The Vascular Leaders Forum (VLF, 1–2 March, Washington DC, USA), hosted by the non-profit organisation VIVA Physicians, is a special consortium which was called in the wake of the meta-analysis published by Katsanos et al late last year, and the discussion of paclitaxel-eluting devices which ensued. The event was live-streamed on the VIVA Physicians website and Vascular News provided live updates here with highlights from the sessions.
2 March, 17:00 GMT
Gary Ansel (Columbus, USA) states that the discussion and consensus reached over the past two days will be communicated in a paper and published “fairly rapidly”.
Christopher White now begins the conclusion of the VLF summit, asking: where do we go from here? “I am so reassured,” White says, “and very proud of us. Our practices get challenged by a meta-analysis that is not the strongest of meta-analyses and yet, instead of being in denial [and] a self-interest, we recognise the validity of this signal, and we commit to solving this problem.”
To Zeller’s previous point (below), White advises “also to remember that the other side of that coin was that the follow-up to ascertaining mortality was greater in the experimental arm, as opposed to the control group. So not only do the patients not get followed up or get the treatment that they maybe need, but then we find out they died more. We are looking at something that is almost like a perfect storm.”
2 March, 16:55 GMT
Thomas Zeller (Bad Krozingen, Germany) says one insight for him from the VLF meeting was that “I underestimated the compliance issue of our patients when we are enrolling patients in studies which are aiming to prove superiority in technology, whether this is a drug or a device. We knew a long time that patient compliance to undergo a follow-up angiogram is lower in the cohort which is doing well, as compared to those who have got symptoms again. What I did not realise, which has been shown here, is that follow-up compliance for any kind of follow-up visit—even non-invasive—is low in those who are doing well. This is a potential bias I personally underestimated in the past, and may have contributed to the outcome of this meta-analysis. We need to really have a look at that as one of the topics of what can be improved.”
2 March, 16:30 GMT
Donna Buckley of the FDA, who has “reviewed most of these devices and recommended them to be approved”, says “Our first response to the meta-analysis was, I think, similar to everybody else’s: we were a little bit surprised.” Given she herself had reviewed the devices and found no cause for concern in the data, Buckley underlines her personal surprise but states, “nonetheless, it was compelling information that we took very seriously.”
Focusing primarily US-based RCTs with data the FDA can validate, Buckley says they have “converged to where we feel like it [the meta-analysis findings] is not a statistical glitch—but we still have all this incongruous information as well with all the statistical analysis that we have gone through.”
Peter Schneider (Honolulu, USA) calls the process that the FDA is undertaking to review the data “judicious, straight-forward and spot on”.
2 March, 16:07 GMT
An audience polls finds 35 of 36 VLF respondents were “surprised by the JAHA meta-analysis suggesting increased mortality in patients treated with paclitaxel-coated devices.” The next poll, however, finds 14 of 35 respondents would be “more conservative” with their recommendations of drug/device combinations for claudication, based on the discussions at VLF.
When a new endovascular strategy is approved and reimbursed in the USA, 24 of 38 attendees respond that their adoption of that strategy “depends on the strength of the data”. “Data rules in this academic group of clinical experts”, says Michael Jaff.
To the question whether respondents have had “unexpected mortality events” in patients they have treated with paclitaxel-coated peripheral devices, 20 (50%) say they do not know, while 17 respondents say they have not. Only three responded affirmatively.
Has your hospital/outpatient centre talked to you about continued use of paclitaxel-coated peripheral devices since the JAHA meta-analysis was published? In the VLF audience, 30 of 39 say they have not. A “resounding” 30 of 39 respond there should not be local restrictions with these devices—only five opting for the answer, “we need more data”.
In response to three attendees who answered a poll with a call for another RCT to clarify the mortality risk, Michael Jaff says “not in my lifetime will that happen”. A majority of 27 out of 40 answered they would need to see merged patient-level datasets from all manufacturers.
