Ten Years Ago to Present: Assessing Trends in Outcomes in a Single-Surgeon Cohort of Adult Spinal Deformity Patients

Background: Adult spinal deformity (ASD) has been shown to compromise health-related quality of life. While advances in spinal realignment have shown promising short-term and even two-year clinical results in this parameter, the longer-term outcomes of ASD over the 10Y period post-corrective surgery remain unclear.

Purpose: To describe the 5 to 10-year outcomes for patients undergoing surgical correction of cervical deformity.

Methods: Inclusion criteria were operative ASD patients (coronal Cobb angle≥20°, SVA≥50mm, PT≥25°, and/or thoracic kyphosis>60°) >18yrs with complete baseline (BL) and up to 10-year (10Y) radiographic/HRQL data. Patients were stratified after ranking by date of initial surgery into tertiles in as Early (1st tertile) vs Late (3rd tertile). Student’s t-test, paired t-test, linear regression, and binary and multivariable logistic regression analyses were used to evaluate the clinical, radiographic, and complications outcomes between tertiles as well as assess linear trends up to 10Y. Kaplan-Meier (KM) survival analysis and Cox regression accounting for age and frailty at presentation was further utilized.

Results: 82 patients with adult spinal deformity were enrolled from 2012 to 2017 (54.5 ± 10.3 years, 57% female, BMI 30.0 ± 7.3 kg/m2). Over the 10Y period, linear regression revealed that patients were more likely to be older (p=.049), and were more likely to be female (p=.036). In terms of baseline HRQL measures assessed, Late patients were only more likely to present with lower NRS-Arm scores compared to their Early counterparts (1.68 vs 2.78, p=.002). Surgically, Late patients were significantly more to be operated on using robotic or navigational assistance, or through minimally invasive technique (all p<.001). Binary analysis revealed that Late patients had significantly lower mean levels fused (p=.002), and on average, Late patients received greater values of cell saver than Early patients (p <.001). In terms of post-operative complications, Early patients were more likely to return to the OR within 30 days (p=.048), and experienced more readmissions within 90 days (p <.001), and overall within 10Y (p <.001). Even when controlling for age, gender, and BMI at presentation, backstep regression analysis demonstrated that belonging to the Late cohort was associated with lower risk of reoperation, or necessity of secondary cervical spine surgery after initial lumbar surgery (both p<.05). Mean 10Y survivorship was 74.7%, with KM analysis revealing that mean survival time for deceased ASD was 84.54 weeks for early patients and 99.17 weeks in Late patients (χ2(1)=.748, p=.016). However, Cox regression adjusting for age and frailty revealed no significant differences in mortality between groups (p>.05).

Conclusion: When assessing patient-reported, operative, and complications outcomes in an adult spinal deformity cohort with up to ten-year follow-up, complications in patients operated on more recently demonstrated lower complication, reoperation, and readmission rates compared to their counterparts. These findings may be due to the significantly increased use of minimally-invasive technique, navigational guidance, or robotic assistance observed in this cohort. However, the findings of this study necessitate additional investigation into the modifiable vs non-modifiable factors that have contributed to safer and more durable surgeries over the past decade.