Area Fraction and Compressive Strength of Newly Formed Bone in Medullarydefects Treated with Calcium Sulfate Pellets Containing Dbm Particles

Introduction: A previous study indicated that the area fraction of new bone after 6 weeks was greater in cavitary defects treated with calcium sulfate (CaSO4) pellets (mean 14.3%) compared to CaSO4 pellets containing demineralized bone particles (DBM) (10.5%) (p=.035) [1]. The CaSO4-treated defects exhibited a unique pattern of circular lamellae of residual CaSO4 and abundant new trabeculae of mineralized bone and unmineralized osteoid with intense remodeling activity. In contrast, defects treated with CaSO4 containing DBM particles had no remaining pellets, mature bony trabeculae, and little remodeling activity after 6 weeks, suggesting that the bone in defects treated with CaSO4 pellets with DBM might have remodeled and matured at a faster rate. The purpose of the present study was to test the hypothesis that after a longer time (26 weeks) the area fraction, compressive strength and histological appearance of new bone would be equivalent in defects treated with CaSO4 pellets compared to CaSO4 pellets containing DBM particles. A new model was developed to allow both histological and mechanical evaluation of the humeral defects. Methods: Under an IACUC-approved protocol, 7 skeletally mature, male hound-type dogs weighing 30-40 kg had a cylindrical, critical-size axial medullary defect (13 X 50mm) created in the proximal humerus, bilaterally [1]. In each dog, one humeral defect was treated with 50 (4.8 X 3.3mm circular) pellets of the control material, CaSO4 dihydrate (Osteoset, Wright Medical Tech.) and the contralateral humerus received 50 similar pellets of the test material, CaSO4 containing canine DBM particles. The test and control treatments were randomized by side. The osteogenic activity of the DBM particles was confirmed in a muscle pouch model in athymic mice. Radiographs were obtained preoperative, immediately postoperative and at 2, 6, 13 and 26 weeks, were graded for changes in pellet density and new bone within the defects. Following euthanasia, high-resolution radiographs were made of the isolated bones. Using a positioning jig, they were cut serially to produce comparable transverse cuts of the right and left bone pairs. Proximal, middle and distal slices of the bones were used to produce undecalcified, plastic-embedded sections, which were analyzed for the area fraction of new bone in the defects by computer analysis of BSE SEM images. Following this analysis, the sections were further ground and stained with basic fuchsin and toluidine blue for characterization of the nature of new bone and residual CaSO4 and DBM particles in the defects by light microscopy. For the mechanical tests, a 17 mm thick transverse section from between the middle and distal levels of the defect was obtained and frozen. In each section, the exact location of the defect was identified from specimen radiographs, and an 8 X 16 mm test cylinder was made along the long axis of the defect using a diamond core drill and a low speed cut-off saw. The specimens were later thawed in saline and unconfined, uniaxial compression tests were performed on an Instron materials testing machine at a compressive strain rate of 0.5mm/min. The compressive strength was determined as the maximum stress in the stress-strain curve. To validate the new mechanical testing protocol, right and left paired humeri were harvested from an additional 7 canines of similar age, size, sex and weight. The normal humeri were sectioned, cored and tested in a manner identical to the subsequent testing of the experimental specimens. These data were compared for right-left differences in the strength of the normal bones and used as base line information on the normal compressive strength of trabecular bone in the proximal humerus. The histological and mechanical data from the experimental and normal humeri were analyzed using nonparametric statistics, including the Friedman and Mann-Whitney tests. P< 0.05 was considered significant. Results: There were no intraoperative or postoperative complications. All of the dogs returned to weight bearing within 2 days of the surgical procedure and completed the study period without any clinical incident related to the treatments. On the clinical radiographs, none of the pellets migrated from the defects. After 2 weeks, pellet density diminished by approximately 25% in the CaSO4-treated defects and 50% in the CaSO4/DBM-treated defects. By 6 weeks, the pellets were no longer visible in any of the radiographs. Multiple sites of circular densities corresponding to previous pellet sites punctuated the CaSO4 pellet-treated defects. In contrast, defects treated with the CaSO4/DBM pellets had few circular densities corresponding to previous pellet sites. At 13 and 26 weeks, there was, overall, a similar radiographic appearance of defects treated with the two pellet types where the density of bone in the defect was similar to that of the surrounding medullary bone. Analysis of the SEM data indicated there was no difference in the area fraction of new mineralized bone when the CaSO4/DBM-treated and the CaSO4-treated defects were compared either in total or by level (Table 1). However, for each treatment, there was significantly more bone in the proximal and middle levels of the defect compared to the distal level (p=.008). The total area fraction of residual CaSO4 was quite low for both the test (0.34 +0.39%) and control (0.30 +0.16%) pellettreated defects.