Diagnostic accuracy and reliability of fine-cut CT scans with reconstructions to determine the status of an instrumented posterolateral fusion with surgical exploration as reference standard

Study Design. Accuracy of a diagnostic test referenced to the gold standard. Objectives. This study evaluated the reliability and accuracy of fine-cut computed tomography scans with coronal and sagittal reconstructions to determine the status of an instrumented posterolateral fusion by using surgical exploration as the reference standard. Summary of Background Data. There is still a need for a reliable and accurate noninvasive method to determine the status of a spinal fusion. Methods. Three spine surgeons reviewed 93 prerevision fine-cut CT scans over 163 fused levels of consecutive patients who had revision surgery after an instrumented posterolateral lumbar fusion. The facet joints and posterolateral gutters at each level were classified as fused or not. The surgeons were unaware of the findings on surgical exploration. Interobserver variability and likelihood ratios for a solid fusion when both, one, or none of the facets and when both, one, or none of the posterolateral gutters were fused, were calculated. Results. There were 42 males and 51 females with a mean age of 57 years range, 19-86 years) at revision. On exploration, there were 32 (19.6%) nonunions over 163 levels. The kappa for interobserver variability for evaluating facet fusions (0.42) was moderate and for posterolateral fusions (0.62) was substantial. The probability of a solid fusion on exploration was higher when both posterolateral gutters were fused on CT scan (89%) than when both facets were fused on CT scan (74%). When both facets and both posterolateral gutters were fused on CT scan, the probability of a solid fusion on exploration is 96%. The absence of fusion of one or both facets or one posterolateral gutter were poor predictors of nonunion on surgical exploration. Conclusions. The CT scan reading of either one or both posterolateral gutters fused or both facets fused were moderately predictive of a solid fusion on surgical exploration. Fine-cut CT scans with reconstructions are moderately predictive of the presence of nonunion when both facets are not fused.