708 - Evaluation of Techniques to Achieve Proper Vertebral Body Alignment in...

Oral Posters: MIS

Presented by: A. Kumar - View Audio/Video Presentation (Members Only)


A. Kumar(1), A. Su(2), V. Sundaram(2), A. Doshi(2), S. Qureshi(3)

(1) Louisiana State University, New Orleans, LA, United States
(2) Mount Sinai Medical Center, Department of Radiology, New York, NY, United States
(3) Mount Sinai Medical Center, Department of Orthopedic Surgery, New York, NY, United States


Introduction: Minimally invasive spine surgery relies heavily on fluoroscopic guidance. Rotational alignment of the vertebral body (VB) is typically assessed using two different techniques. The first involves assessment of the pedicles to ensure that they are symmetrical in size and proximity to the lateral wall of the VB. The second method evaluates the position of the spinous process (SP) to ensure it falls in the center of the lateral walls of the VB. The precision of these techniques may be confounded by anatomic variation and deformation of these structures. This paper aims to evaluate the reliability of these techniques by evaluating the symmetry of these structures on CT scan.

Materials and Methods: One hundred lumbar spine CT scans from patients complaining of back pain, without evidence of scoliosis or spondylolisthesis, were reviewed; a total of 500 lumbar vertebrae. Measurements included pedicle height and width, distance from the pedicle to the lateral wall of the VB and the angle of the SP measured from the perpendicular of a line running through the widest point of the VB. The length of the SP was also measured and these values allowed us to calculate the offset from midline of the tip of the SP as would be seen in an AP fluoroscopic image. Considering the measurement error of a Cobb angle is quoted at 5 degrees, we labeled SP angles less than this as relatively perpendicular to the VB.

Results: There was no statistical difference appreciated comparing bilateral pedicle height and width or pedicle-lateral wall distance for each VB from L1 to L5. The SP angle was found to deviate relatively equally to the right or left and this varied from level to level even within the same patient. Notably, the deviation of the SP was greater than 5 degrees in 14.4% of the vertebrae. In the L1 body, SP angle was greater than 5 degrees in 12% with the average deviation in these of 5.73 degrees, correlating to an offset of 3.1mm. L2 SP deviation was greater than 5 degrees only 4% of the time, averaging 6.05 degrees meaning an offset of 4mm. The L3 the SP deviated 16% of the time averaging 6.36 degrees meaning an offset of 3.7mm. The L4 SP also deviated 16% of the time, averaging 7.3 degrees with an offset of 3.7mm. The L5 SP deviated 24% of the time averaging 8.48 degrees meaning an offset of 4.9mm.

Conclusion: Our CT guided anatomical study of the lumbar vertebrae shows that pedicle dimensions and pedicle-lateral wall distance are symmetrical and therefore reliable measures of VB rotation. The novel finding is that the SP is not an accurate marker of midline as it may significantly deviate to one side 4-24% of the time by a margin of 3.7 to 4.9mm. This occurs more frequently as we move caudal in the lumbar spine. Using the SP to determine midline of the VB may lead to misalignment during localization, increasing the risk of medial or lateral breach of the pedicle wall.