As published in the New England Journal the ALBIOS trial has already received a lot of attention and was also one of the hot topics at the ISICEM in Brussels last week. So what’s the story.
First there is the original article published in the recent NEJM. Gattinoni et al. have looked into the potential advantages in giving 20% albumin and cristalloid solutions to hypoalbuminic patients with severe sepsis compared to cristalloid solutions only. In the albumin group they aimed for a serum albumin concentration of 30g/L (a number hardly ever seen in any ICU patient) until discharge from the ICU or 28 days after randomization. 1818 patients were included and to make things short: The trial found no difference in 28-day mortality (primary outcome), 90-day mortality (secondary outcome), or any other relevant clinical endpoint (number of patients with organ dysfunction, degree of dysfunction, length of stay in ICU and the hospital).
So far for the original article. But now there is also this wonder-some supplementary appendix where the group has performed an unplanned subgroup analysis in septic shock patients only where a significant difference in the 90-day mortality was found – in favor for albumin.
Indeed there is a lower number of deaths in the albumin group, but not in the p value when adjusted for clinically relevant variables. It might also be interesting to note that the 90-day mortality was a secondary endpoint and no subgroup analysis was performed of 28-day mortality among patients with septic shock.
My personal view on this topic is that the original article and the supplemental appendix provide no good reason to favor albumin in the treatment of septic patients in general. The evidence provided to support the application of hypertonic albumin in septic shock patient is also not very convincing.
In regards of the SAFE study from 2004 the use of albumin will remain controversial and it will be interesting to see further trials upcoming in this field. It is also worthwhile remembering that albumin remains reasonably expensive and as Prof. Takala, Bern Switzerland, mentioned in Brussels last week: the money saved on avoiding unnecessary albumin infusions might be better invested in other ICU resourced proven to improve patient outcome. Comments?