Secondary Logo

Journal Logo

The Spine Blog

Friday, July 3, 2020

Biphasic calcium-phosphate: as good as iliac crest bone graft?

Spine surgeons have been in search of a high-quality bone graft substitute for decades. Iliac crest bone graft (ICBG) has been the gold standard to which other graft options have been compared, and only BMP-2 has been shown to be equally effective. While enthusiasm for BMP-2 was initially high, use has dropped off over concerns about cost and safety. Local bone graft, allograft, demineralized bone matrix, and inorganic scaffolds (i.e. calcium phosphate derivatives) have been studied, and none are equivalent to ICBG, especially in challenging fusion environments (i.e. 3 or more levels, smokers, osteoporosis, uninstrumented fusions, etc.) In an effort to create a reasonably priced, effective bone graft substitute, Nuvasive Inc. developed a biphasic calcium-phosphate (BCP) product that they report to be osteoinductive. The Dutch Clinical Spine Research Group performed a trial in which 100 lumbar and thoracolumbar instrumented fusion patients were treated with ICBG on one side and BCP on the other. Sixty-six patients underwent single level fusion, 20 patients had 2 level fusion, and 14 had 3 or more levels fused (average of 4.7 levels fused in this group). Sixty-two percent had an interbody fusion as well. Fusion at each level on each side was evaluated with a CT scan at one year after surgery and rated as fused, doubtful fusion, and nonunion. Eighty-seven patients underwent 1 year CT scan and were included in the analysis. Overall, 71% of segments were fused on one or both sides. Fifty-five percent of the BCP levels were graded as fused compared to 52% of the ICBG levels. The study was powered to show non-inferiority of BCP at an absolute level of 15% with a 90% confidence interval, and this was demonstrated.

Various inorganic matrices have been evaluated as bone graft substitutes, and generally these have not been as effective as ICBG. Prior to concluding that BCP is as effective as ICBG, the multiple limitations of this study need to be considered. One of the most concerning is that surgeons were limited to using 8 cc of autograft (ICBG + local bone graft) per level in order to use an equivalent volume of BCP. The authors required that at least 50% of this had to be ICBG, meaning that only 4 cc of ICBG might be used in a single level fusion This situation does not reflect reality, as surgeons typically harvest much more than 4-8 cc of ICBG. The low fusion rate of 52% for ICBG may be a result of intentionally using an insufficient quantify of ICBG. A similar criticism about low quantities of ICBG was levelled against the Medtronic IDE studies evaluating BMP-2. This paper also combined patients undergoing single and multilevel fusions, including some patients who likely had fusions of more than 5 levels (the authors did not report the range of fusion levels). A five level fusion is a completely different fusion environment than a one level fusion, and it is not clear why these patients were combined in a single analysis. Similarly, it is not clear why patients both with and without interbody fusions were included, as the use of an interbody cage likely increases the posterolateral fusion rate as well. The study design in which the two different graft types were used in the same patient is also problematic as, contrary to the authors assertions, the two sides are not independent of each other. If one side fuses solidly and prevents loosening of the hardware prior to the other side fusing, it likely increases the likelihood of the other side fusing as well. This design also precludes any comparison of PROs between the graft types. The authors suggested that this study design prevented between-group differences as each patient serves as his own control, however, a good randomization scheme also essentially eliminates between-group differences. The real question that this paper did not address is in which situations would the use of BCP as a bone graft extender be helpful. It is probably not necessary in one or two level instrumented fusions as the fusion rate is fairly high with local bone graft alone. It could play a role in multilevel fusions or other challenging fusion environments, though this paper did not address those situations. This paper represents an industry-sponsored trial that seems to have been designed in a way to show that their product was equivalent to ICBG. Given the limitations described above, this is probably not a fair conclusion.

Please read this paper in the July 15 issue. Does this change your thoughts about BCP as a bone graft substitute?

Adam Pearson, MD, MS
Associate Web Editor