Wednesday, June 12, 2013

Beware Observational Conclusions


"High achieved SBP after standardized antihypertensive therapy in hyperacute intracerebral hemorrhage was independently associated with poor clinical outcomes. Aggressive antihypertensive treatment may ameliorate clinical outcomes."

Jerry Hoffman has mentioned the exercise to, while reading, expand any instance of the word "may" to "may or may not" – to help erase the positive bias of speculative conclusions.  And, this is one of those perfect circumstances where the lukewarm endorsement from this abstract conclusion ought to be predicated with a pound of cautionary conditionality.

These authors call their study SAMURAI-ICH, and it's a prospective, observational study regarding the safety of early blood pressure reduction in intracerebral hemorrhage.  What this really means is they thought aggressive BP lowering was going to be awesome – despite only having various bits of inconclusive evidence – so they made wholesale practice changes, and then started a registry to monitor outcomes.  So, you can see the bias already.

And, verily, there is an association between their ability to lower blood pressure in ICH and favorable outcomes.  Now, their "favorable outcomes" cohort was also young, less disabled at baseline, and had smaller ICH hematoma volume.  Through the magic of statistical models, they attempt to control for all the various prognostic catastrophes, and thusly they arrive at their significant association.

But, finally, this observation doesn't in the slightest explain whether the blood pressure control improved outcomes – or whether it was simply easier to lower blood pressure in patients whose cerebrovascular physiology was less deranged by smaller insults, and who went on to have good outcomes.  Aggressive antihypertensive treatment "may or may not" ameliorate clinical outcomes, indeed.

Beware observational conclusions!

"Systolic Blood Pressure After Intravenous Antihypertensive Treatment and Clinical Outcomes in Hyperacute Intracerebral Hemorrhage : The Stroke Acute Management With Urgent Risk-Factor Assessment and Improvement-Intracerebral Hemorrhage Study"
www.ncbi.nlm.nih.gov/pubmed/23704107

No comments:

Post a Comment

Comments on posts > 10 days old will be moderated; blame spammers.