I liked your analysis. You make perfect sense as usual. However, I am
concerned about "referral" bias in numbers quoted from level 1
referral trauma centers. You alluded to some of it in your commentary on the
website too. The true positivity rate (sensitivity) will be enhanced and the
false positive rate will be lower when the prevalence of the disease is
higher. But, from the same referral centers, if we stratify the population
up-front, into those with a "low" and "high" pre-test probability of
c-spine injuries and then evaluate the true positive rates as well as NPV of
a lateral c spine, one may find that the likelihood of missing a clinically
significant injury is exceedingly low.
My underlying concern is still the same, as I stated earliar.Very often, in
these pre-verbal children, odontoid and AP radiographs end up being
inadequate. What do you do? Keep trying to get a better view?. Obtain a CT?
Get a submental view/obliques? Fall back on your pre-test probability and
play the odds? I very often find myself following the last path.
If they are not lucid, one really does not care too much about clearance of
cspine anyway. They have to have continued immobilization regardless!.
Jay Pershad, MD
For more information, send mail to [log in to unmask] with the message: info PED-EM-L
The URL for the PED-EM-L Web Page is: