Abstract
Aims:
In the absence of neurological deficits, the decision for surgery in spinal tuberculosis (STB) depends on the individual experiences of the surgeons, which may differ widely. There is currently no universal consensus on an objective definition of instability in STB. The Tuberculosis Spine Instability Score (TSIS) was developed to discriminate between the stable and unstable spine in STB. In the current study, we analyze the reliability, responsiveness, and construct validity of the TSIS to define clear guidelines for management of STB.
Methods:
Individuals presenting with STB were evaluated to assess the quality of the TSIS along with testing for validity, reproducibility, and responsiveness. Construct validity was expressed as the Pearson correlation coefficient. Intraobserver test-retest reliability and interobserver reliability was expressed using intraclass correlation. Longitudinal validity was assessed through responsiveness and an effect size calculation.
Results:
There were 162 individuals (98 females, 64 males) with STB identified with a mean age of 33.39 years (SD 16.58). The TSIS showed good construct validity with substantial correlation with the Spinal Instability Neoplastic Score (Pearson coefficient 0.827). Near perfect interobserver and intraobserver reliability was obtained with intraclass correlation coefficient values of 0.941 (95% CI 0.921 to 0.957) and 0.985 (95% CI 0.980 to 0.989), respectively. Evaluation of longitudinal validity was performed in 64 individuals at six months apart. With the smallest detectable change measure smaller than the minimal important change measure, the instrument was found to be responsive. The effect size over the six-month period was 1.039.
Conclusion:
The TSIS proves to be an excellent discriminative scoring tool with good validity, reliability, responsiveness, and high specificity. It can differentiate between the stable and unstable spine in STB and will provide objective assessments to aid surgical decision-making in this scenario.
In the absence of neurological deficits, the decision for surgery in spinal tuberculosis (STB) depends on the individual experiences of the surgeons, which may differ widely. There is currently no universal consensus on an objective definition of instability in STB. The Tuberculosis Spine Instability Score (TSIS) was developed to discriminate between the stable and unstable spine in STB. In the current study, we analyze the reliability, responsiveness, and construct validity of the TSIS to define clear guidelines for management of STB.
Methods:
Individuals presenting with STB were evaluated to assess the quality of the TSIS along with testing for validity, reproducibility, and responsiveness. Construct validity was expressed as the Pearson correlation coefficient. Intraobserver test-retest reliability and interobserver reliability was expressed using intraclass correlation. Longitudinal validity was assessed through responsiveness and an effect size calculation.
Results:
There were 162 individuals (98 females, 64 males) with STB identified with a mean age of 33.39 years (SD 16.58). The TSIS showed good construct validity with substantial correlation with the Spinal Instability Neoplastic Score (Pearson coefficient 0.827). Near perfect interobserver and intraobserver reliability was obtained with intraclass correlation coefficient values of 0.941 (95% CI 0.921 to 0.957) and 0.985 (95% CI 0.980 to 0.989), respectively. Evaluation of longitudinal validity was performed in 64 individuals at six months apart. With the smallest detectable change measure smaller than the minimal important change measure, the instrument was found to be responsive. The effect size over the six-month period was 1.039.
Conclusion:
The TSIS proves to be an excellent discriminative scoring tool with good validity, reliability, responsiveness, and high specificity. It can differentiate between the stable and unstable spine in STB and will provide objective assessments to aid surgical decision-making in this scenario.
| Original language | English |
|---|---|
| Pages (from-to) | 540-547 |
| Number of pages | 8 |
| Journal | Bone and Joint Journal |
| Volume | 107-B |
| Issue number | 5 |
| Early online date | 1 May 2025 |
| DOIs | |
| Publication status | Published - May 2025 |