Exploratory meta-analysis on dose-related efficacy and morbidity of bone morphogenetic protein in spinal arthrodesis surgery.

Abstract

Bone morphogenetic protein (BMP) is frequently used for spinal arthrodesis procedures in an "off-label" fashion. Whereas complications related to BMP usage are well recognized, the role of dosage is less clear. The objective of this meta-analysis was to assess dose-dependent effectiveness (i.e., bone fusion) and morbidity of BMP used in common spinal arthrodesis procedures. A quantitative exploratory meta-analysis was conducted on studies reporting fusion and complication rates following anterior cervical discectomy and fusion (ACDF), posterior cervical fusion (PCF), anterior lumbar interbody fusion (ALIF), transforaminal lumbar interbody fusion (TLIF), posterior lumbar interbody fusion (PLIF), and posterolateral lumbar fusion (PLF) supplemented with BMP. A literature search was performed to identify studies on BMP in spinal fusion procedures reporting fusion and/or complication rates. From the included studies, a database for each spinal fusion procedure, including patient demographic information, dose of BMP per level, and data regarding fusion rate and complication rates, was created. The incidence of fusion and complication rates was calculated and analyzed as a function of BMP dose. The methodological quality of all included studies was assessed according to the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. Data were analyzed using a random-effects model. Event rates are shown as percentages, with a 95% CI. Forty-eight articles met the inclusion criteria: ACDF (n = 7), PCF (n = 6), ALIF (n = 9), TLIF/PLIF (n = 17), and PLF (n = 9), resulting in a total of 5890 patients. In ACDF, the lowest BMP concentration analyzed (0.2-0.6 mg/level) resulted in a fusion rate similar to the highest dose (1.1-2.1 mg/level), while permitting complication rates comparable to ACDF performed without BMP. The addition of BMP to multilevel constructs significantly (p < 0.001) increased the fusion rate (98.4% [CI 95.4%-99.4%]) versus the control group fusion rate (85.8% [CI 77.4%-91.4%]). Studies on PCF were of poor quality and suggest that BMP doses of ≤ 2.1 mg/level resulted in similar fusion rates as higher doses. Use of BMP in ALIF increased fusion rates from 79.1% (CI 57.6%-91.3%) in the control cohort to 96.9% (CI 92.3%-98.8%) in the BMP-treated group (p < 0.01). The rate of complications showed a positive correlation with the BMP dose used. Use of BMP in TLIF had only a minimal impact on fusion rates (95.0% [CI 92.8%-96.5%] vs 93.0% [CI 78.1%-98.0%] in control patients). In PLF, use of ≥ 8.5 mg BMP per level led to a significant increase of fusion rate (95.2%; CI 90.1%-97.8%) compared with the control group (75.3%; CI 64.1%-84.0%, p < 0.001). BMP did not alter the rate of complications when used in PLF. The BMP doses used for various spinal arthrodesis procedures differed greatly between studies. This study provides BMP dosing recommendations for the most common spine procedures.