To date, evidence suggests that surgical treatment can improve pain and disability in adults with symptomatic scoliosis. However, these studies are based on averages across large groups of patients, and not all surgically treated patients achieve average or above-average outcomes. The reasons that some adults with scoliosis markedly improve with surgery, while others fail to improve, are not clear. Comparison of patients at extremes of outcome measures might provide insights into factors affecting outcomes. The objective of this prospective study presented by Justin S. Smith, MD, PhD, University of Virginia Health System, Charlottesville, Virginia, USA, was to compare best vs worst clinical outcomes in patients with adult spinal deformity (ASD) following surgical treatment.

The patients (n = 227) were recruited from the International Spine Study Group (ISSG) multicenter database for ASD. At baseline and follow-up, the patients completed health-related quality-of-life measures, including the Scoliosis Research Society-22 (SRS-22) questionnaire, Oswestry Disability Index (ODI), SF-36 physical and mental components, and back and leg pain numeric ratings. Included patients had a primary diagnosis of ASD (aged > 18 years) and were part of the ISSG operative treatment group.

At 2-year postoperative follow-up, the patients had significant improvements from baseline in mean leg pain score (4.2 vs 2.4; P < .001), mean back pain score (7.0 vs 2.3; P < .001), mean ODI (41 vs 25; P < .001), and mean SRS-22 (2.9 vs 3.7; P < .001). Other improvements included mean coronal balance, sagittal balance, maximum Cobb angle, pelvic tilt, and pelvic incidence-lumbar lordosis (PI-LL) mismatch (P < .001 for all).

Based on the ODI, the best 2-year outcome was defined as ODI ≤ 15 (n = 43) and the worst, ODI ≥ 50 (n = 51). Based on the SRS-22, the worst outcome was defined as SRS-22 ≤ 2.5 (n = 27) and the best, SRS-22 ≥ 4.5 (n = 25).

On univariate analysis, baseline factors distinguishing the best and worst outcomes based on the ODI were diagnosis of depression (P = .028), back pain score (P = .003), body mass index (BMI; P = .002), sagittal vertical access (SVA) > 5 cm (P = .009), and occurrence of ≥ 1 major complications (P = .001). Distinguishing factors at follow-up were leg pain score (P < .001), back pain score (P < .001), PI-LL mismatch (P = .042), and SVA > 5 cm (P = .062).

The baseline factors distinguishing the best and worst outcomes based on the SRS-22 were diagnosis of depression (P < .001), back pain score (P = .006), prior spine surgery (P = .007), American Society of Anesthesiologists grade (P = .004), and occurrence of ≥ 1 minor or major complication (P = .034). Distinguishing factors at follow-up were leg pain score (P < .001) and back pain score (P < .001).

On multivariate analysis, distinguishing factors for best and worst outcomes based on ODI were baseline BMI, ODI, and follow-up SVA. Distinguishing factors for best and worst outcomes based on SRS-22 scores were baseline depression, minor or major complication, and baseline SRS-22 (Table 1).

Table 2 shows the factors distinguishing the best and worst outcomes in the ISSG database as compared with the Spinal Deformity Study Group (SDSG) database.

Factors distinguishing best and worst outcomes of surgery for ASD were primarily patient related, such as obesity, depression, pain severity, and comorbidities, based on the ISSG and SDSG. However, the ISSG data also suggest that in addition to patient factors, residual spinopelvic malalignment (PI-LL mismatch) and occurrence of major complications are distinguishing factors.