Does conflict of interest affect the reported fusion rates of bone graft substitutes and extenders?

Bone graft extenders are being used more in spine surgery as a substitute for iliac crest bone graft. However, potential conflict of interest could impact average fusion rates. The purpose of this study was to evaluate whether fusion rates reported in the literature were different in papers evaluating bone graft substitutes and extenders when there was potential conflict of interest versus no potential conflict of interest.

Pubmed was searched for studies evaluating fusion rates when bone graft extenders including demineralized bone matrix, hydroxyapatite, and tricalcium sulfate were used. Studies were screened for one or two level fusions and for degenerative spinal conditions. The average fusion rates of subgroups were compared using unpaired Student's t-tests.

1928 studies were evaluated. 86 studies were included in the study. The fusion rates varied from 4 to 100%. There were 24 studies with a potential conflict of interest and 62 studies with no conflict of interest. The average fusion rate of all the studies was 84.63% with standard deviation of 18.33%. The average fusion rates of those studies with conflict of interest was 80.93% versus 86.06% without conflict of interest. This was not statistically significant (p>0.07). The average fusion rate of studies evaluated by CT scan was 79.8% versus 87.9% without CT. The average fusion rate of studies that employed an independent reviewer to evaluate the fusion was 82.61% versus 85.63% for studies with no independent reviewer.

There is a great variability in the reported fusion rates of bone graft extenders. Counter to expectations, average fusion rates were lower in the studies where there was a potential conflict of interest. The use of CT scans and an independent reviewer seem to account for the lower reported fusion rates, and may be a means of negating the potential conflicts of interest in fusion studies.

2.