The minimally effective dose of bone morphogenetic protein in posterior lumbar interbody fusion: a systematic review and meta-analysis

Abstract

BACKGROUND CONTENTThe risks and benefits of recombinant human bone morphogenetic protein-2 (BMP) in posterior lumbar interbody fusion (PLIF) and transforaminal lumbar interbody fusion (TLIF) have been widely reported. However, the BMP dose associated with such reports varied widely. Additionally, data on the location of BMP placement on complications and fusion are lacking. PURPOSETo determine the minimally effective dose (MED) of BMP which results in optimal fusion rates while minimizing complications; to determine the effects of the location of BMP placement has on fusion rates and complications. STUDY DESIGNSystematic review and meta-analysis. STUDY SAMPLEAdult patients undergoing PLIF/TLIF for degenerative indications. OUTCOME MEASURESRates of radiculitis, fusion, osteolysis, heterotopic bone formation, and new cancer diagnosis. METHODSPubMed, Embase, and Cochrane Database were used to identify studies published between January 1, 2011 and April 30, 2019 reporting BMP usage in adult patients who underwent PLIF/TLIF degenerative indications. A qualitative and quantitative synthesis was performed to evaluate the MED of BMP and the effect of location of BMP placement on fusion and complications. Complications were defined as osteolysis, heterotopic bone growth, radiculitis, and rate of new cancer diagnosis. Complications and fusion outcomes were each pooled according to commercially available BMP doses. Additionally, complications and fusion outcomes were pooled according to 4 location groups (interbody cage only, interbody cage + posterolateral gutter [PLG], cage + interspace, and interspace + PLG). Heterogeneity was assessed with Q and I2 statistics. RESULTSTwenty-two articles, totaling 2,729 patients were included. Sixteen studies reported fusion and 15 reported complications. Among fusion studies, the mean BMP/level ranged from 1.28 to 12 mg/level. Among complication studies, the mean BMP/level ranged from 6.7 to 23.6 mg/level. The pooled overall fusion rate was 94.0% (91.4–95.8 confidence intervals). There was no significant difference in fusion and complication rates between different BMP doses. Thirteen studies included data on the location of BMP placement with 1,823 patients. At each BMP location, the fusion rate was not significantly different across the dose ranges (1.28–12 mg/level). We found the fusion rate to be marginally higher in the interspace + PLG group compared to the other groups. When BMP was placed in the interbody cage there was a mild increase in the rate of osteolysis compared to other placement locations. CONCLUSIONSFusion and complication rates did not differ significantly between different doses of BMP with the lowest MED for fusion as low as 1.28 mg/level. The location of BMP placement does not significantly affect fusion or complication rates.