Finally, 32 of 37 say “current trial enrolment in claudicants” should continue, while five say it should stop, and only 5 of 37 plan to change their clinical practice as a result of the meta-analysis.
2 March, 14:50 GMT
Michael Jaff (Massachussets General Hospital, Boston, USA) describes reading the Vascular News breaking story of the Katsanos meta-analysis on his phone, as he was about to walk into a consultation with a patient. In a talk titled “Public hysteria and potential adverse impact on patients: Is there a better way to publish data?”, Jaff discusses the manner in which data should be released, and the “leak that lead to a frenzy even before the paper was published”.
“This meta-analysis was leaked to media outlets well before the paper actually came out”, Jaff says, adding “that leak lead to a frenzy even before the paper was published”. Pointing to the JAHA embargo rules, Jaff says that the rules are subject to some degree of interpretation.
While acknowledging that “all of us [researchers]” have given early interviews or data release to the media, and a paper released in press simultaneous to its presentation or journal publication is a “great feather in the cap of any academic researcher”, Jaff nevertheless states: “as far as I am concerned, unless the authors felt that there was an immediate risk to public health and safety—in which the leak should have been to regulatory authorities—it should not have been leaked to media outlets to be discussed at meetings weeks before the paper came out.”
Highlighting a sentence in the meta-analysis conclusion which reads “Further investigations are urgently warranted”, Jaff argues this wording “was interpreted by many to mean there is a serious threat to public health. And, given we have just spent the last day trying to figure out what the heck is going on […] it is hard to say that ‘urgent’ was was warranted. That alarmist comment, I think, added to the frenzy that we are all dealing with today.”
In the ensuing panel discussion, Stephen Greenhalgh, London, UK, speaking on behalf of Interventional News (that first broke the story) and Vascular News, commented on the difference between responsible and irresponsible media coverage. He made the point that there was a high-level editorial decision to report “on this important signal” once there was an awareness of the impending publication of the Katsanos et al meta-analysis in JAHA.
Greenhalgh said it had been important to include reactions from key opinion leaders and the physician community (many of whom were gathered at VLF 2019) in the breaking-news article that reported on the findings from the meta-analysis to attain a “level of balance”. Importantly, he also clarified that both Interventional News and Vascular News did not publish the article until after the journal had published the paper, so that the embargo was not, in fact, broken by the newspaper, despite the findings of the paper being shared ahead of time. “My organisation worked with the journal on the timing of our article relative to the publication of the original meta-analysis,” he said. The approach taken by both Interventional News and Vascular News was only to publish a factual account without editorial commentary, he noted, saying at VLF 2019 “so far, I think we have had responsible media covering this [issue]”.
2 March, 13:50 GMT
Paul Rudolf (Washington DC, USA) now discusses “informed consent” regarding the use of paclitaxel-coated and -eluting devices, and expands on the legal implications of the Katsanos et al meta-analysis.
“Yesterday, I was really struck by the patient that was up here. The patient basically said ‘If I trust a doctor, and I have confidence in the doctor, and I believe what the doctor tells me, then I am fine.’ I think that has been the experience of most physicians, and statistically, patients that trust their doctors just don’t sue them. I think it is a case of doctor’s gaining the trust of their patients, which is nothing new to anyone in this room, but was just something that really stuck me yesterday.”
He continues, “The other thing I was struck by is that, as usual, we always find that medicine is an art, not a science. Everything we do is informed by scientific evidence, but at the end of the day we have to make informed decisions with out patients, and those are always individualised decisions.”
Saying he did not want to trivialise the findings by Katsanos et al regarding the mortality signal, Rudolf says: “Even if the data in the meta-analysis is true, at five years only one out of 14 patients would be affected by the mortality signal, which means that 13 out of 14 patients that you would be treating with a drug-eluting balloon or stent would still do just fine and not be affect by that.”
