[Safety evaluation of pedicle screw placement in proximal thoracic vertebrae via funnel technique]

To evaluate the safety of funnel technique for the placement of proximal thoracic pedicle screws (T1-T5).

Between March 2006 and October 2010, 105 patients undergoing pedicle screw placement of proximal thoracic vertebrae were retrospectively reviewed. There were 57 males and 48 females with an average age of 40.3 years old. According to the techniques of placing pedicle screws, they were classified into the funnel technique group (group A, n = 54) and bare hand technique group (group B, n = 51). Their perioperative and postoperative complications were examined simultaneously. Postoperative CT (computed tomography) scans were performed in all cases to determine the instrument malposition. The percentage of malposition was analyzed between two groups and among different levels.

A total of 402 proximal thoracic pedicle screws were placed in 105 cases. No injuries of spinal cord, large blood vessels or lung were reported. There were 13 intraoperative pedicle fractures and 4 dural tears. A total of 45 screws were misplaced in 105 patients (11.19%). In group A, the percentage of malposition was 10.14%. And there were no statistic significance among different levels (P > 0.05). In group B, the percentage of malposition was 12.31% and no statistic significance existed among different levels (P > 0.05). No intra-group statistical significance was found for group A or B. Medial malposition was found in 17 cases. And lateral malposition occurred in 25 screws while only 3 screws penetrated anterior cortex. Among 59 patients with neurological deficits, 37 cases showed postoperative improvements according to the AISA (American Spinal Injury Association) scale.

The funnel technique is both safe and reliable for the placements of proximal thoracic pedicle screws.