424 adult patients were randomised to a resuscitation protocol aimed at either normalising CRT vs. normalising or decreasing lactate levels.
The study was powered to detect an absolute reduction of 28-day mortality by 15%. Of course, this is crazy… nothing would be expected to work this well. As such, this study was always at a high risk of a type II error (i.e. not finding a difference when one truly exists). ¿Vamonos?
Out of interest, the CRT involved training and standardisation as follows
...CRT was measured by applying firm pressure to the ventral surface of the right index finger distal phalanx with a glass microscope slide. The pressure increased until the skin was blank and then maintained for 10 seconds. The time for return of the normal skin color was registered with a chronometer, and refill time greater than 3 seconds was defined as abnormal.
Once again, the primary outcome was all cause mortality at 28 days. There were lots of prespecified secondary outcome measures.
¿Resultados?
By day 28, a total of 35% in the CRT group vs 43 % in the lactate group had died (hazard ratio, 0.75 [95% IC, 0.55 to 1.02]; P=.06; risk difference, -8.5% [95% CI, -18.2% to 1.2%]
P value was 0.06. Therefore, a negative study…
This highlights the silly dichotomous nature that we attribute to P values. Less than 0.05 means “something important” whereas above 0.05 means nothing… totalmente loco amigo.
0.05 has always been a single arbitrary cut-off of probability. It tells us nothing about the effect size, whether a study was powered correctly or whether the outcomes are clinically important. It doesn’t tell us about bias, confounding, reverse causality or if the correct statistical analysis was used. The reverence we hold to P values must stop!
In the end, this study produces quite a bit of noise in favour of capillary refill time as a target for resuscitation. But perhaps it highlights the lack of utility of lactate. To be fair, we would never use a single marker in isolation. So perhaps this is all a bit silly… almost as silly as the title, ANDROMEDA-SHOCK trial.
¡Olé!
Covering: Hernandez G, Ospina-Tascon G Damiani LP, et al. Effect of a Resuscitation Strategy Targeting Peripheral Perfusion Status vs Serum Lactate Levels on 28-Day Mortality Among Patients with Septic Shock: The ANDROMEDA-SHOCK Randomized Clinical Trial. JAMA. 2019;321:654-64. [link to article]
Dr Brian Doyle is an emergency physician originally from the United States but now very much calls Tasmania his home. Unfortunately, it will now be a bit more difficult to deport him from the country as he passed his Australian citizenship test a few years ago. (He was able to answer that Phar Lap won the Melbourne rather than the Davis Cup). His main interests are mostly the clinical aspects of emergency medicine but also in education, ultrasound and critical appraisal of the literature. He spends much of his time annoying people to help out with conferences. |