Comment on “A Randomised Trial of Endovascular and Open Surgery for Ruptured Abdominal Aortic Aneurysm. Results of a Pilot Study and Lessons Learned for Future Studies” by Hinchliffe RJ, Bruijstens L, MacSweeney STR, Braithewaite BD

Article Outline

• Reference
• Copyright

The majority of articles on emergency EVAR report the experience in patients selected on a combination of favourable aspects, such as a suitable arterial morphologic configuration to allow easy access and optimal sealing of the stent-graft, the presence of an experienced endovascular team and often also a hemodynamically stable individual. This is not the typical patient presenting with a ruptured abdominal aortic aneurysm and these papers do not provide a realistic outcome of endovascular management of this serious condition. In recognition of this problem, the present authors have made a commendable effort to systematically analyse the outcome of emergency EVAR in all-comers with a ruptured AAA. Their study design was carefully chosen. The patient had to be sufficiently fit to tolerate an open repair, and the randomisation was made at presentation of the patient at hospital, but before CT-scanning. In addition CT-scanning, according to the study protocol, could be withheld depending of the hemodynamic condition of the patient. Two-third of the patients could not be randomised because consent to participate in the study was not given, an endovascular team was not available, or because of severe hemodynamic instability precluding CT-scanning. However, these aspects represent a random phenomenon, not leading to a selection bias, i.e. did not result in a systematic inclusion of patients with a better or worse prognosis in the study group compared to the entire group of 103 patients presenting during the study period.

Most of the important aspects of endovascular management, such as the use of aorto-uni-iliac endografts and local anaesthesia during the first part of the procedure were accurately described. The frequent observation of shorter, wider and more angulated infrarenal necks in ruptured compared to electively treated aneurysms was emphasised. The issue of severely aneurysmatic common iliac arteries, frequently complicating the endovascular procedure, was not referred to.

Mortality figures were reported by intention-to-treat and the proportion of deaths was understandably much higher in the group treated by endovascular technique than in any of the selected case series referred to by the authors. However, the mortality rates in the endovascular and open treatment patients who actually underwent repair of their aneurysm, could be compared to a recently published multicenter cohort series in which patients were preferentially (if possible) treated by emergency EVAR. The mortality rates in the EVAR-group was 11% lower, and in the open repair group 4% lower than in the currently published series, which may be due to common inter-series variation and to differences in design of the two studies.¹

The present study provides a realistic appraisal of the outcome of emergency EVAR and a reasonable comparison with open repair. Nevertheless, the important question one could ask the authors is why the study was prematurely terminated? A randomised controlled study (RCT) is the most powerful scientific instrument available for clinical assessment. Stopping an RCT before the predetermined number of patients is recruited, or before the trial safety committee observes one of the tested treatments associated with a significantly worse outcome, is against the principles of proper conduct of clinical studies. A pilot study and a randomised controlled trial are by definition entirely different assessments and the two types cannot be combined. The fact that the authors did not observe the expected difference between study-arms after enrolment of one-third (only 32 patients) of the calculated 100 patients needed for an accurate conclusion, cannot justify the termination of a randomised study. By doing so, the study organizers precluded the observation of trends during later stages of the study. Readers may expect from published articles that carry the flag of an RCT in their title, to receive at least level II scientific evidence about the efficacy of a treatment. The present study has demonstrated neither equality or superiority of one of both treatments. The real RCT on ruptured AAA-treatment still has to be undertaken.

Back to Article Outline

Reference 

1. Peppelenbosch N, Geelkerken RH, Soong C, Cao P, Steinmetz OK, Teijink JAW, et al. Endograft treatment of ruptured abdominal aortic aneurysms using the Talent aortouniiliac system. An international multicenter study. J Vasc Surg. 2006;43:111–122
   • View In Article