Implications of lumbosacral transitional anatomy on vertebral numbering: evaluation of 3147 adult full-length spine radiographs.

Abstract

Objective: When evaluating anomalous vertebral anatomy, limited consensus exists on how to best label vertebrae, leading to numbering discrepancies between radiologists and surgeons. Errors in vertebral numbering can have devastating implications for patients, especially when intraoperative numbering errors occur. Analyzing whole-spine radiographs and identifying patterns of transitional anatomy relative to vertebral numbering could allow for a consistent numbering method.

Methods: This single-institution cross-sectional study included patients older than 18 years of age who underwent full-length spine imaging at the University of Colorado Hospital from January 31, 2018, to March 31, 2020. Patients with deformity or congenital fusion were included. A retrospective analysis was performed on full-length spine radiographs obtained in 3147 patients. The number of presacral mobile segments, number of ribs, and presence of hypoplastic or incomplete ribs were identified and recorded. Results were reviewed by a committee of musculoskeletal radiologists, neurosurgeons, and orthopedic spine surgeons, with verification through interobserver analysis.

Results: Among 3147 patients (age range 18-89 years), 2868 (91.1%) had the conventional 24 presacral mobile segments (7 cervical, 12 thoracic, 5 lumbar). Transitional anatomy, defined as having fewer or more than 24 presacral segments, was observed in 279 patients (8.8%). Specifically, 174 patients (5.5%) had 25 presacral segments, 104 patients (3.3%) had 23, and 1 patient (0.03%) had 26. Regarding the number of ribbed vertebrae (thoracic), 2976 patients (94.6%) had 12 ribs, including hypoplastic ribs, while 143 (4.5%) had 11 ribs, and 28 (0.88%) had 13 ribs. The co-occurrence of both lumbosacral transitional anatomy and an abnormal number of ribbed vertebrae (thoracic) was least prevalent and observed in 56 patients (1.8%). Considering the first nonribbed vertebra as the first lumbar vertebra aligned with the most prevalent pattern of transitional anatomy.

Conclusions: Evaluation of 3147 patients with full-length spine imaging demonstrated a lower prevalence of ribbed vertebral body anatomical variations compared with transitional lumbosacral anatomy. Based on these findings, the authors suggest further investigation of numbering vertebrae in a cranial to caudal progression, designating the first ribbed vertebra as thoracic (T1) and the first nonribbed vertebra in the lumbar spine as lumbar (L1). The clinical impact of a standardized vertebral numbering system is yet to be determined.