Deep Learning Method for Precise Landmark Identification and Structural Assessment of Whole-Spine Radiographs.

This study measured parameters automatically by marking the point for measuring each parameter on whole-spine radiographs. Between January 2020 and December 2021, 1017 sequential lateral whole-spine radiographs were retrospectively obtained. Of these, 819 and 198 were used for training and testing the performance of the landmark detection model, respectively. To objectively evaluate the program's performance, 690 whole-spine radiographs from four other institutions were used for external validation. The combined dataset comprised radiographs from 857 female and 850 male patients (average age 42.2 ± 27.3 years; range 20-85 years). The landmark localizer showed the highest accuracy in identifying cervical landmarks (median error 1.5-2.4 mm), followed by lumbosacral landmarks (median error 2.1-3.0 mm). However, thoracic landmarks displayed larger localization errors (median 2.4-4.3 mm), indicating slightly reduced precision compared with the cervical and lumbosacral regions. The agreement between the deep learning model and two experts was good to excellent, with intraclass correlation coefficient values >0.88. The deep learning model also performed well on the external validation set. There were no statistical differences between datasets in all parameters, suggesting that the performance of the artificial intelligence model created was excellent. The proposed automatic alignment analysis system identified anatomical landmarks and positions of the spine with high precision and generated various radiograph imaging parameters that had a good correlation with manual measurements.

The iterative implementation of a comprehensive enhanced recovery after surgery protocol in all spinal surgery in Korea: a comparative analysis of clinical outcomes and medical costs between primary spinal tumors and degenerative spinal diseases.

OBJECTIVE: Most studies on the enhanced recovery after surgery (ERAS) protocol in spine surgery have focused on patients with degenerative spinal diseases (DSDs), resulting in a lack of evidence for a comprehensive ERAS protocol applicable to patients with primary spine tumors (PSTs) and other spinal diseases. The authors had developed and gradually adopted components of the comprehensive ERAS protocol for all spine surgical procedures from 2003 to 2011, and then the current ERAS protocol was fully implemented in 2012. This study aimed to evaluate the impact and the applicability of the comprehensive ERAS protocol across all spine surgical procedures and to compare outcomes between the PST and DSD groups. METHODS: Adult spine surgical procedures were conducted from 2003 to 2021 at the Seoul National University Hospital Spine Center and data were retrospectively reviewed. The author divided the study periods into the developing ERAS (2003-2011) and post-current ERAS (2012-2021) periods, and outcomes were compared between the two periods. Surgical procedures for metastatic cancer, infection, and trauma were excluded. Interrupted time series analysis (ITSA) was used to assess the impact of the ERAS protocol on medical costs and clinical outcomes, including length of stay (LOS) and rates of 30-day readmission, reoperation, and surgical site infection (SSI). Subgroup analyses were conducted on the PST and DSD groups in terms of LOS and medical costs. RESULTS: The study included 7143 surgical procedures, comprising 1494 for PSTs, 5340 for DSDs, and 309 for other spinal diseases. After ERAS protocol implementation, spine surgical procedures showed significant reductions in LOS and medical costs by 22% (p = 0.008) and 22% (p < 0.001), respectively. The DSD group demonstrated a 16% (p < 0.001) reduction in LOS, whereas the PST group achieved a 28% (p < 0.001) reduction, noting a more pronounced LOS reduction in PST surgical procedures (p = 0.003). Medical costs decreased by 23% (p < 0.001) in the DSD group and 12% (p = 0.054) in the PST group, with a larger cost reduction for DSD surgical procedures (p = 0.021). No statistically significant differences were found in the rates of 30-day readmission, reoperation, and SSI between the developing and post-current ERAS implementation periods (p = 0.65, p = 0.59, and p = 0.52, respectively). CONCLUSIONS: Comprehensive ERAS protocol implementation significantly reduced LOS and medical costs in all spine surgical procedures, while maintaining comparable 30-day readmission, reoperation, and SSI rates. These findings suggest that the ERAS protocol is equally applicable to all spine surgical procedures, with a more pronounced effect on reducing LOS in the PST group and on reducing medical costs in the DSD group.

A cost-utility analysis between decompression only and fusion surgery for elderly patients with lumbar spinal stenosis and sagittal imbalance.

Lumbar spinal stenosis (LSS) and sagittal imbalance are relatively common in elderly patients. Although the goals of surgery include both functional and radiological improvements, the criteria of correction may be too strict for elderly patients. If the main symptom of patients is not forward-stooping but neurogenic claudication or pain, lumbar decompression without adding fusion procedure may be a surgical option. We performed cost-utility analysis between lumbar decompression and lumbar fusion surgery for those patients. Elderly patients (age > 60 years) who underwent 1-2 levels lumbar fusion surgery (F-group, n = 31) or decompression surgery (D-group, n = 40) for LSS with sagittal imbalance (C7 sagittal vertical axis, C7-SVA > 40 mm) with follow-up ≥ 2 years were included. Clinical outcomes (Euro-Quality of Life-5 Dimensions, EQ-5D; Oswestry Disability Index, ODI; numerical rating score of pain on the back and leg, NRS-B and NRS-L) and radiological parameters (C7-SVA; lumbar lordosis, LL; the difference between pelvic incidence and lumbar lordosis, PI-LL; pelvic tilt, PT) were assessed. The quality-adjusted life year (QALY) and incremental cost-effective ratio (ICER) were calculated from a utility score of EQ-5D. Postoperatively, both groups attained clinical and radiological improvement in all parameters, but NRS-L was more improved in the F-group (p = 0.048). ICER of F-group over D-group was 49,833 US dollars/QALY. Cost-effective lumbar decompression may be a recommendable surgical option for certain elderly patients, despite less improvement of leg pain than with fusion surgery.