16 The Journal of Thoracic and Cardiov Ricco and colleagues 1 are to be congratulated for their comprehensive reporting of the operative results of the French thoracic aortic stent grafting experience over the 2-year period from 1999 through 2001. This is the only report of its kind that depicts the results of an entire country’s experience. It was designed by the French National Health Insurance Fund, with the data verified by third-party advisors. The report is both timely and provocative in the wake of the US Food and Drug Administration’s recent approval of the use of the Gore TAG endograft for thoracic aortic aneurysms in an elective setting. A total of 166 thoracic aortic stent grafts were placed in 29 centers, with no center performing more than 15 cases, and most centers inserting 5 or less grafts in the 2-year period. Although presumably capturing all or most thoracic aortic stent-grafting cases performed in France during the time period, the report is limited in both its scope and its analysis. The most relevant limitations are the report’s retrospective nature and its inherent lack of follow-up data. Despite these flaws, the article reveals several important findings. Devices used included both industrial products (Gore [53%], Medtronic [32%], Stenford [4%], and Boston Scientific [0.5%]) and custom-made grafts designed by the implanting physicians (10%), a profile that is representative of trends around the world. The morbidity and mortality rates by device implanted or by physician specialty (ie, surgeon, radiologist, or cardiologist) are not provided. The overall mortality and major complication rates were significant at 10% and 29%, respectively (although 3 patients were not followed to 30 days). Type I or II endoleaks occurred in 16.3%, but among patients who had proximal descending thoracic aortic coverage (23 patients), 26% had a type I endoleak. Fifteen of the 23 had concomitant left subclavian artery transposition. All type I leaks required additional intervention either by means of open or endovascular techniques. The details of how and when these endoleaks were diagnosed are not provided. Two cases of proximal stent graft migration occurred, but it is not clear whether this migration occurred intraoperatively or postoperatively. None of the type II endoleaks (18 cases) were treated. Follow-up is not provided to demonstrate the natural history or prognosis of these type II leaks in this series. Type III endoleaks were not reported, although 3 of the type I leaks apparently occurred between 2 stent grafts. Paraplegia occurred in 6 (3.6%) patients. Although the anatomic detail and extent of aneurysm coverage is not provided for the patients who had paraplegia, 3 had concomitant abdominal aortic aneurysm repairs. Five of these 6 patients died. Neither spinal drainage nor perioperative evoked potential monitoring was used in any case, a practice that would seem prudent in high-risk cases. Chronic type B aortic dissection was present in 20% of the cases, and the mortality rate among this cohort was 15.2%. There were 17 cases of acute traumatic disruption treated with endografting, with a mortality rate of 6%. There were 3 cases of postoperative aortoesophageal or aortotracheal fistulae, a complication that is less common after conventional open repair. Endografts were placed in 3 patients with known septic aneurysms, and 2 of these patients died within 30 days. Finally, the authors appropriately highlight the fact that at least 17% of the aneurysms treated among the entire group were smaller than 50 mm, and the mortality rate among these cases was 5%. In summary, the report by Ricco and colleagues provides an unbiased glance of what the results of this new technology are likely to be when applied across a wide range of pathology, experience, and expertise. Despite the many advances in the
Read full abstract