Pedicle screw placement accuracy impact and comparison between grading systems.
Marios T, et al. Surg Neurol Int. 2017.
Show full citation
BACKGROUND: Pedicle screw instrumentation is widely used for spinal stabilization. However, the accuracy for free-hand screw placement ranges from 69% to 94%. This study assesses the value of the existing classification systems, and investigates their impact on the ability to assess the accuracy of free-hand screw placement.
METHODS: Data were collected retrospectively from the medical records of 34 patients who received 224 pedicle screws placed utilizing a free-hand technique. Screw placement was evaluated employing the 2-mm increment and Zdichavsky et al. classification systems. Kappa coefficient and Landis and Koch interpretations were employed for statistical analysis.
RESULTS: The 2-mm increment classification system resulted in a total of 18 (8.03%) misplaced screws. Lateral screw misplacement was observed in 13 (5.8%) instances, with medial pedicle wall penetration being noted in 5 (2.23%). Of the 18 misplaced screws, 4 (22.22%) were classified as minor (≤2 mm), 12 (66.67%) as moderate (2-4 mm), and 2 (11.11%) as severe (>4 mm) (K = 0.882). The Zdichavsky et al. grading system categorized 208 (92.84%) pedicle screws as Ia, 10 (4.46%) as Ib, 1 (0.45%) as IIa, 2 (0.90%) as IIb, 2 (0.90%) as IIIa, and 1 (0.45%) as IIIb grade; this resulted in a total of 16 (7.14%) misplaced screws (K = 0.980). One patient exhibited a new postoperative radiculopathy attributed to poor screw placement. There were no additional early or late postsurgical complications attributed to screw misplacement.
CONCLUSION: The free-hand pedicle screw placement technique is both safe and effective. Postoperative computed tomography studies; however, are useful to confirm the accuracy of screw placement. Although, the available grading systems proved reliable, easy to use, and clearly reflected the individual surgeon's skills, they do not clearly document whether screws are safely placed.
PMID 28713634 [PubMed] PMCID PMC5502290