Backfilling of iliac crest defects with hydroxyapatite-calcium triphosphate biphasic compound: a prospective, randomized computed tomography and patient-based analysis

Abstract Background context Hydroxyapatite-calcium triphosphate (HCT) biphasic compounds are known to be efficacious in filling bone voids. No large study to date has assessed their radiographic efficacy in iliac crest voids with computed tomography (CT) analysis at a 2-year follow-up. Purpose To assess whether backfilling iliac crest defects with HCT biphasic compound decreases donor site pain and what effect backfilling has on CT appearance of the donor ilium. Study design Prospective randomized clinical trial. Patient sample Adult patients with spinal disorders undergoing spinal arthrodesis requiring posterior iliac crest bone grafting. Outcome measures Physician-administered visual analog scale (VAS) and pre- and postoperative CT analysis was performed. Methods This prospective, randomized, single-blind study followed patients requiring nonstructural posterior iliac crest harvest as part of spinal disorder treatment for 2 years. The harvest technique preserved both cortical tables and their periostea. All patients were randomized to backfill of HCT or no backfill. All patients had a CT of the pelvis immediately postoperative and at the 2-year follow-up. Computed tomography analysis was performed by a board-certified neuroradiologist. Analysis included qualitative assessment of the ilia appearance and defect density quantified in Hounsfield units. All patients completed VAS of their donor site pain (0–10, from low to high) at 6 weeks and 2 years postoperatively. Results Thirty-seven of 40 (17 women and 20 men) subjects returned for a mean 23.9-month follow-up (range, 22–29 months). The average age was 51.7 years (range, 27–79 years). Eighteen patients were in the backfill group (BF) and 19 were in the control group (C). There was no statistically significant difference in pain at 6 weeks or 2 years between the two groups. Bone density significantly decreased from postoperative to 2 years in BF (implying resorption of HCT and replacement of host bone) and significantly increased in C (implying reformation of host bone). Both groups had similar cortical defect repair. The backfill group had significantly better medullary defect repair (p