, MD; David Dobarro, MD; Jimenez Valero Santiago, MD; Angel Sanchez-Recalde, MD; Guillermo Galeote, MD; Luis Riera, MD; Isidro Moreno-Gomez, MD; Jose Mesa, MD; Ignacio Plaza, MD; Jose Lopez-Sendon, MD

Abstract re-posted below.

Background. In order to improve technique and to prevent serious procedural complications during transcatheter aortic valve implantation (TAVI), it is crucial to identify the causes of death of patients undergoing this procedure.
Objective. The objective of this study was to identify the causes of death during the procedure and at 1 month in patients with severe aortic stenosis undergoing TAVI.

Methods. 12 published studies with information about the causes of death in patients undergoing TAVI were selected. Overall, 1,223 patients were included in these studies, and 249 deaths were reported (119 at 1 month and 130 at > 1 month post-procedure).

Mortality during the procedure and at 1 month was 2.3% and 9.7%, respectively. The proportion of cardiac deaths was higher at < 1 month in comparison with > 1 month (56% versus 34%, respectively; p = 0.001). At 1 month, the most frequent causes of death were cardiac failure/multi-organ failure (24%), sudden death/cardiac arrest (17%), vascular and bleeding complications (17%), stroke (11%), sepsis (11%), and cardiac tamponade (10%). During the procedure, the most frequent causes of death were cardiac tamponade (39%), cardiac failure (21%), cardiac arrest (18%), and vascular and/or bleeding complications (18%).

In patients treated with the CoreValve system (Medtronic, Minneapolis, Minnesota) versus those treated with Edwards valves (Cribier-Edwards, Edwards-SAPIEN or SAPIEN XT valve, Edward Lifesciences, Irvine, California), deaths at 1 month due to vascular and bleeding complications were less frequent (3% versus 22%, respectively; p = 0.019), but those due to cardiac tamponade (26% versus 6%, respectively; p = 0.019), and because of aortic regurgitation (10% versus 0%, respectively; p = 0.03) were more frequent.

Conclusion. In this pooled analysis, mortality at 1 month after TAVI was 9.7%. The causes of death were widely variable, and of both cardiac and non-cardiac origin. There were some important differences between both devices in the cause of mortality.

How does this compare with conventional aortic valve replacement surgery (AVR)?

In cardiac surgery – surgeons use database calculators.  The most popular one is called the STS risk calculator to determine or estimate the surgical risk for specific patients.  This calculator is based on thousands and thousands of patients over decades of research to give approximate surgical risk of morbidity (post-operative complications) and mortality by looking at the planned procedure as well as patient risk factors (age, poor heart function, co-morbidities).  Other calculators include a European calculator called EUROscore, and a score used  by the VA (veteran’s hospitals.)

Well, how accurate are these calculators?

Interestingly enough – at the same time as the TAVI article, an article (Basreon et. al) discussing and comparing each of these calculators to actual results was published in the June 23 issue of the Annals of Thoracic Surgery.  (I’ve re-posted the abstract below.)

In their research – Basreon et. al. found the overall peri-operative mortality for aortic valve replacement surgery to be 5.9%  which is well under the 9.7% reported in the article by Moreno, et. al (re-posted above) for TAVI.

While the argument can be made that the higher than expected peri-operative mortality in the TAVI group may be secondary to other factors (patient condition at time of TAVI) without more information on patient demographics – it is hard to know.

I, for one, would like to know the average ages of patients in both groups – was the TAVI group made up of non-surgical fragile, 95 year-olds?  What specific factors raised their EUROscores? Was it overall heart function, or was it a specific co-morbidity?

It’s difficult to know since it’s a composite of other research data from multiple studies (and since TAVI is widely used in Europe, accounting for as many as 40% of all aortic procedures in Germany, for example) – this data may also reflect many of these patients (who are not frail elderly, for example.)

Reading through the Moreno study – there is little discussion of the actual patient population, except for one small paragraph (re-posted below). Both of these limitations are probably due to the nature of the study – where investigators were pooling the results of several different studies – which is a good strategy to get a wide overview.  However, this can make it impossible to go back and look at questions like ours, particularly if the investigators on the original, smaller studies didn’t record / report this information.

[my comments in brackets/ italics].  I have placed data within the article in bold or italics. 

“In this study, pooling the results of 12 series, mortality at 1 month in patients treated with TAVI was 9.7%, and mortality during the procedure was 2.3%. These data compare favorably with the predicted surgical mortality, since EuroSCOREs ranged from 12–28%.  [this is the risk calculator that Basreon et. al found that grossly overestimated risk in the study re-posted below.]

