Study design: A systematic review.

Objective: To evaluate the efficacy, safety, and outcomes of osteoconductive bone graft extenders (BGEs) compared with iliac crest bone graft (ICBG) in posterolateral thoracolumbar spinal fusion.

Summary of background data: ICBG is the current "gold standard" for achieving spinal arthrodesis. However, morbidity associated with its harvesting has led to the increased use of BGEs.

Methods: An electronic literature search was conducted through April 2011 using MEDLINE, EMBASE, CENTRAL, and Cochrane Library. Risk of bias and methodological assessment was performed using the Cochrane Risk of Bias Tool. Higgins I(2) test was used to assess for heterogeneity. Pooled weighted relative risk (RR) ratios were calculated to compare fusion and adverse event rates. Weighted standardized mean differences were calculated to compare functional outcome and pain scores.

Results: Thirteen studies were included representing a total of 768 patients. Overall study quality was low (mean Cochrane Risk of Bias score, 4.8 out of 12; range, 3-6). Fusion rates were comparable between the BGE and ICBG groups (RR, 0.96; 95% confidence interval [CI], 0.89-1.03; P = 0.28). Higgins I(2) test (58%) suggested substantial heterogeneity in the pooling of studies. The pooled rate of donor site pain in the ICBG group was 11.2% (95% CI, 7.4%-15.1%). Reported adverse events, excluding donor site pain, were significantly lower in the BGE group (RR, 0.42; 95% CI, 0.28-0.64; P < 0.0001). Functional outcomes were not significantly different between the 2 groups.

Conclusion: Osteoconductive BGEs combined with local spine autograft and/or bone marrow aspirate have comparable fusion rates, similar functional outcomes, lower complication rates, and a lower risk of donor site pain than ICBG. Caution should be taken in interpreting these findings, given the low quality of the studies and the heterogeneity in the results. Randomized controlled studies using blinded assessments are required to help elucidate more conclusive evidence.