The Accuracy of Computer-Assisted Pedicle Screw Placement in Degenerative Lumbrosacral Spine Using Single-Time, Paired Point Registration Alone Technique Combined with the Surgeon’s Experience
Prasert Iampreechakul MD*, Chana Chongchokdee MD*, Wuttipong Tirakotai MD*
Affiliation : * Department of Neurological Surgery, Prasat Neurological Institute, Bangkok, Thailand
Objective : Evaluate the accuracy of computer-assisted pedicle screw placement in patients with degenerative lumbrosacral
spine using single-time, paired point registration alone technique in combination with the surgeon’s experience.
Material and Method: A computer-assisted pedicle screw insertion in lumbrosacral spine were performed in 62 consecutive
patients (363 screws) using single-time, paired point registration without surface matching. After finding the entry point and
trajectory of the pedicle under image guidance, the surgeon then inserted pedicle screws by his experience. Postoperative
DynaCT scans were obtained and interpreted by two neuroradiologists. The Kappa statistic was used to measure the degree
of interobserver agreement. The screw position was graded as follows: Grade A = entirely within the pedicle; B = medial or
lateral pedicle wall breach less than 2 mm; C = medial or lateral pedicle wall breach equal to 2-4 mm; D = medial or lateral
wall breach more than 4 mm. Clinical outcomes including a numeric pain score, neurologic symptoms, and complications
were reviewed from all charts of patients. Additionally, the registration error, registration time, screwing time, and estimated
blood loss were analyzed.
Results : A total of the 363 pedicle screws, the first neuroradiologist interpreted grade A in 95.6%, grade B in 4.1% and grade
C in 0.3%, while the second neuroradiologist interpreted grade A in 95.3%, grade B in 3.6%, and grade C in 1.1%. There
was no incidence of grade D in this present study. No neurologic or vascular injuries occurred from pedicle screw placement.
The mean registration error was 1.54 + 1.28 (range, 0.9-2.5) mm with the mean time required for the registration process
for each patient was 3.64 + 1.92 (range, 2-8) minutes. The mean screwing time for each patient was 20.29 + 9.44 (range,
13-40) minutes. The mean pain score improved from 6.45 + 1.74 points preoperatively to 3.04 + 0.82 points postoperatively.
In the radiculopathy group, motor power gradually improved in all patients.
Conclusion : The use of single-time, paired point registration without surface mapping combined with surgeon’s perception
for computer-assisted pedicle screw insertion has proven to be a safe and effective technique in degenerative lumbrosacral
spine. The author’s technique did not rely solely on the image guidance. However, the navigation system would be very
beneficial to improve decision-making, especially in surgery of patients with distorted spinal anatomy. Besides, this technique
may decrease the overall operative time without compromise the accuracy of pedicle screw placement.
Keywords : Pedicle screw, Computer-assisted surgery, Navigation, Single-time registration, Lumbrosacral spine
