An Up To Date Meta-Analysis on Paclitaxel Containing Devices in Femoropopliteal Arterial Disease, And A Commentary On The Entirety of the Data to Date.
Clinical question
Is paclitaxel safe when used in drug eluting stents and drug coated balloons for revascularization of the femoropopliteal artery?Take away point
Paclitaxel containing devices are safe and effective devices, especially in their current usage and at their current generation. Previous versions of the devices may have conveyed some mortality risk, but it’s important to understand both the meta-analyses and the individual studies that comprise the data reporting, and how clinical practice is nuanced and so does the literature..Reference
Katsanos, K. (2024). Paclitaxel meta-analyses in the lower limbs: Missing the trees for the forest. Journal of Vascular and Interventional Radiology.Click here to access article
Briody, H., Kearns, C. A., & Lee, M. J. (2024). Mortality, safety and efficacy of paclitaxel-containing balloons and stents in the femoropopliteal artery: systematic review and meta-analysis of randomized controlled trials since 2018. Journal of Vascular and Interventional Radiology.
Click here to access article
Commentary is from University Hospital Patras, School of Medicine, Rio, Greece.
In revascularization efforts for peripheral arterial disease, paclitaxel has proven to be an excellent anti-restenotic agent in the femoropopliteal artery when used in drug coated balloons and drug eluting stents in multiple randomized controlled trials. While effective, its safety profile was called into question in a 2018 meta-analysis. The meta-analysis, published by Katsanos et al, found statistically significant increases in all-cause death in patients with paclitaxel coated devices, and urged further investigations. In 2023, a patient level, industry funded meta-analysis concluded there was no excess mortality. Based on this study, the FDA published their official statement that there is no discernible excess of mortality.
The current meta-analysis was performed to provide an independent analysis of safety and effectiveness outcomes, including but not limited to mortality. The literature search was performed from 2018 (a choice made so this study would act as an update of the aforementioned Katsanos study) and included 19 randomized controlled trials for a total of 4,284 participants. When selecting studies, the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) was utilized (see figure). Effectiveness outcomes were primary patency which was analyzed using risk ratios via inverse variance, random-effects model. Safety outcomes were all cause mortality, target limb amputation, target lesion revascularization, clinically driven target lesion revascularization, and thrombosis. These were also analyzed using risk ratios via inverse variance, random effects model, as well as heterogeneity assessment via visual inspection of forest plots, chi squared test and I^2 statistic.
The results found no evidence for all-cause mortality in paclitaxel containing devices in the femoropopliteal region from 12 to 60 months, nor did they find any other safety concerns. The study redemonstrated the effectiveness of paclitaxel devices in maintaining primary patency with a pooled risk ratio of 1.55. Taken together, this meta-analysis concluded that paclitaxel containing devices work, and are safe to use. They compare this study directly to the Katsanos study due to its similar methodology and design, however they offer no suggestions as to the source of the different findings.
In response to the multiple meta-analyses looking at the use of paclitaxel containing devices in the femoropopliteal arteries with no evidence of increased all-cause mortality, Dr. Konstantinos Katsonas provided a companion commentary. In the piece, he presents 27 studies from 2008-2021 which show an increase in all-cause mortality in the paclitaxel containing device arm, arguing that the recent studies demonstrate dilution of the mortality signal, but do not eradicate it completely. He presents a meta-regression of long-term risk ratio of all cause death against publication year and shows that the observed mortality risk decreases with publication year, which would seem to suggest earlier generation devices or clinical practices may be the cause for the mortality signal. He further subgroups the studies by paclitaxel dosage and shows that the relative mortality risks may be different among the different devices with variable paclitaxel dosing. He concludes that the reader must not miss the trees for the forest, meaning that the individual studies may have points of interest that become diluted in meta-analysis.
The academic discussion around paclitaxel can make it difficult for providers to know if these devices are safe or not. This most recent meta-analysis is an excellent consolidation of the most recent literature on paclitaxel containing devices in femoropopliteal artery treatment. The results add to the body of evidence that these paclitaxel devices are safe and effective tools in peripheral arterial disease, especially at their current generations. Providers can use these devices and expect good outcomes for their patients without an increased mortality risk.
That being said, Dr. Katsonas does raise interesting questions. The 2018 Katsonas et al meta-analysis did show a significant increase in all-cause mortality. In his commentary, this effect appeared again when looking at all studies from 2008-2021. While there were shortcomings in the Katsonas paper, the fact remains that a mortality signal was present and hasn’t been fully accounted for. The fact that it seems to disappear when looking at studies from only 2018 onward could be indicative of the root cause. Moreover, the more recent studies did not subgroup based on device or paclitaxel dosing. The data he presents on the decrease of all-cause mortality risk over the years, as well as the trend of lower risk ratios with lower paclitaxel dosing, are suggestive that the mortality signal may be device-, device-generation, and/or paclitaxel dosage-dependent. Of course, this is only speculation based on available data, but could make excellent areas of inquiry should the question of safety in paclitaxel containing devices remain.
