OBJECTIVETo identify the biomechanical feasibility of the thoracic extrapedicular approach to the placement of screws.

METHODSFive fresh adult cadaveric thoracic spine from T1 to T8 were harvested. The screw was inserted either by pedicular approach or extrapedicular approach. The result was observed and the pullout strength by pedicular screw approach and extrapedicular screw approach via sagittal axis of the vertebrale was measured and compared statistically.

RESULTSIn thoracic pedicular approach, the pullout strength of pedicle screw was 1001.23 N+/-220 N (288.2-1561.7 N)ls and that of thoracic extrapedicular screw approach was 827.01 N+/-260 N when screw was inserted into the vertebrae through transverse process, and 954.25 N+/-254 N when screw was inserted into the vertebrae through the lateral cortex of the pedicle. Compared with pedicular group, the pullout strength in extrapedicular group was decreased by 4.7% inserted through transverse process (P larger than 0.05) and by 17.3% inserted through the lateral cortex (P less than 0.05). The mean pullout strength by extrapedicular approach was decreased by 11.04% as compared with pedicular approach (P less than 0.05).

CONCLUSIONSIt is feasible biomechanically to use extrapedicular screw technique to insert pedicular screws in the thoracic spine when it is hard to insert by pedicular approach.