In the randomized PARTNER trial, a significant reduction (~20% absolute risk reduction) in the 1-year mortality was obtained for patients with severe aortic stenosis and considered not suitable for surgery due to a very high surgical risk when treated with TAVI in comparison with medical treatment. [as you know from previous discussions – medical treatment is exceedingly ineffective for this condition.  It would be more helpful if authors had better defined their ‘very high’ risk patient group since multiple studies show that cardiologists, etc. overestimate patient’s surgical risk.] 

Importantly, mortality significantly reduced through the years, from 2004 to 2010, probably reflecting not only the learning curve and the technical improvements, but also a better patient selection process. [meaning patients that are ineligible for surgery may also be ineligible for TAVI in some cases.]   This reduction in mortality over time has also been observed in single-center experiences.

Although not statistically significant, mortality at 30 days was higher in patients treated by transapical approach in comparison with transfemoral approach. Probably, the worse clinical profile of patients undergoing transapical TAVI justifies, at least in part, these findings.[7,12]

Ann Thorac Surg. 2011 Jun 23. [Epub ahead of print] Comparison of Risk Scores to Estimate Perioperative Mortality in Aortic Valve Replacement Surgery.


Division of Cardiology, Veterans Administration Medical Center and University of Minnesota, Minneapolis, Minnesota.

Abstract  (bolding/ italics from original article)


Transaortic valve implantation has recently been introduced as an alternative to aortic valve replacement (AVR) for high-risk patients with aortic stenosis. However, accurate assessment of surgical risk is critical for appropriate patient selection. We compared the accuracy of The Society of Thoracic Surgeons (STS) risk score, the European System for Cardiac Risk Evaluation (EuroSCORE), and the Veterans Administration (VA) risk score in predicting perioperative mortality after AVR.


We included 537 consecutive patients who underwent AVR for severe aortic stenosis at the Minneapolis VA Medical Center between 1997 and 2008. Observed and predicted perioperative (30-day) mortality rates were compared. Hosmer-Lemeshow goodness-of-fit test and receiver operating characteristic curves were performed to assess the performance of the scores.


Perioperative mortality rate was 5.9% (n = 32). Predicted mortality rates for the EuroSCORE, STS score, and VA score were 15.6%, 3.6%, and 6.7%, respectively (p = 0.001). The EuroSCORE overestimated mortality in all patients, most notably among those with ejection fraction less than 35% (49% predicted versus 9% observed). The EuroSCORE had poor calibration (goodness-of-fit test p < 0.008), whereas the STS and the VA scores were well calibrated. However, all three scores displayed good discrimination characteristics per the areas under the receiver operating characteristic curves: STS score 0.73 (95% confidence interval: 0.69 to 0.77); VA score 0.66 (95% confidence interval: 0.62 to 0.70); and EuroSCORE 0.68 (95% confidence interval: 0.64 to 0.72; p > 0.05).


The EuroSCORE substantially overestimates perioperative mortality risk in AVR, particularly in patients with low ejection fraction. These data have implications when deciding the appropriate intervention (transaortic valve implantation versus AVR) for high-risk aortic stenosis patients.

In general – the majority of the literature cites peri-operative mortality for AVR at 2.0 – 5.0% (but this is an average of ALL patients, making the calculators our best estimate of predicted risk.)

So what does this mean?

Clearly, when the data from Moreno et. al shows a thirty-day (peri-operative) mortality of almost 1 in 10 patients – it’s a signal we need to proceed with caution, and continue to follow the research.

1.  Since the authors report many of these patients at very high surgical risk (presumably due to cardiac status as well as co-morbidities) and 2. we know that in most people aortic stenosis progresses slowly  – it stands to reason that we need to consider intervening earlier in the course of the disease.  (Before the heart is significantly weakened).

For people with Aortic Stenosis – I’d want to get second/ third opinions from a cardiac surgeon before proceeding with any catheter based valve procedures.  I’d bring all of my information, and studies (echocardiogram results, lab results, medication lists) to have a serious talk to the surgeon about my surgical risk – (and ask him to calculate and show my risk based on the STS calculator).  I certainly wouldn’t let anyone make any decisions about my health/ care based on my age alone.  [we’ve talked about a ‘good 80’ versus a ‘bad 80’ or even a ‘bad 50’ before.]

Then, I would weigh all of the information – and do my best to make an informed decision.

Other posts about Aortic Stenosis/ TAVI/ AVR:

1. TAVI overview

2. TAVI: a stent scenario

3. TAVI recommendations

4. Aortic Stenosis and TAVI

5. Aortic Stenosis as Heinz 57

6. Aortic stenosis and surgery


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s