Dr. Katsonas’ final point is to not miss the “trees” (individual studies) for the “forest” (meta-analysis), as individual studies can yield insights that may be diluted in meta-analyses. However, he doesn’t really support this in his commentary, as most of his points are made by using grouping statistics based on multiple studies. Regardless, it is an important point to keep in mind when reading meta-analyses. Procedural practice and published literature are nuanced, and evidence-based medicine should be supported by both high-level meta-analyses as well as individual studies that befit the patient population and institutional experience.
Post Author
Sean Rogers, MD
Interventional Radiology Fellow, PGY-6
University of Massachusetts Chan Medical School
Study design
Meta-analysis, commentary/perspectiveSetting
Meta-analysis was performed at Beaumont Hospital, Dublin Ireland and the Royal College of Surgeons in Ireland.Commentary is from University Hospital Patras, School of Medicine, Rio, Greece.
Figure
Summary
In revascularization efforts for peripheral arterial disease, paclitaxel has proven to be an excellent anti-restenotic agent in the femoropopliteal artery when used in drug coated balloons and drug eluting stents in multiple randomized controlled trials. While effective, its safety profile was called into question in a 2018 meta-analysis. The meta-analysis, published by Katsanos et al, found statistically significant increases in all-cause death in patients with paclitaxel coated devices, and urged further investigations. In 2023, a patient level, industry funded meta-analysis concluded there was no excess mortality. Based on this study, the FDA published their official statement that there is no discernible excess of mortality.
The current meta-analysis was performed to provide an independent analysis of safety and effectiveness outcomes, including but not limited to mortality. The literature search was performed from 2018 (a choice made so this study would act as an update of the aforementioned Katsanos study) and included 19 randomized controlled trials for a total of 4,284 participants. When selecting studies, the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) was utilized (see figure). Effectiveness outcomes were primary patency which was analyzed using risk ratios via inverse variance, random-effects model. Safety outcomes were all cause mortality, target limb amputation, target lesion revascularization, clinically driven target lesion revascularization, and thrombosis. These were also analyzed using risk ratios via inverse variance, random effects model, as well as heterogeneity assessment via visual inspection of forest plots, chi squared test and I^2 statistic.
The results found no evidence for all-cause mortality in paclitaxel containing devices in the femoropopliteal region from 12 to 60 months, nor did they find any other safety concerns. The study redemonstrated the effectiveness of paclitaxel devices in maintaining primary patency with a pooled risk ratio of 1.55. Taken together, this meta-analysis concluded that paclitaxel containing devices work, and are safe to use. They compare this study directly to the Katsanos study due to its similar methodology and design, however they offer no suggestions as to the source of the different findings.
In response to the multiple meta-analyses looking at the use of paclitaxel containing devices in the femoropopliteal arteries with no evidence of increased all-cause mortality, Dr. Konstantinos Katsonas provided a companion commentary. In the piece, he presents 27 studies from 2008-2021 which show an increase in all-cause mortality in the paclitaxel containing device arm, arguing that the recent studies demonstrate dilution of the mortality signal, but do not eradicate it completely. He presents a meta-regression of long-term risk ratio of all cause death against publication year and shows that the observed mortality risk decreases with publication year, which would seem to suggest earlier generation devices or clinical practices may be the cause for the mortality signal. He further subgroups the studies by paclitaxel dosage and shows that the relative mortality risks may be different among the different devices with variable paclitaxel dosing. He concludes that the reader must not miss the trees for the forest, meaning that the individual studies may have points of interest that become diluted in meta-analysis.
Commentary
The academic discussion around paclitaxel can make it difficult for providers to know if these devices are safe or not. This most recent meta-analysis is an excellent consolidation of the most recent literature on paclitaxel containing devices in femoropopliteal artery treatment. The results add to the body of evidence that these paclitaxel devices are safe and effective tools in peripheral arterial disease, especially at their current generations. Providers can use these devices and expect good outcomes for their patients without an increased mortality risk.
That being said, Dr. Katsonas does raise interesting questions. The 2018 Katsonas et al meta-analysis did show a significant increase in all-cause mortality. In his commentary, this effect appeared again when looking at all studies from 2008-2021. While there were shortcomings in the Katsonas paper, the fact remains that a mortality signal was present and hasn’t been fully accounted for. The fact that it seems to disappear when looking at studies from only 2018 onward could be indicative of the root cause. Moreover, the more recent studies did not subgroup based on device or paclitaxel dosing. The data he presents on the decrease of all-cause mortality risk over the years, as well as the trend of lower risk ratios with lower paclitaxel dosing, are suggestive that the mortality signal may be device-, device-generation, and/or paclitaxel dosage-dependent. Of course, this is only speculation based on available data, but could make excellent areas of inquiry should the question of safety in paclitaxel containing devices remain.
Dr. Katsonas’ final point is to not miss the “trees” (individual studies) for the “forest” (meta-analysis), as individual studies can yield insights that may be diluted in meta-analyses. However, he doesn’t really support this in his commentary, as most of his points are made by using grouping statistics based on multiple studies. Regardless, it is an important point to keep in mind when reading meta-analyses. Procedural practice and published literature are nuanced, and evidence-based medicine should be supported by both high-level meta-analyses as well as individual studies that befit the patient population and institutional experience.
Post Author
Sean Rogers, MD
Interventional Radiology Fellow, PGY-6
University of Massachusetts Chan Medical School
