Although the 30-day mortality data from EVAR Trial 1 and the Dutch Randomized Endovascular Aneurysm Management (DREAM) trial were similar, the conclusions drawn by the two groups of investigators differed dramatically.As reported in the previous issue of Vascular News, the 30-day mortality results of the UK EVAR 1 trial showed that endovascular aneurysm repair (EVAR) reduced death at 30 days by around two-thirds vs. open repair. The authors of the EVAR Trial 1, of which Roger Greenhalgh (Imperial College London, and Hammersmith Hospitals NHS Trust, UK) is Principal Investigator, emphasized, “These early results with EVAR, applied to large aneurysms in suitable patients, provide justification for continued use of this technique in controlled or trial settings; however, the early promise of endovascular repair cannot be guaranteed and might not endure in the long term. These 30-day mortality results are a license to continue scientific evaluation of EVAR, but not to change clinical practice.”
The results of the Dutch Randomized Endovascular Aneurysm Management (DREAM) trial, published in The New England Journal of Medicine (October 14 2004), also showed a lower operative mortality rate with the endovascular procedure at 30 days. However, the authors of the DREAM trial, which include Dr Jan Blankensteijn and Dr Monique Prinssen of the University Medical Center, Utrecht, the Netherlands, drew the following conclusion from their study: “On the basis of the overall results of this trial [DREAM], endovascular repair is preferable to open repair in patients who have an abdominal aortic aneurysm that is at least 5cm in diameter.”
The DREAM trialists later cover this statement by supporting the EVAR Trial 1 view, saying, “Long-term follow-up is needed to determine whether this advantage [for EVAR] is sustained.”Other important differences between EVAR Trial 1 and the DREAM trial are that the results of the DREAM trial did not reach statistical significance and that the DREAM trial under recruited, while EVAR Trial 1 over recruited. In the discussion in the DREAM paper, the authors clarify their findings by disclosing that the size of their study group was 12% lower than anticipated due to time restrictions imposed by the trial’s sponsor.
The DREAM trial found 30-day operative mortality was 4.6% in the open repair group (8 of 174 patients) and 1.2% in the endovascular repair group (2 of 171 patients). These percentages are similar to those found by the larger EVAR 1 Trial, which reported a 30-day operative mortality of 4.7% (24 of 516 patients) in the open repair group and 1.7% (9 of 531 patients) in the EVAR group.
The DREAM investigators make use of the EVAR Trial 1 data in their paper to bolster their own results: “Our trial and the EVAR 1 Trial are almost equivalent in terms of patient selection (patients with low surgical risk) and outcome criteria. Combining the results of the two trials yields the most accurate approximation of the risk ratio for in-hospital death to date: an operative mortality of 5.8% in the open repair group (40 of 690 patients) and 1.9% for the endovascular repair group (13 of 702 patients), resulting in a risk ratio of 3.1.”
In an accompanying editorial in The New England Journal of Medicine, Dr Frank Lederle, Professor of Medicine at the Minneapolis VA Medical Center and trial coordinator of the US Open Versus Endovascular Repair (OVER) trial, noted that the DREAM trial investigators observed reductions in 30-day mortality with EVAR as compared with open repair but that this “did not reach statistical significance owing to the relatively small sample size”. He continued by stating that he rejected the DREAM authors conclusion that endovascular repair is preferable to open repair in patients who have an abdominal aortic aneurysm that is at least 5cm in diameter because only the risk associated with repair is addressed, not the benefit. As he explains, “The more innocuous therapy is favored in comparison of procedural complications, even if the therapy is in-effective Just as we would not compare angioplasty and coronary bypass without considering subsequent cardiac events, we cannot compare open repair without evaluating the long-term risk of aneurysm rupture and graft complications.”
Commenting for Vascular News on the EVAR Trial 1, Lederle described this trial as “a great achievement”, but again emphasized the importance of awaiting the long-term results.
One-third of the EVAR1 patients will have been followed for four years by next June when EVAR 1 and EVAR 2 are due to report. DREAM is now also to be followed longer which is to be applauded.