Lumbosacral Transitional Vertebrae Detected on Routine Sacroiliac Joint Radiographs in a Rheumatology Cohort: Comparison of Castellvi and Jenkins Systems
In brief
Jenkins classification flags 90% of patients, versus 30% with Castellvi
4% with Types II to IV. Neither classification was linked to documented low back pain after adjustment, and radiology reports mentioned none of the findings. The age-related rise in Jenkins Type I findings calls its 10-mm threshold's clinical meaning into question.
- Journal
- Journal of clinical medicine (Q1)
- Published
- 29 October 2026
- Study design
- Retrospective cohort
- Evidence level
- Level 3, Low (CEBM 3b)
- Authors
- Jakub Jankowski, Natalia Maślanek, Marcin Glin, Marcin Jasiewicz, Agnieszka Pawełczyk, Michał Polguj, et al.
- PMID
- 43030580
- DOI
- 10.3390/jcm15218436
Why clinicians should know about it
- Picked for Anatomy (paper of the day, 16 November 2026): Population prevalence and classification of lumbosacral anatomical variation
Abstract
Background/Objectives: Lumbosacral transitional vertebrae (LSTV) are an under-recognized anatomical variant potentially associated with low back pain (LBP). Comparative data on the established Castellvi classification and the more recently proposed Jenkins system are also limited. This study compared LSTV detection by the two systems on routine sacroiliac joint radiographs and evaluated associations with age, LBP, and rheumatologic diagnoses. Methods: This retrospective study included 817 patients who underwent sacroiliac joint radiography at a rheumatology outpatient clinic between 2022 and 2024. Radiographs were reassessed using a standardized protocol. Age-related patterns and alternative transverse process-sacral ala distance thresholds were evaluated. Associations between LSTV and documented LBP were adjusted for age, sex, radiographic sacroiliitis grade, and the presence of a recorded rheumatologic diagnosis. Results: LSTV was identified in 248/817 patients (30.4%) by Castellvi and in 735/817 (90.0%) by Jenkins. The discrepancy was driven predominantly by Type I variants (20.9% vs. 80.5%); after their exclusion, both systems identified the same 77/817 patients (9.4%) as having Types II-IV. Jenkins-positive prevalence rose from 68.4% in patients younger than 20 years to 100% in those aged ≥70 years (Cochran-Armitage Z = 6.80, p < 0.001), whereas no corresponding trend was observed for Castellvi. An exploratory TP-sacral ala threshold of <5 mm reduced Jenkins-positive prevalence to 61.7%. Neither classification was associated with documented LBP after multivariable adjustment. LSTV was not mentioned in any original sacroiliac joint radiology reports. Conclusions: The Castellvi and Jenkins systems yielded markedly different LSTV prevalence estimates, primarily because of their differing Type I criteria. The high prevalence and strong age dependence of Jenkins Type I findings suggest that the 10-mm distance criterion may capture age-dependent anatomical configurations in addition to transitional morphology and highlight the need for independent validation before its broader clinical use.
Abstract as published, via PubMed.
For healthcare professionals. The summary is generated by AI from the published abstract, and the evidence level is assigned automatically from the study design on the Oxford CEBM hierarchy. Neither is medical advice. Read the full paper before changing practice.