Commentary on ‘External Validation of Models Predicting Survival After Ruptured Abdominal Aortic Aneurysm Repair’

Abstract

The decision to palliate, that is to withhold potentially effective treatment, is justified by the belief that the treatment is of no benefit and that it would carry excess mortality risk. Nevertheless, concerns remain over when, why, and whether palliating patients with abdominal aortic aneurysm (AAA)makes sense, as the decision is still largely based on the subjective interpretation of harsh reality. Critical appraisal of sound evidence suggests that there are no reliable or valid criteria to establish whether to treat an aneurysm in either the elective or emergency situation. It has been suggested that there will always be some patients at “highest risk,” “unsuitable for intervention,” in whom the risk of elective large AAA repair will outweigh potential survival benefits. Given that the likelihood of death from non-aneurysm related causes is greater than that from the aneurysm itself, and that life expectancy would not increase despite intervention, palliation should be offered to these patients. However, how the “highest risk”or “unfitness” profile should be objectively and reliably measured, to withhold elective aneurysm repair, remains debatable and not standardized. Even more challenging is the decision to turn down patients with ruptured AAA (rAAA) as suggested in the paper by the Amsterdam Acute Aneurysm Trial group (AIAX trial) published in this issue of European Journal of Vascular and Endovascular Surgery. Authors tested a number of well known, already developed, predictive models, some refined for patients with rAAA, to identify subgroups at high mortality risk, based on a variety of demographics, comorbidity, and severity conditions (e.g. Hardman index, Vancouver scoring system, updated Glasgow Aneurysm Score [GAS], and Edinburgh Ruptured Aneurysm Score). Nevertheless, the study showed only limited and insufficient performance of all the current prognostic models when applied to 449 real patients with rAAA treated in 10 Netherlands hospitals in modern times (after 2004). Onemain finding derived from this validation study was the general tendency to overestimate the true death rates according to predicted values, and also using the updated GAS model shown to be the most reliable and accurate in prediction of death (60% in GAS vs. 54% observed death rate).