Archive for the ‘JAMA’ Tag

Oxygen – too much of a good thing ?

We’ve known for sometime that high inspired oxygen fractions (>0.8) are associated with atelechasis (and decruitment / shunt) and lung injury and suspected that hyperoxia also leads to the generation of toxic oxygen free radicals, which may have deleterious effects in other pathologies. A recent paper adds to this story;

Association between arterial hypoxia following resuscitation from cardiac arrest and In Hopsital mortality. Hope, J et al. JAMA. 2010;303(21):2165-2171 (doi:10.1001/jama.2010.707).

This paper has been appraised by a colleague, Paul Townsley, and you can download his presentation from the blue box on the right. Although the methodology has some potential flaws (which is probably inevitable), the paper provides evidence of definite harm of hyperoxia in ischemic brain injury.

Of course, there is an alternative weight of evidence of the harm associated with even short periods of desaturation in traumatic brain injury (and presumably also brain injury from poor perfusion). How to square this circle? The devil as always is in the detail. Controlling hyperoxia is probably important, but whether the benefits of avoiding hyperoxia outweigh the additional risk of periods of desaturation, or how often those periods occur within patient populations, is unknown. There is sufficient evidence to warrant a carefully designed RCT in the post resuscitation phase, (and maybe also in ARDS while we’re at it).

Dexmedetomidine vs Midazolam

Dexmedetomidine vs Midazola for sedation of critically ill patients : A randomised control trial. Riker, R. et al. JAMA 301 (5) 489-99.

This prospective, double blind, multicentred RCT compared the alpha agonist dexmedotomdine against midazolam. It was designed to assess safety & efficacy. A total of 375 patients expected to be ventilated for > 24hrs were randomised to either agent targeted to a RASS score of between -2 to +1. Primary outcome measure was period of time within target range, secondary end points included delirium scores & duration of ventilation Continue reading

Tight Glucose Control – A Meta-analysis

Benefits & Risks of Tight Glucose Control in Critically Ill Adults; A Meta-Analysis. Weiner et al. JAMA. 2008;300(8):933-944.

This meta-analysis of 28 studies of “tight glucose control” on AICU inevitably includes papers studying different protocols on different patients and hence is limited in its applicability. I’m not sure that meta-analysis serves any purpose not bettered by a brief reading of the key studies of this concept, as there is significant heterogenity within the studies and the vast majority of the numbers come from a few well known trials. The failure of any study to replicate Van de Berge’s original 30% mortality reduction now begins to suggest that the benefit of tight glucose control is at best marginal and perhaps non-existent. There is of course a potential harm from the well documented increased risk of hypoglyceamia. The meta-analysis generates an RR of 0.93 (0.85-1.03) for tight glycaemic control. They could identify no stratified group that showed a consistent mortality benefit, including medical ICU patients (as in the original study).

The accompanying editorial is more useful and summarises the problems with investigating glucose control. Namely…

Continue reading