BIS – magic prediction post arrest ?

A fascinating paper very well reviewed by Dr. James Dawson at the QMC journal club;

The bispectral index and suppression ratio are very early predictors of neurological outcome during therapeutic hypothermia after cardiac arrest Intensive Care Med. 2010 Feb;36(2):281-8. Epub 2009 Oct 22.

There is a pdf version of Dr. Dawson’s presentation in the blue box. Unfortunately for copyright reasons I’ve had to remove many of the slides documenting patient characteristics etc, although if you’re an ESICM member you can get the orginal paper on their site. Although there were some concerns raised about methodology this trial suggested the ROC curve for BIS for prediction of neurological outcome after cardiac arrest was 0.91 (which is pretty impressive). Points raised by those at the meeting were;

  • Muscle activity has profound influence upon BIS
  • No breakdown of who got propofol and who got lorazepam
  • Propofol ‘metabolised’ more slowly when patient hypothermic so may have had influence
  • Change in protocol during study
  • Accepted all dysrrhythmias to study, not just VT/VF group
  • Little standardisation in first few hours after arrest
  • What is the value in the findings ? Would we really withdraw patients with poor predicted outcomes when the results suggest we’d get this wrong 10% of the time?

1 comment so far

  1. danharvey on

    Although I wasn’t at the meeting (sorry to have missed it), I would add that determining adequate power for ROC curve generation is tricky. They presumably included all those who died in the negative outcome group as well James ?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: