Author + information
- Jeffrey J. Cavendish, MD,
- William F. Penny, MD, FACC,
- Michael M. Madani, MD,
- Shahin Keramati, MD,
- Ori Ben-Yehuda, MD, FACC,
- Daniel G. Blanchard, MD, FACC,
- Ehtisham Mahmud, MD,
- Anthony Perricone, MD and
- Sotirios Tsimikas, MD, FACC ()
We agree that device size may account for some cases of anastomotic device failure, although these data became available only this month (February 2004), showing a 20.8%, 8.5%, 14.7%, and 0.0% failure rate in gray, green, blue, and purple sizes, respectively, although these were not statistically different from each other (1). In our report (2), mostly green or blue devices were used, but after appreciating that some patients were developing anastomotic device stenosis, only blue and purple were used. The quality of the veins and distal targets was felt to be adequate. In addition, all occlusions were at the proximal anastomoses, rendering poor distal runoff or technical problems with the distal anastomosis less likely. All patients received aspirin 325 mg and clopidogrel 75 mg on the day of the surgery. The aspirin was given indefinitely, and the clopidogrel was continued for two to three months.
Although the historical incidence of graft failure is higher than in our report, our study was a clinical and not an angiographic study. Thus, we did not evaluate the true incidence, which is likely underestimated. In addition to the studies pointed out in our study, a recent angiographic (three month) report shows that 11%, 5%, and 7% of grafts were either occluded, had >50% stenosis or <50% stenosis, respectively (19 of 81, or 23% saphenous vein graft [SVG]) (1). Of note, 18.6% of patients had recurrent cardiac symptoms, including three cardiac deaths. We have also identified an additional three patients and six SVGs with variable amounts of anastomotic device stenosis or occlusion in whom coronary artery bypass graft was performed in 2002, the time period reported in our article. Two patients were treated with sirolimus-eluting stents (2 SVGs and native left circumflex obtuse marginal coronary artery, respectively), and the other with medical therapy.
An important clinical consequence of anastomotic device stenosis is its aggressive nature, which appears diffuse and severe with involvement of all SVGs. This is in contrast to in-stent restenosis, which is more variable and only rarely leads to involvement of all stents. In this regard, patients with anastomotic device stenosis can present with global ischemic episodes that lead to myocardial infarction and death, which is also unusual with coronary in-stent restenosis. With the publication of additional reports describing anastomotic device stenosis (1,3), we would reemphasize increased oversight by regulatory agencies and adequately powered, randomized controlled trials of such devices, similar to coronary stent trials.
We believe that these devices will ultimately find their niche in clinical practice, as the imperative to reduce stroke during coronary artery bypass grafting is obviously critically important and has been reflected by the rapid acceptance of such devices by cardiovascular surgeons. However, improvements in the design of such devices seem to be required to improve outcomes (i.e., anti-proliferative drug coatings, improved loading and deliverability, and so forth). Pending further studies showing long-term efficacy, we continue to advise limiting their use to patients with a clearly increased risk of stroke during aortic cross-clamping. In patients in whom the device needs to be placed, the use of the largest devices, prolonged dual anti-platelet agents, careful follow-up, communication between cardiothoracic surgeons and cardiologists when the device is implanted, and ischemia testing in the first two to six months are warranted.The treatment of anastomotic device stenosis will continue to evolve as we learn more about this new iatrogenic disease.
- American College of Cardiology Foundation