S2 Alar Screw Insertion Accuracy and Factors Associated With Screw Loosening and Lumbosacral Nonunion.

OBJECTIVE:To investigate S2 alar screw (S2AS) accuracy and factors associated with S2AS loosening and lumbosacral nonunion. METHODS:We retrospectively reviewed patients who underwent lumbosacral fusion surgery with S2AS addition under fluoroscopy. S2AS loosening and lumbosacral nonunion were analyzed using a 1-year postoperative computed tomography. S2AS insertion accuracy was originally classified as accurate, short, anterior perforation, lateral perforation, and sacroiliac joint (SIJ) deviation among lateral perforation. Clinical data including sex, age, body mass index, fused segments, fusion procedure, primary or revision surgery, Japanese Orthopedic Association scores and complications were collected. Factors associated with S2AS loosening and lumbosacral nonunion were analyzed. RESULTS:A total of 37 patients (74 screws, age: 63.78 ± 13.57 years, female/male: 14/23 patients, body mass index: 23.11 ± 2.53, fused segments: 1-4 levels, revision: 38%) were included. S2AS loosening and lumbosacral nonunion were observed in 18 screws (13%) and 8 patients (22%) respectively. Only 35 screws (47%) were inserted accurately in our classification. Short, lateral perforation, and anterior perforation were observed in 14 screws (19%), 22 screws (30%), and 3 screws (4.1%). SIJ deviation was seen in 15 screws (20%) Factors associated with S2AS loosening were older age (P = 0.038), fusion levels (P = 0.011), and SIJ deviation (P < 0.001). S2AS loosening affects S1 pedicle screw (S1PS) loosening (P = 0.001). Furthermore, S2AS loosening is a risk factor for lumbosacral nonunion (P = 0.046). CONCLUSIONS:S2AS insertion under fluoroscopy is inaccurate. S2AS loosening induces S1PS loosening and lumbosacral nonunion. Surgeons should avoid deviating to SIJ, especially in older patients and relatively longer fusion.