Rib-based Distraction Device Implantation Before Age 3 Associated With Higher Unplanned Rate of Return to the Operating Room

Predicting early return to the operating room in early-onset scoliosis patients using machine learning techniques

PURPOSE

Surgical treatment of early-onset scoliosis (EOS) is associated with high rates of complications, often requiring unplanned return to the operating room (UPROR). The aim of this study was to create and validate a machine learning model to predict which EOS patients will go on to require an UPROR during their treatment course.

METHODS

A retrospective review was performed of all surgical EOS patients with at least 2 years follow-up. Patients were stratified based on whether they had experienced an UPROR. Ten machine learning algorithms were trained using tenfold cross-validation on an independent training set of patients. Model performance was evaluated on a separate testing set via their area under the receiver operating characteristic curve (AUC). Relative feature importance was calculated for the top-performing model.

RESULTS

257 patients were included in the study. 146 patients experienced at least one UPROR (57%). Five factors were identified as significant and included in model training: age at initial surgery, EOS etiology, initial construct type, and weight and height at initial surgery. The Gaussian naïve Bayes model demonstrated the best performance on the testing set (AUC: 0.79). Significant protective factors against experiencing an UPROR were weight at initial surgery, idiopathic etiology, initial definitive fusion construct, and height at initial surgery.

CONCLUSIONS

The Gaussian naïve Bayes machine learning algorithm demonstrated the best performance for predicting UPROR in EOS patients. Heavier, taller, idiopathic patients with initial definitive fusion constructs experienced UPROR less frequently. This model can be used to better quantify risk, optimize patient factors, and choose surgical constructs.

LEVEL OF EVIDENCE

Prognostic: III.

Changes in research quality and surgical trends at the international congress on early-onset scoliosis

PURPOSE

The founding of the International Congress for Early Onset Scoliosis (ICEOS) and first annual meeting in 2007 represented a significant milestone in advancing the care of patients with EOS. Due to the complexity and rarity of EOS, this annual conference is the premiere venue for physicians, researchers, and advanced practice providers to identify and understand the best treatments for children with EOS. This study examines the trend of various treatment modalities presented at ICEOS and the changes in research quality since its inception.

METHODS

Podium presentations from the 2007 through 2021 ICEOS annual meetings were reviewed to determine the number of study patients, use of a study group, and key features of study design. Treatment strategies being evaluated were recorded and included non-operative treatments (casting/bracing), traditional growing rods (TGR), vertical expandable prosthetic titanium rib (VEPTR), Shilla growth guidance, magnetically controlled growing rods (MCGR), and vertebral body tethering (VBT). Linear regressions were performed to analyze changes in research topic and study group utilization.

RESULTS

A total of 532 abstracts were reviewed. An average of 97.5 ± 81.3 patients were included per study with a significant increase from 42.3 ± 89.7 in 2007 to 337.6 ± 587.4 in 2021 (r² = 0.632, p < 0.001). A total of 130 (24.4%) abstracts resulted from multicenter study groups with the proportion increasing significantly from 13.0% in 2007 to 36.4% in 2021 (p = 0.039, r² = 0.289). The majority (96.2%) of study group-based projects were from either the Growing Spine Study Group (GSSG), Chest Wall and Spine Deformity Study Group (CWSDG), Children's Spine Study Group (CSSG), or the Pediatric Spine Study Group (PSSG). Additionally, a significant increase in studies utilizing patient-reported outcome measures (PROMs) was observed (r² = 0.336, p = 0.023). Significant increases in the proportion of presentations discussing MCGR (r² = 0.738, p < 0.001) and VBT (r² = 0.294, p = 0.037) as surgical treatments were observed.

CONCLUSION

The trends in EOS device implantation observed in registry studies align with the trends in research presented at ICEOS including the increased proportion of studies focusing on MCGR and VBT over the past decade. An attempt to increase the quality of research presented at ICEOS through multicenter study groups, increased patient recruitment, and utilization of PROMs has been seen since its inception.

LEVEL OF EVIDENCE

V.