Study design: A retrospective comparison of three different types of bone grafts for posterior spinal fusion in adolescent idiopathic scoliosis.

Objective: To determine the efficacy of bone marrow and demineralized bone matrix as a bone graft substitute for spinal fusion.

Summary of background data: Several reports have documented a high morbidity associated with harvesting autologous iliac crest bone graft (ICBG) for spinal fusion. Composite bone graft consisting of demineralized bone matrix and aspirated bone marrow may reduce the morbidity and still retain the osteoinductive properties of iliac crest autograft.

Methods: Three different bone grafting techniques were used by a single surgeon in 88 consecutive patients who had posterior spinal fusion for adolescent idiopathic scoliosis. Segmental instrumentation with dual-rod fixation was used in all cases. Selection of type of graft was determined historically by the time when the operations were performed. Autologous ICBG was used in Group A, freeze-dried corticocancellous allograft in Group B, and composite graft of autologous bone marrow and demineralized bone matrix in Group C. Seventy-seven patients were reviewed, with a minimum of 2 years' follow-up (mean, 3 years 7 months; range, 2 years-9 years 5 months). Radiographs were assessed for pseudarthrosis and loss of correction of 10 degrees or more. Loss of 10 degrees of correction has been previously identified as an indicator of potential pseudarthrosis or fusion instability. Both of these criteria were used to compare success of fusion.

Results: Failure caused by pseudarthroses was seen in two patients (2.6%), one in Group A and one in Group B. Eleven patients lost greater than 10 degrees of correction, but only one demonstrated pseudarthroses. The 13 patients with pseudarthroses or loss of correction constitute the failure group for purposes of graft assessment. The failure rate was 12.5% in Group A (ICBG), 28% in Group B (freeze-dried corticocancellous allograft), and 11.1% in Group C (composite graft of autologous bone marrow and demineralized bone matrix). Eliminating patients with crankshaft phenomenon did not substantially change the results. There was no morbidity associated with bone marrow aspiration.

Conclusions: Fusion rates were comparable for GroupA (ICBG) and Group C (composite graft of autologous bone marrow and demineralized bone matrix). The composite graft is our preferred graft for fusions in adolescent idiopathic scoliosis.