Rudolf also says that 40-45% of malpractice law suits hinge on an inadequate discussion of the complications of the procedure prior to treatment. He emphasises that in the case of paclitaxel, informed consent is the most important factor to consider when thinking about the legal implications of treatment with paclitaxel devices, and also says that as it is mortality that is being discussed, physicians should bear in mind that any possible legal case would be bought by the patient’s estate and family.
As doctors treating patients with PAD in the lower leg indicated for use with paclitaxel devices already have “complicated informed consents” due to the risk of amputation and recurring disease, Rudolf says that, from his perspective, “I am not sure how different the informed consent will be. In my view, this is just something that is added on to what is already there.”
One difference in informed consent between before the Katsanos et al meta-analysis was published and since is that, if medical therapy is a good treatment option for patients, then that should be included to a greater extent in the discussion with patients.
The most important piece of advice Rudolf has for physicians is to stay up to date on device manufacturer information, FDA recommendations (which he says “should be followed, but are not binding”), and practice guidelines.
Informed consent should be patient specific, Rudolf adds. So disucssio of this mortality signal may be more pertinent to a 50 year old patient than to an 88 year old patient, he says, and previous exposure to paclitaxel devices may also heighten the significance of discussing the meta-analysis results.
2 March, 13:42 GMT
Marianne Brodmann (Graz, Austria) takes to the stage to discuss how the meta-analyssi is being dealt with in Europe.
“You have already heard how we have a more hetergeneous approach in Europe than the USA, because you have heard that the SWEDEPAD and BASIL trials are on hold. What I am talking about is the general approach in Europe, mainly the central European approach,” she begins.
“We have a great tradition in Europe with DCBs and with large RCTs”, Broadmann says.
From early data, Broadmann says that the main concern with DCBs in Europe is of the local effect of paclitaxel (aneurysm risk).
She says that the FDA letter of 17 January 2019 saying that the regulatory body were starting their own investigation caused some concern amongst European peripheral interventionalists.
“We still think it is a safe technology we can use,” Broadmann says of paclitaxel-coated and -eluting devices, following a summary of patient-level data presented at the Leipzig Interventional Course (22-25 January 2019, Leipzig, Germany).
However, despite this cautious optimism, Broadmann says that she saw this morning a headline in a German newspaper, the Frankfurter Allgemeine, reading [translated]: “Within a hair’s breadth of a scandal.” See image below.
However, Broadmann says the content of the article is not as “hard” as the headline, and includes statements from the German Society of Angiology, but that “we never know how patients will respond to such articles.”
2 March, 13:34 GMT
Yoshiaki Yokoi (Osaka, Japan), of the IN.PACT Admiral (Medtronic) SFA Japan trial, explains the ramifications of the JAHA meta-analysis in Japan.
“In Japan, every trial conducted in the USA we have to repeat, to get approval [in Japan]. We have a delayed approval process,” he explains. Based on the IN.PACT Admiral SFA Japan trial, the DCB was approved. However, the US and Japanese approval differed slightly in their indications: whilst in the USA the IN.PACT Admiral paclitaxel-coated balloon could be used in de novo, restenotic, or in-stent restenotic lesions with up to 360m in SFA, in Japan the device could be used in de novo or restenotic lesions, with up to 200mm in SFA. There was no indication in Japan for in-stent restenosis, “a unique regulation”.
Looking at paclitaxel-coated balloon use in his own hospital since the publication of the Katsanos et al meta-analysis in JAHA, Yokoi says, “Fortunately, no obvious paclitaxel-related death was found in our hospital.”
Yokoi says he believe analysis should be separated between DCBs and DESs, and the “paclitaxel dose is 20 times more [for DCB] than in the DES.”
“There us a concern about what the long-term, downstream effect of paclitaxel is”, Yokoi says.
To be “hopefully published soon”, Yokoi tells the VLF audience that he and his colleagues have analysed the cause of deaths in the Japanese cohorts of the IN.PACT Japan trial, the LUTONIX Japan trial, the Zilver PTX trial, and the Eluvia trial up to five years.
