The adoption of a new classification system: time-dependent variation in interobserver reliability of the thoracolumbar injury severity score classification system

Abstract

Study design: Prospective clinical assessment of the interobserver reliability of the Thoracolumbar Injury Classification and Severity Score (TLISS) in a series of consecutive patients.

Objective: To evaluate the time-dependent changes in interobserver reliability of the TLISS system.

Summary of background data: Reliability of an injury classification system is fundamental to its usefulness. A system that can be taught and implemented effectively will be highly reliable. Vaccaro et al recently introduced a novel thoracolumbar injury classification and treatment recommendation system called the "Thoracolumbar Injury Classification and Severity Score." An improvement over previous traumatic thoracolumbar systems, it has been designed to be both descriptive as well as prognostic. To define better the benefits of this system, the purpose of our study was to assess the time-dependent changes associated with implementation of the TLISS system at 1 institution.

Methods: Seventy-one consecutive patients presenting with acute thoracolumbar injury were prospectively assessed at a single training institution. Plain radiographs, computed tomography, and magnetic resonance imaging were independently reviewed, and each case was classified according to the TLISS system. Seven months later, 25 consecutive patients presenting with acute thoracolumbar injuries were prospectively assessed at the same institution. TLISS classification criteria were again applied after reviewing plain radiographs, computed tomography, and magnetic resonance imaging. The unweighted Cohen kappa coefficient and Spearman correlation values were calculated to assess interobserver reliability at each assessment time. Interobserved reliability at the time of the first assessment was then compared with interobserver reliability from the second assessment.

Results: Statistically significant (P < 0.05) improvements in interobserver reliability were observed. Both the unweighted Cohen kappa coefficient and Spearman correlation values increased across all comparable fields: TLISS subscores (mechanism of injury, posterior ligamentous complex), total TLISS, and TLISS management scores.

Conclusions: The significant improvements observed in interobserver reliability of the TLISS system suggest that the classification system can be taught effectively and be readily incorporated into daily practice. The strong correlation values obtained at the second assessment time suggest that the TLISS system may be reproducibly used to describe thoracolumbar injuries.