#87 Interobserver Reliability of CT Scans to Assess Radiographic Fusion Criteria with a Novel Titanium Interbody Device

General Session: Cutting Edge Technology - Imaging Guidance

Presented by: P. Slosar


P.J. Slosar (1)
L. Marrero (1)
J. Kaiser (2)
D. Sacco (3)

(1) Spine Care Medical Group, San Francisco Spine Institute, San Francisco, CA, USA
(2) California Advanced Imaging Medical Associates, Novato, CA, USA
(3) National Orthopedic Imaging Associates, Greenbrae, CA, USA


Introduction: The accuracy of CT scan analysis of titanium cages for assessment of lumbar interbody fusion has been questioned in the past. Previously published reports on this issue have generally focused on older technologies using paired threaded titanium cages. Radiologists reported difficulties assessing fusion bone due to cage artifact and small graft windows. A unique interbody fusion device has been developed with an acid-etched titanium surface for surface bone integration and a wide aperture for bone graft. The purpose of this study was to determine the reliability of CT scans to assess fusion bone formation and other radiographic variables with this novel titanium interbody device.

Methods: Patients underwent ALIF at L3-S1 had CT scan analysis as part of a separate clinical outcomes study. All patients received the same implant type (Endoskeleton TA®; Titan Spine LLC.). Each cage was packed with 2 sponges of InFuse® (Medtronic) (3.0 mg per fusion level). 1-3 cc. of HA/ TPC (MasterGraft Matrix®) collagen sponge was used as a graft extender to fill any remaining gaps within the fusion cage. Fine cut CT scans / reconstructed images were performed randomly at 6, 9, or 12 months. Two independent radiologists reviewed the scans and were blinded to clinical results and the other's interpretations. Interobserver reliability was calculated using the kappa statistic.

Results: 33 patients/ 56 spinal fusion segments were analyzed. There were 17 males and 16 females with an average age of 46 years (range 23-66 years). Six patients (18%) were nicotine users. 17 patients underwent CT scanning at 6 months, 9 patients at 9 months, and 7 patients at 12 months. There were no differences in results reported at the separate time points nor between males, females, or nicotine users. The kappa coefficient was calculated and noted in Tables 1-2. The kappa for the overall study was 0.88 (p< .001) demonstrating strong agreement between the 2 radiologists in 345 of the 392 data points reviewed. The agreement for fusion formation was 0.77 (p< .001). All other measurements reached statistical significance.

Conclusion: Clinicians in the past have been reluctant to rely on CT scans with titanium cages due to concerns that the interpretation of the images may not be accurate. The interbody device used in this study demonstrated minimal artifact, minimal subsidence, and trabecular bone was easily identified through the implant in the overwhelming majority of CT scans reviewed. This study demonstrates critical radiographic fusion variables can be reliably assessed within this unique titanium fusion device with a high degree of interobserver agreement.

Fusion Grades

Other Fusion Criteria