Speaking of these results, Yokoi informed delegates that the LEVANT Japan and IN.PACT Japan trials did not show any significant increase of mortality in the DCB arm. Bases on these data, the DCB was accepted by regulatory bodies in Japan, and use started from January 2018. The maximum dose of paclitaxel indicated for use in Japan is 30–40mg. Yokoi says that “A slow flow after DCB use is observed, and there is concern about downstream effects of paclitaxel. We are under investigation of previous Japanese RCTs in regards to mortality and paclitaxel-related complications.”
His personal view is that in most DCB studies, primary patency was systematically used as the primary endpoint, and that this metric may be misleading for the evaluation of real clinical outcome. He concludes by stressing, again, that the downstream effect should be thoroughly investigated.
2 March, 13:20 GMT
Mårten Falkenberg (Gothenburg, Sweden), of the SWEDEPAD study, gives a talk entitled “A clinical trial paradox: Reaction by EU and American investigators to this analysis”.
The SWEDEPAD trials address the “Achilles heel” for endovascular treatment of PAD. Falkenberg says “I think we all agree drug-eluting technology is the most promising new tool” for PAD treatment, but also says that “randomised trials are underpowered to address patient-entered clinical endpoints.” He says there is a need for an independent, large clinical trial with patient-centred endpoints to address these limitations.
“In order to have patient-centred important outcomes, we decided to split the trial.” Falkenberg says. SWEDEPAD 1 will enrol patients with CLI, and SWEDEPAD 2 will entol patients with intermittent claudication. Currently, 1,480 are included in SWEDEPAD 1, and 810 patients are enrolled in SWEDEPAD 2, meaning enrollment is approximately two thirds complete.
Both studies halted enrolment following the publication of the Katsanos et al meta-analysis. “We are at the moment struggling to make up our mind on how to proceed”, Falkenberg concludes.
2 March, 13:00 GMT
Misti Malone, chief of the peripheral interventional group at the FDA, provides FDA insight of the paclitaxel mortality signal. She thanks audience members for participating in this collaboration, and explains that, as is understood in the Food Drug and Cosmetics Act, safe does not mean risk free when it comes to medical devices, but that it is important to gain greater clarity on the mortality signal Katsanos et al reported in their meta-analysis.
Factors to consider when assessing a mortality signal:
- Likelihood of the event
- Quality of data
- Strength of causal relationship
- Risk severity
- Risk duration
- Risk reversibility
- Benefit magnitude
- Patient vulnerability
- Alternative therapies
- Implications for related devices
The FDA is in the process of replicating the meta-analysis, using both similar and alternative statistical methodologies to those used by Katsanos and colleagues. This analysis will include all 28 RCTs in the original meta-analysis.
Speaking of this investigation, Malone says: “Following preliminary review of the data, we believe that a signal still persists, and that this warrants additional investigation. We acknowledge that this may impact other disease states, such as AV and CLI.”
The FDA recognises limitations of the available data. According to Malone, these include a lack of standardised categorisation and adjudication of death; unknown causes of death and co-morbidities; treatment with other devices; trial period and ever-changing medical practice.
Malone says that the FDA will use this information to assess the risk for future recommendations, “including a potential advisory panel which may suggest potential modifications to trial design, such as informed consent and labelling, and/ or additional post-market data collection.”
1 March, 20:16 GMT
Thomas Zeller (Bad Krozingen, Germany) says that after his talk on the THUNDER trial and long-term follow-up, “I believe we will understand the potential flaws of the Katsanos study.”
Naming the THUNDER trial and the FEMPAC trials as the beginning of the paclitaxel story, Zeller describes the importance of understanding what happened in the THUNDER trial with regards to mortality: “At five years, we found no late catch-up with regards to clinical efficacy, favouring the DCB concept. …It is important to note, at two years we excluded patients who underwent more than one TLR. This could have been endovascular reintervention or bypass surgery.” By five years, this number of excluded patients rose to 25, 11 of which were in the paclitaxel cohort—”and that is the follow-up bias, basically,” Zeller says, “which is important to realise, but was not included in the Katsanos analysis.”
