First talk from Prof Takkenberg, who needs no introduction to any of you. A very important talk in which she questioned the primacy of RCTs and spoke up on the importance of 'real world' registry data, often dismissed as poor quality
Correctly, she highlighted that 1 issue with RCTs is the very selective nature of the patients enrolled - often leading to much better outcomes in trials than we see subsequently in real life - TAVI & SAVR example here
Concluded by emphasising that RCTs and observational / real world data should be viewed in combination and as complimentary, not as RCTs are the untouchable gold standard that should never be questioned...
Starts off with an explanation of the valve structure.
Based on the highly successful Perimount valve design, with new features to reduce leaflet calcification & for ring to 'expand' to facilitate future V-in-V TAVI when AVR degenerates
This slide is a little over-hyped imho...the valves were explanted after just 8 months...we need a longer timeframe, but I understand the premise - these leaflets are hopefully less prone to calcification
Maybe I'm just spoilt by excellent outcomes where I work, but I was worried by these data
30-day mortality 2.6% in a cohort with mean age just 60, and 94pts are <60yrs old? We didn't have STS scores that I saw, but this seemed really high to me
Again, the data (that Prof Meuris' team conducted) showing significantly less leaflet calcium on the Inspiris compared to Perimount valve after 8 months in an animal model...my thoughts on this later...
The pitch was very much that these valves may be used in patients that would otherwise receive a mechanical valve - mAVR means more bleeding, tAVR means more re-do procedures
There was NO mention of the cost of the Inspiris Resilia valve itself and how that compares to a mechanical valve...see end of thread for why that matters...
These slides were used to illustrate that anticoagulation comes at a price, soemtimes due to low INR needing admission for correction or bleeding complications causing admission and possibly major harm
So, my thoughts? Well, this was a cleverly crafted session. The constant theme through the talks was to question the primacy of RCTs & showcase the utility of real world data
Now, as it happens, I *agree* that RCTs have major flaws, especially some done recently...
Cherrypicking patients to maximise likelihood of a significant difference between two arms is not useful, if the outcomes achieved in an RCT - upon which FDA / CE approvals may be based - simply cannot be recreated in our daily practice
However - where I disagreed is what we do about this. The answer is not to embrace retrospective observational studies even more but to improve RCTs! We *have* to demand better of our randomised studies. There's a reason randomised studies are considered best
Randomisation evenly & randomy distributes all variables - known & unknown - that could influence outcomes
So, must do better to enrol women, to enrol ethnic minorities & to reduce exclusion criteria...of course, *obvious* reasons trial sponsors may not be keen
Our surgeons were not very happy. But, on this occasion, I had some sympathy with the managers too. The new valve is DOUBLE the cost of the old valve and there are ZERO data (by definition) *proving* that it lasts longer than the original Perimount Mgane Ease
I'm sure I'm coming across as too cynical. I don't mean to. I applaud Industry for spending R&D funds on improving bioprostheses & Edwards has been a leader in the field
But, it would be nice to have a well-conduced RCT of Inspiris vs mAVR in 50-60yr olds
End! 😀
• • •
Missing some Tweet in this thread? You can try to
force a refresh
It takes time to read the paper, read the supplementary appendix, analyse the results, think about them etc!
Some thoughts...🧵
On Sunday I wrote a thread about asymptomatic severe AS and what we knew already from RECOVERY & AVATAR RCTs and what the guidelines currently advocate
Transthoracic echocardiography (TTE), when performed with care and diligence, can reveal a lot about the valve. TOE isn't necessary in all cases to determine leaflet pathology.
A worked example below:
In the PLAX view, you can assess the scallops of the leaflets
In a true PLAX view with aortic valve clearly visible, you mostly see the A2-P2 interface. Here, you can see a clear & large prolapse of the posterior leaflet
If you tilt upwards towards the PLAX RV outflow (pulmonary valve) view you see mostly the A1-P1 interface
Here, you can see the valve looks slightly different & no prolapse is seen
PE has garnered a reputation for huge profits as they typically buy into a company, aggressively ⬇️ costs whilst ⬆️ profit margins, leverage debt if needed, and then sell, often at a large profit
Does this matter when it comes to delivering healthcare services?
Sixty years ago this month - October 1963 - US cardiologist Dr Robert Bruce published a paper detailing his efforts to devise a multi-stage treadmill test
Little did he know this would go on to become the Bruce protocol the most widely used for exercise testing
A thread...
Prior to this, 'stress testing' to evaluate cardiac function was performed using the Masters 2-step technique, first described in 1935. This crude but simple test involved repeated steps up & down over 90 seconds
Bruce, amongst others, recognized the limitations of this test & set about developing an exercise treadmill test
His initial work focused upon a single stage protocol, but he soon realized this wasn't stressing fitter patients enough & the test was taking too long!