Variability in Assessing Spinopelvic Parameters With Lumbosacral Transitional Vertebrae: Inter- and Intraobserver Reliability Among Spine Surgeons.


Journal Article

STUDY DESIGN: Prospectively collected survey study OBJECTIVE.: The aim of this study was to determine the consistency with which spinopelvic parameters (SPP) are determined in patients with lumbosacral transitional vertebrae (LSTV). SUMMARY OF BACKGROUND DATA: The incidence of LSTV in the general population is as high as 35.6%. The often fixed nature of LSTV relative to the pelvis, but lumbar-type appearance, may lead to differential use of the S1 endplate when performing SPP assessment. This could have significant impact on SPP derived from these landmarks, resulting in considerable variation in surgical planning and decision-making. METHODS: Twenty patients demonstrating LSTV on standing lateral 36-inch spinal radiographs were randomly arranged and independently analyzed by 16 experienced spine surgeons using the same computer software. Pelvic incidence (PI), pelvic tilt (PT), lumbar lordosis (LL), and T1 pelvic angle (TPA) were captured. Two weeks after the first assessment, surgeons repeated the measurements after image sequence re-randomization. Intraclass correlation coefficient (ICC) was calculated to evaluate interobserver reliability (IOR) for each SPP. Intraobserver reliability (IAOR) was assessed through an average Pearson correlation coefficient for each parameter for each surgeon. RESULTS: Sixteen surgeons completed initial measurements. IOR was poor for TPA (0.35, 95% confidence interval [CI] 0.20, 0.58) and PI (0.42, 95% CI 0.26, 0.65) and fair for LL (0.67, 95% CI 0.51, 0.82), and PT (0.63, 95% CI 0.47, 0.81). Fourteen surgeons completed phase-2 measurements to assess IAOR. Average parameter PPC showed excellent IAOR (LL 0.86, TPA 0.77, PI 0.78, PT 0.86). Kappa coefficient showed fair agreement for raters choosing the same endplate for measurement (Phase 1: 0.38, Phase 2: 0.32). By patient, the percentage of raters that chose the S1 endplate for measurement varied from 6.3% to 85.7%. CONCLUSION: Significant variability exists when surgeons measure SPP in patients with LSTV. These parameters are critical in determining the goals of surgical reconstruction and such variability may have considerable implications for radiographic goals and outcomes of surgical reconstruction. LEVEL OF EVIDENCE: 4.

Full Text

Duke Authors

Cited Authors

  • Khalsa, AS; Mundis, GM; Yagi, M; Fessler, RG; Bess, S; Hosogane, N; Park, P; Than, KD; Daniels, A; Iorio, J; Ledesma, JB; Tran, S; Eastlack, RK; International Spine Study Group,

Published Date

  • June 15, 2018

Published In

Volume / Issue

  • 43 / 12

Start / End Page

  • 813 - 816

PubMed ID

  • 29846363

Pubmed Central ID

  • 29846363

Electronic International Standard Serial Number (EISSN)

  • 1528-1159

Digital Object Identifier (DOI)

  • 10.1097/BRS.0000000000002433


  • eng

Conference Location

  • United States