Reliability analysis based on expert panel case series review and grading per the Enneking and Weinstein-Boriani-Biagini classification systems.
To assess the reliability of the Enneking and Weinstein-Boriani-Biagini classification systems.
The Enneking and Weinstein-Boriani-Biagini (WBB) classifications were developed to stage and facilitate treatment planning in patients with primary spine tumors. To date, their interobserver and intraobserver reliability has not been assessed–a fundamental step in facilitating broader clinical and research use.
Clinical information, imaging studies, and biopsy results were compiled from 15 selected patients with primary spinal tumors. Eighteen spine surgeons independently estimated and scored the cases for Enneking grade, tumor and metastasis categories, Enneking stage, Enneking-recommended surgical margin, WBB zones and layers, and WBB-recommended surgical procedures, with a second assessment performed after random resorting of cases. Interobserver and intraobserver reliability of each category were assessed by percent agreement or proportional overlap. The Fleiss, Cohen, and Mezzich κ statistics (κ) were then applied, determined by the type of variable analyzed.
The κ statistics for interobserver reliability were 0.82, 0.22, 0.00, 0.57, 0.47, 0.31, 0.58, and 0.54 for the fields of Enneking grade, tumor and metastasis categories, Enneking stage, Enneking-recommended surgical margin, WBB zones and layers, and WBB-recommended surgical procedures, respectively. The κ statistics for intraobserver reliability were 0.97, 0.53, 0.47, 0.82, 0.67, 0.63, 0.79, and 0.79 for the same respective fields. According to Landis and Koch, the ranges of κ values of 0.00 to 0.20, 0.21 to 0.40, 0.41 to 0.60, 0.61 to 0.80, and >0.80 imply slight, fair, moderate, substantial, and near-perfect agreement, respectively.
Results indicate moderate interobserver reliability and substantial and near-perfect intraobserver reliability for both the Enneking and WBB classification in terms of staging and guidance for treatment, despite a less than moderate interobserver reliability in interpreting the Enneking local tumor extension and WBB sector. Before incorporating the classifications in the clinical practice and research studies, further work is required to investigate the validity of the classifications.
The interobserver and intraobserver reliability of the Enneking and Weinstein-Boriani-Biagini staging systems for the evaluation and management of individuals with primary spinal tumors was assessed. Moderate interobserver reliability was noted for both systems. Substantial and near-perfect intraobserver reliability was noted for the Enneking and WBB systems, respectively.
From the *Vancouver General Hospital, Vancouver, Canada; †Ospedale Maggiore, Bologna, Italy; ‡University of Saskatchewan Saskatchewan, Canada; §Instituto Regina Elena, Rome, Italy; ¶Mayo Clinic, Rochester, MN; ∥University of Toronto, Toronto, Canada; **University of Iowa, Iowa City, IA; ††Johns Hopkins University, Baltimore, MD; ‡‡H. Lee Moffitt Cancer Center, Tampa, FL; §§Thomas Jefferson University Hospital, Philadelphia, PA; ¶¶University of Utah, Salt Lake City, Utah; ∥∥Catholic University of Parana, Curitiba, Brazil; ***University of Pittsburgh, Pittsburgh, PA; †††MD Anderson Cancer Center, Houston, TX; ‡‡‡Northwestern University, Chicago, IL; and §§§Medtronic Spinal and Biologics, Memphis, TN.
Acknowledgment date: December 17, 2007. Revision date: July 7, 2008. Acceptance date: September 21, 2008.
The manuscript submitted does not contain information about medical device(s)/drug(s).
Funds were received in support of this work. No benefits in any form have been or will be received from a commercial party related directly or indirectly to the subject of this manuscript.
Supported by the Spine Oncology Study Group and funded by an educational/research grant from Medtronic Spinal and Biologics.
The Spine Oncology Study Group acknowledges prior work on this topic by Bradford L. Currier, MD, Department of Orthopedic Surgery, Mayo Clinic, Rochester, MN.
Address correspondence and reprint request to Charles G. Fisher, MD, Vancouver General Hospital, 2733 Heather Street, D6 Heather Pavilion, VGH, Vancouver, BC V5Z 3J5; E-mail: Charles.Fisher@vch.ca