During assessment after injury, the log roll examination, in particular palpation of the thoracolumbar spine, has low sensitivity for detecting spinal injury. The manoeuvre itself requires a pause during trauma resuscitation. The aim of this study was to assess the utility of the log roll examination in unconscious trauma patients for the diagnosis of soft tissue and thoracolumbar spine injuries.Methods
A retrospective cohort study was undertaken, reviewing the cases of unconscious (Glasgow Coma Scale (GCS) <9) and/or intubated major trauma (Injury Severity Scale (ISS) >12, abbreviated injury scale 2008) patients from the Alfred Trauma Registry, over a 2-year period from January 2011 to December 2012. Log roll examination findings, as documented in the medical record, were compared with CT reports. Out of the 624 screened records, 222 (35.6%) were excluded as the log roll or CT/MRI had not been performed.Results
There were a total of 2028 major trauma presentations to the Alfred Hospital Emergency and Trauma Centre during the study period. Excluded cases comprised 147 patients who did not have a documented log roll, and 75 patients who did not have a CT or MRI. Of the 402 cases that met inclusion criteria, 35.3% had a thoracolumbar fracture, and the sensitivity of log roll examination was found to be 27.5%, with a specificity of 91%. The negative likelihood ratio for abnormalities on log roll was low (0.8).Conclusions
Examination of the back in unconscious trauma patients could be limited to visual inspection only to allow identification of penetrating wounds and other soft tissue injuries (including of the posterior scalp) and removal of foreign bodies, in patients planned for CT scans. The low sensitivity and poor negative likelihood ratio suggest that a normal log roll examination does not accurately predict the absence of bony injury to the thoracolumbar spine.