Morphometric comparison of the pedicle rib unit to pedicles in the thoracic spine

A radiographic study of the pedicle rib unit morphology as compared with measurements of the pedicle in cadaveric thoracic spines.

To evaluate the morphology of pedicle rib units in the thoracic spine in normal thoracic human spines, with particular attention to T4-T9, and to compare the dimensions of the pedicle rib unit with corresponding dimensions of the adjacent pedicles.

Despite the clinical successes reported with thoracic pedicle screw-rod constructs, controversy exists as to the safety of this technique in pedicles that are anatomically too small for transpedicular fixation. An alternative method of extrapedicular screw fixation within the pedicle rib unit was evaluated in a previous study and found to be anatomically feasible. Although the morphology of the pedicle rib unit was described in two previous studies, the measurements were obtained in scoliotic spines. Therefore, no study has sought to define the transverse dimension and chord length of the pedicle rib unit, and compare it with the corresponding pedicle dimensions.

Six fresh unfixed adult cadavers were obtained randomly. No history of spine disease was noted, and cause of death was unrelated to spinal disorder or trauma. The mean age was 84, with a range of 76 to 90. There were two females and four males. Computerized tomographic (CT) images of the thoracic spine were obtained. For cadavers in which the gantry resulted in oblique axial sections, reformatting was performed for more accurate measurements. Measurements of the transverse diameter and chord length of the pedicle rib unit were obtained and compared with measurements of the transverse pedicle width and chord length.

The transverse width and chord length of the pedicle rib unit were significantly larger than corresponding pedicle measurements at all levels, a consistent finding when comparing the mean of all levels, the mean of T4-T9, and the mean of each individual level T4-T7.

This study confirms the marked difference in size of the pedicle rib unit as compared with the pediclein both transverse width and chord length. This allows for a space that accommodates much larger major screw diameters, longer screw lengths, and because of the nature of screw placement, a greater screw convergence. Thus, there is anatomic potential for extrapedicular vertebral body fixation in the thoracic spine. However, care must be taken in placement of screws following precisely our previously described method of extrapedicular screw insertion. Penetration of surrounding structures, most notably the aorta on the left, is a potential risk when deviating from the method. Biomechanical evaluations are presently being conducted to evaluate the use of extrapedicular thoracic screw fixation.