Monday, November 4, 2013

The Trauma Log Roll is Dead

Among unproven interventions, back-boarding, cervical collars, and log-rolling have been part of the dogma of trauma since Alfred Nobel invented the electric slide.  We've finally started to put an end to uncomfortable and unwarranted back-boarding, we've re-designed cervical collars, and this article takes on log-rolling.  The assertion of these authors is clinical examination of an otherwise alert major trauma patient is unreliable, does not obviate imaging, and may thereby be omitted from the initial secondary survey.

Unfortunately, this is a very specific, limited, retrospective registry review.  Only patients from the trauma registry at the Alfred Hospital were included: major trauma (ISS >15) and admitted for 24 hours, or isolated thoracolumbar injuries requiring 72 hours of hospitalization.  This identified 1,161 patients with thoracolumbar fractures, and these authors further pared it down to 538 who were GCS >15 as their proxy for potentially reliable examination.  How many of these alert, appropriate trauma patients with thoracolumbar fractures complained of pain on log-roll and spinal palpation?


So, yes, if the clinical examination is only 60.3% sensitive for significant thoracolumbar fractures, then we ought to stop bothering to log-roll our patients.  But, generalizing the evidence from this retrospective review in a highly selected population is grossly irresponsible.  It is reasonable, as the accompanying letter states, if the decision has already been made in a major trauma to progress to full-body computed tomography – a test more sensitive and specific for spinal fractures than clinical examination – log-roll and complete physical examination may be deferred.  The theoretical risks to log-roll – lack of true thoracolumbar stability, possibility of disturbing internal hemostasis – if there is no benefit, are appropriate considerations if physical examination does not change clinical evaluation.  It is, however, excessive to universally posit, as the letter authors do, "Log-rolling a blunt major trauma patient is inappropriate in the primary survey."

"Can initial clinical assessment exclude thoracolumbar vertebral injury?"‎

"Log-rolling a blunt major trauma patient is inappropriate in the primary survey"


  1. Dr. Radecki, there appears to be some hidden cruft at the end of the first PubMed link keeping it from loading (hidden URL encoded characters %E2%80%8E).

  2. The cruft has been decrufted.

    I mean, oh no, you've discovered my hidden NSA trojan horse tracking code!

  3. Initiating download of all your files and


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