Regarding the risk ratio, when adjusted to take into account the lesser number of patients available for follow-up at five years, “the end of the story is that the risk ratio is significantly lower when patients where you really have the appropriate clinical information,” says Zeller, adding this is “one of the flaws in the Katsanos study, and I believe this is not only true for the THUNDER trial.”
Summarising his critique of the meta-analysis as it relates to the THUNDER study, Zeller argues: “paclitaxel-coated balloons are more effective than POBA, individual plasma concentrations vary, a constant dose to clinical response correlation does not exist in my opinion, and the risk of five-year mortality depends on the statistical method applied.”
1 March, 18:30 GMT
William Gray (New York, USA) now takes to the stage to present a critical appraisal of the JAHA analysis.
He summarises: “The pharmacokinetics and toxicity of paclitaxel is well studied and characterised, and there are no prior data that indicate that very, very low doses that are present as part of DCB and DES use produce a mortality effect. Therefore, in order to claim otherwise, and describing a novel, and grave, paclitaxel toxicity, never been described in the history of the world, one would need to present Level 1 evidence: prospective, randomised, and appropriately sized.
“A non-patient level meta-analysis with multiple flaws—many fatal—will likely not suffice.”
He lists several flaws of the meta-analysis:
- Katsanos’ paper assumes a random-distribution model, which he says is “appropriate”, in its statistics, but “did not make adjustments for a non-parametric (skewed) distribution. This is inappropriate.”
- There is a selection bias due to the lack of a complete follow-up. One year outcomes in the combined 28 studies show no differences in mortality. Two- and five-year data includes a subset of 12 and three trials, respectively, of the original 28. A more appropriate time-to-event analysis cannot be performed and could not be performed on the data available to Katsanos. At one year, trials included with long-term data show a mortality effect, whereas the trials without long-term data show no mortality effect
- “The dose-time product is problematic.” He says the selection bias “plays out here as well”, and that the variable “Time” in the dose-time calculation is disproportionately available for the longer-term follow-up studies and worse mortality, biasing the calculation. He adds: “Once the study population is divided by DCB length and size, adjustments for possible differences in population cardiovascular risk factors would need to be made. It’s no longer a randomised application or trial. It is now a trial of sub-setted patients using different sizes of balloon. That needs to be back-adjusted, and it was not.”
- The need for—and promise of—meta-analysis is not the assessment of low frequency events that are not typically individually powered. In the case of DCB and DES studies, death was part of a composite safety endpoint. According to unnamed statisticians Gray spoke to, “A prospective study to independetly assess mortality in DCB/ DES for PAD would need to enrol between 2,000 and 6,000 patients to avoid the risk of Type 1 error.” Whilst the meta-analysis begins with 4,663 patients across 28 RCTs, at two years, there are 2,316 patients across 12 RCTs, and at year five, there are 863 patients acrss three RCTs.
- The lost-to-follow up are not properly accounted for in this trial; Gray does some back of the envelope calcuations to show that taking the lost-to-follow up patients into account, the relative risk from the five year RCTs included in the study is lower than cited in the Katsanos meta-analysis.
- Gray claims the PTA group is likely not paclitaxel naive for the entirety of the analysis. This is because paclitaxel device approvals in Europe and the USA preceded almost all of the one year period of TLR experienced in the PTA arm. “That is, the trial finished, and by the time the one year TLR arm was finished, there was available paclitaxel in the world.” Gray says that as most of his practice uses paclitaxel to treat restenosis, “I think the authors would need to recognise this, and perform some sliding scale analysis of the impact of variable proportions of paclitaxel usage in the TLR patients. Without that, I think this whole analysis blows up, because this assumes that these patients are paclitaxel naive for the duration, rather than move them into the column of paclitaxel mortality.”
1 March, 18:15 GMT
Konstantinos Katsanos (Patras, Greece) presents “The Why, The How and the Results” of the JAHAmeta-analysis, saying “I guess most of you have already read the paper.”
“We only looked at randomised controlled studies—exactly to try to tease out only the effect of paclitaxel,” Konstantinos explains, “considering that ideally in a well-designed randomised study, all the other confounders are supposedly accounted for.”
“The tissue half-life of paclitaxel being administered by a DCB or DES seems to be in the order of several weeks, if not months. Above all, I was sort of surprised to find out but there has been recently identified a drug-drug interaction of the metabolism of clopidogrel and paclitaxel—and remember now we are referring to PAD patients.”
Katsanos directs attention to a dataset he finds “highly important”—the IN.PACT SFA suvival analysis: “The hazard ratio of the two arms on the log scale from the beginning up to five years, and you can see that those are non-proportional hazards. The curves are diverging, and then they are converging again. So hazard ratios may not be showing, and may not be documenting the full picture.”
“I think we should not contaminate randomised studies with individual patient data of single arm registries”, Katsanos argues, “before we make sense of the randomised studies. I think what would be a mistake—and this seems to be the avenue taken by the industry—to try to contaminate randomised studies with the inclusion of single-arm registries. These single-arm registries have different follow-up, different characteristics. We need to stick to randomised evidence and we need to agree on the proper way to analyse this evidence.”
Read more about Katsanos et al‘s findings here.
1 March, 16:29 GMT
Elazer Edelman, who is the director of the MIT Institute for Medical Engineering and Science and professor of medicine at Harvard University, gives an overview of drug-delivery kinetics and tells the panel: “It does not make any sense to indict the drug for having an effect at five years. […] I do not think we can indict the drug because there is no mechanism that I can envision, that would account for mortality. It does not mean there is not a mortality signal and it does not mean we should not be doing more studies, but at the end of the day we need more data, and we need more people to support the pre-clinical and non-clinical work.”
1 March, 14:16 GMT
Ramon Varcoe (Sydney, Australia) kicks off the second session of VLF 2019 with an overview of “Unintended consequences of various trial designs and their potential effect on mortality and outcomes”.
Considering how trial design can affect outcomes and specifically mortality, Varcoe notes that as randomised controlled trials for interventional studies—as opposed to drug trials—are not as easily blinded, several types of biases can be hard to avoid. “Outcome assessors (for mortality, this is usually study coordinators or investigators) may unconsciously or intentionally alter their assessment”, Varcoe notes.
“How can we determine whether the mortality in the Katsanos review was due to paclitaxel or to trial design?” Varcoe asks, noting that all the studies in the meta-analysis had a “high risk of bias” in this area. Performing an identical meta-analysis of RCTs in the SFA, but excluded all drug-coated devices, to check whether an association was still found. With 22 studies vs. the 28 reviewed by Katsanos et al, and approximately half the subjects (2,451 vs. 4,432), at 12 months Varcoe and colleagues found “a higher risk ratio which favoured the control groups in terms of mortality outcomes—and this was highly statistically significant”. At two years, the control review by Varcoe dropped to half the amount of studies and about a third of the study subjects, rendering the findings underpowered. However, Varcoe noted, “there if a definite trend” which leads in the same direction, and continues out to five years.
“It is my view that these findings cast some doubt as to the causal link of the association between paclitaxel and mortality.” Instead, Varcoe suggests, “it is much more likely” that this association is due to the introduction of bias, more tenacious follow-up of subjects in the experimental arms, and higher rates of medical interaction in RCT arms with more frequent target lesion revascularisations.
1 March, 13:35 GMT
Josh Beckman (Nashville, USA) speaks on variables that may affect mortality in patients with PAD.
“My talk is really simple”, Beckman says, listing four factors that “are enough to tell us about the variables that affect mortality”:
- How bad is the PAD?
- What are the risk factors that cause PAD?
- What is the atherosclerotic burden doing?
- What can the patient tell you themselves about how they feel?
“I think, actually asking people how they are doing is a way to integrate every aspect of their problems.” Pointing to a 2006 study by P Garg et al, Beckman notes in this study, to predict twice the risk of mortality, “all you had to do was ask: do you leave your house? So functional capacity is a nice way to integrate everything we are talking about, in terms of risk factors, disease severity, and polyvascular disease.”
“In summary, the primary cause of mortality in PAD is cardiovascular, followed by cancer. […] Prior revascularisation per se did not affect mortality.”
1 March, 13:26 GMT
Raghu Kolluri (Columbus, USA) takes to the podium to talk about mortality rates in randomised vascular device trials. Sticking to the femoropopliteal studies, Kolluri begins by saying: “Mortality is not reported in most of the trials, and we don’t have enough long-term data—this is evident from the Katsanos paper itself. At two years, you have twelve studies to compare, and at five years you have only three studies to compare.”
Speaking of the 2016 Circulation paper where the mortality data in the treatmentarm (treatment with the Zilver PTX paclitaxel-eluting stent from Cook Medical) and the control arm (percutaneous transluminal angioplasty [PTA]) were inadvertently swapped, Kolluri says: “When I looked at the numbers, the Katsanos analysis actually included the flipped numbers, so I don’t think this would have mattered in this situation.”
Kolluri cites several recent patient-level analyses which fail to find a statistically significant difference in all-cause mortality between DCB treatment and a control. These include the pooled analysis of all IN.PACT Admiral DCB clinical programmes (from Medtronic), the RANGER SFA randomised three-year data (Boston Scientific), the Zilver PTX data (Cook Medical), and the Philips’ Stellarex data, all reported at the Leipzig Interventional Course (LINC; 22-25 January, Leipzig, Germany).
“The biggest issue I had with this question that the organising committee [of VLF] put to me [What are the mortality rates in the published randomised vascular device trials?] is that mortality was not reported in many trials, and mortality was not the primary endpoint,” Kolluri says.
“The other interesting study that is important to note is the VIBRANT study. VIABAHN compared to the BMS [bare metal stent] showed [a] higher death rate—almost three times—which does not make sense in regards to how that can happen […] I am intrigued. None of these [deaths] were related to the device, it is important to note that.”
Talking of the COMPASS trial, Kolluri notes that the all-cause death rates at 21-months were 5% in each arm, even without any intervention.
In summary, Kollui says: “The paclitaxel, at two years, demonstrated a 7.2% [mortality rate].” He goes on to list the mortality rates of: bare metal stent (4.4% to 9.8%), atherectomy to three years (12.5%), atherectomy (6.9% to 12.5%), PTA (0% to 12.2%), and medical management (0% to 4%).
He concludes: “Honestly, I don’t know why this is happening, I am intrigued, and I am here to learn.”
1 March, 13:10 GMT
An audience poll finds that out of 40 attendees at the VLF, 28 respond they are “concerned about the potential link between paclitaxel-coated devices used in PAD interventions and mortality”.
1 March, 13:00 GMT
Gary Ansel (Columbus, USA) opens VLF meeting with an introduction to the goal of the event. Providing a background, Ansel says “all VLF symposium are funded solely and entirely by the VIVA organisation.”
“With the recent JAHA meta-analysis that suggests increased risk of mortality from commonly utilised endovascular devices, we felt strongly that physicians needed to assume a leadership in investigating this potential risk in as transparent, high-level and inclusive a method as possible. To this end, this meeting is being live streamed, so that all those interested may observe the proceedings.”
Ansel says the “most important part of this Vascular Leaders Forum is the hours of open discussion.”
“We are all here to ensure that what we do is best for our patients—after all, that is why we do what we do.”
1 March, 12:30 GMT
First session set to begin at 13:00 GMT.