Regarding AIS, the Lenke classification remains the standard.¹ Comparative studies determined the Lenke classification to have good–excellent (kappa >0.75) interobserver and intraobserver reliability for curve type (kappa 0.92–0.83), lumbar modifiers (kappa 0.80–0.84), and sagittal thoracic modifiers (kappa 0.94–0.97), respectively, revealing significant improvements from the previously universally accepted King–Moe classification.²–⁴ Further, a retrospective study investigating 606 AIS patients determined that 90% of operative cases had surgically recognized structural regions of the spine predicted by the Lenke curve type.⁵ While the Lenke classification offers comprehensive radiographic evaluation, improved decision‐making, and good‐excellent reliability, retrospective and comparative studies have suggested that its complexity (42 curvature types) and inability to define end construct levels hinder feasibility for surgical planning.⁶,⁷

Despite controversy over operative decision‐making (31% prevalence),⁸ variations in operative management have reduced (18% vs 12%, p = 0.001) since the Lenke classifications implementation.⁹ According to the Lenke classification, major curves should be included in the fusion construct, along with the structural minor curves, leaving nonstructural minor curves to spontaneously correct.¹⁰ Selective fusion involves fusing one of two curves that crosses the midline, and the lower instrumented vertebra (LIV) extending no lower than L2. A meta‐analysis determined that for Lenke 1C curves, selective thoracic fusions decreased postoperative main thoracic Cobb angle (mean difference [MD]: −27.78° [−30.71° to −24.85°]; p <0.01), postoperative thoracolumbar/lumbar Cobb angle (MD: −16.24° [−17.99° to −14.48°]; p <0.01), and improved coronal balance (MD: 0.47 cm [0.07–0.87]; p = 0.02).¹¹ While selective fusion maximizes postoperative mobility, corona ‐decompensation is a concern. Nonselective thoracic fusion, where LIV may extend past L2, sacrifices mobility to more definitively prevent coronal decompensation. Regarding upper instrumented vertebra (UIV) determination, the primary objective is to minimize shoulder imbalance and prevent proximal junctional kyphosis (PJK) development. UIV caudal to proximal upper‐end vertebra (Lenke 1; odds ratio [OR] = 15.91 [2.18–115.95]; p = 0.0063) and cephalad to upper end vertebra (Lenke 5, OR = 9.07 [1.77–46.45]; p = 0.0081) were significant PJK risk factors.¹²

After wide adaptation of the Lenke classification, the Scoliosis Research Society (SRS) classification was established to guide categorization and management of ASD. A comparative study determined the SRS classification to have moderate interobserver (kappa: 0.64), regional sagittal modifier (kappa: 0.73), and degenerative lumbar modifier (kappa: 0.65) classification reliability. Interobserver reliability for determining UIV (kappa: 0.56) and LIV (kappa: 0.77) was substantial.¹³ While this system describes structural curvature, it overlooks other important clinical factors relevant to ASD decision‐making (e.g. age, BMI, disability). Building off of the SRS classification, Schwab and colleagues created a system which guides operative decision‐making by identifying clinically significant radiographic parameters associated with patient‐reported outcome measures (PROMs).¹⁴ Comparative studies have determined the SRS Schwab classification to have good–excellent interobserver (kappa: 0.73–0.87) and intraobserver (kappa: 0.83–0.94) reliability.¹⁵,¹⁶ The most clinically significant SRS Schwab radiographic parameters are pelvic incidence minus lumbar lordosis (PI‐LL), pelvic tilt (PT), sagittal vertical axis (SVA), and T1 pelvic angle (TPA).¹⁷,¹⁸ A prospective comparative study determined PT, SVA, and PI‐LL to correlate most strongly with disability in operative and nonoperative cohorts, with PT ≥22° (r = 0.38), SVA ≥47 mm (r = 0.47), or PI‐LL ≥11° (r = 0.45) most strongly predictive of ODI >40 (indicative of severe disability).¹⁹ Correlations between age and PROMs also exist (r >0.510, p <0.001), with younger patients requiring more rigorous alignment objectives.²⁰ Prospective studies have validated the use of SRS Schwab modifiers. Patients with improved PT, PI‐LL, SVA, and TPA were associated with improved SRS 22 (total, pain, activity, appearance), ODI, and physical component scale (PCS) (p <0.05).²¹,²² SRS Schwab modifiers also correlated with operative decision‐making. A prospective study of 527 consecutive patients determined patients with abnormal sagittal spinopelvic modifiers required major osteotomies, iliac fixation, interbody fusions, and/or decompression procedures (p <0.001).²³