While MIS TLIF may limit soft-tissue damage during surgery, one of the challenges has been obtaining a solid fusion without open bone grafting techniques. The MIS approach yields very little local bone graft, and most MIS surgeons focus on obtaining an interbody rather than an intertransverse fusion. The use of BMP-2 has been an appealing solution to this problem as it is associated with high fusion rates, however, it has also been reported that placing BMP-2 in the interbody space can result in heterotopic bone formation in the foramen and epidural space, causing radicular symptoms. Other bone graft substitutes have performed poorly, leaving MIS TLIF surgeons to accept low fusion rates or use BMP-2 and risk heterotopic bone formation. One potential solution to this problem is to minimize the BMP-2 dose such that fusion rates remain acceptably high, while hopefully reducing the risk of heterotopic bone formation. Little is known about the lowest effective dose of BMP-2 in MIS TLIF or if there is a threshold below which heterotopic bone formation is minimized. In order to better assess this issue, Dr. Lytle and colleagues from Michigan State University retrospectively reviewed 690 patients treated with MIS TLIF using BMP-2 from 2009-2014, representing a total of over 1100 fused interspaces. All included patients had at least 6 months of follow-up with dynamic radiographs. About one third also had a CT scan performed beyond 6 months out from surgery. They reported an overall fusion rate of about 95%. They performed a multivariate logistic regression analysis that showed dose of BMP-2 was the only significant independent predictor of fusion. Number of levels fused was bordering on being a significant predictor, while smoking and diabetes were not. They then performed an analysis to determine if there was a threshold above which increasing the dose did not yield further gains in fusion rate. They found that increasing the dose from below 1 mg of BMP-2 per level to between 1-2 mg per level significantly increased the odds of fusion two-fold. Increasing the dose to greater than 2 mg per level also increased the fusion rate, but this difference was not significant. They did not report clinical outcomes, complication rates, or incidence of heterotopic bone formation. They concluded that the ideal dose of BMP-2 for MIS TLIF is between 1 mg and 2 mg.
The authors have done a nice job assembling a large retrospective cohort to look at an important question. The data do suggest that a dose lower than the commonly used 4.2 mg/level (dose in a small InFUSE kit) can result in high fusion rates. However, the authors did not evaluate clinical outcomes, complications or heterotopic bone formation, so it is not clear that the lower dose results in fewer problems. It is a reasonable assumption that a lower dose should result in less of an effect, both in terms of fusion rate and complications, though the threshold at which complications are reduced was not explored by this study. In addition to not evaluating the downside to using BMP-2, this study was also limited by a non-standardized evaluation of fusion. Both the timing of the definitive imaging and type of imaging was left up to the treating surgeons, and these are both potential confounders. If patients did not follow-up beyond 6 months, it is possible that they subsequently went onto fusion but were classified as non-union based on the 6 month imaging. Additionally, it seems likely that symptomatic patients were more likely to undergo CT scan, a more sensitive study for pseudarthrosis, so pseudarthrosis could have been detected more frequently in symptomatic patients. If either of these factors were associated with BMP-2 dosage, than confounding could be a problem. These are typical limitations of a retrospective cohort study of patients undergoing routine clinical care (as opposed to care defined by a study treatment protocol). Hopefully this study will be the impetus for an RCT evaluating different BMP-2 dosages in MIS TLIF. Surgeons currently have relatively scant data on which to make decisions about BMP-2 dosing, and error in either direction can be problematic.
Please read Dr. Lytle's study on this topic in the July 15 issue. Does this change how you consider BMP-2 dosing in MIS TLIF?
Adam Pearson, MD, MS
Associate Web Editor