Bone harvested by intramedullary reaming offers a minimally invasive alternative to harvesting bone from the iliac crest, which has long been considered the gold standard for autogenous bone grafting. The biologic potential of intramedullary reaming material has been studied both in vitro and in vivo. The material provides osteogenic, osteoinductive, and osteoconductive properties that are comparable to the material harvested from the iliac crest. In addition to the ability to obtain a large volume of bone, the graft harvested by the Reamer-Irrigator-Aspirator has been shown to be rich in growth factors, including BMP-2, TGF-β1, IGF-I, FGFa, and PDGFbb.
The iliac crest has long been considered to be the gold standard donor site for harvesting autogenous bone for use in situations requiring a biologically active bone graft. Complications and morbidity associated with autogenous iliac crest bone grafting is well documented. Several decades ago surgeons suggested the use of bone debris obtained from reaming of the intramedullary canal of long bones as an alternative source of autogenous bone graft material. With standard reamers, bone graft may be extracted from the intramedullary canal as the reamer is pulled out of the bone, and some additional graft may be recovered from the flutes of the reamer heads. Although in these reports the use of bone graft harvested by intramedullary reaming was successful, there has been no direct clinical comparison of bone graft material harvested from the intramedullary canal with that harvested from the iliac crest.
The Reamer-Irrigator-Aspirator (RIA, Synthes, Paoli, PA) was initially designed to decrease embolic complications associated with standard reaming; however, clinicians quickly capitalized on its ability to harvest a large quantity of autogenous bone graft. Using the RIA, graft material is removed from the canal by continuous suction and irrigation. Corticocancellous bone and other particulate debris can be harvested with a simple in-line filter system. In addition, a large volume of aspirate, including the irrigation fluid, is collected in a separate container. Compared with a standard reamer, the RIA device offers a much more efficient technique for harvesting material from the intramedullary canal. The biologic potential of bone harvested from the intramedullary canal with standard reamers has been previously studied, and with the advent of the use of RIA there has been renewed interest in exploring the biologic properties of this graft source.
Iliac crest bone graft is considered the gold standard because it is known to be osteogenic, osteoconductive, and osteoinductive. To be osteogenic, a bone graft substance must contain viable osteoblasts or stem cells that can differentiate along the bone-forming lineage. To be osteoconductive, the material must have a 3-dimensional architecture that promotes bone formation. Finally, to be osteoinductive, the substance must contain bone morphogenic proteins. For bone harvested from intramedullary reaming to serve as an alternative graft source it must possess these same properties.
Cellular content of intramedullary reamings
The intramedullary canal contains bone marrow and vasculature that make up two microenvironments or “niches,” termed the vascular and osteoblastic “niches.” Most stem cells found in the bone marrow are hematopoietic stem cells (HSC), whereas the smaller population of stem cells are the mesenchymal stem cells (MSC).
The Osteoblastic Niche
One population of HSCs is found lining the endosteal bone in close proximity to osteoblasts, which suggests a close relationship between hematopoiesis and osteogenesis. A significant function of the osteoblastic niche is to support the HSCs and hematopoiesis. When bone marrow is transplanted into an irradiated organism, it is necessary for osteoblasts to be present to obtain a successful grafting of the transplanted marrow. The osteoblastic niche is responsible for providing the signals that regulate the population of HSCs in undifferentiated states. Of these signals, bone morphogenic protein (BMP) and parathyroid hormone (PTH) are key proteins that regulate the formation of osteoblastic cells lining trabecular bone.
The Vascular Niche
The vascular niche of the intramedullary bone marrow provides a pathway for the mature hemopoietic cells generated in the osteoblastic niche to reach the peripheral circulation through the fenestrated walls of the bone marrow sinusoids. Within the vascular niche, a cell line of mesenchymal origin is found surrounding and in direct contact with the endothelial cells of the microvasculature. Because of the rich intramedullary blood supply, bone harvested from the intramedullary canal may contain a larger quantity of blood vessel fragments and thus a greater number of these MSCs than bone harvested from the iliac crest.
Pericytes are distinct, polymorphic cells that have multiple, branching cytoplasmic processes that encircle the microvascular capillary. There is evidence to suggest that the pericyte acts as a osteoblast progenitor cell when stimulated in vitro; however, it is not clear under what circumstances the pericyte will go down this particular differentiation pathway in situ.
Doherty and colleagues placed isolated pericytes in a diffusion chamber and followed their differentiation and showed that the pericytes expressed markers that are indicative of osteoblast formation (osteopontin, osteonectin, bone sialoprotein, and osteocalcin). The high levels of osteopontin secreted at the onset of mineralization detected within the diffusion chamber have been shown in other studies to be indicative of mature osteoblast formation. Brighton and colleagues showed that pericytes differentiated down the osteoblastic pathway and synthesized an increased amount of alkaline phosphatase (2–3 times greater) when placed under low oxygen tension (3%) conditions versus 21% to 60% oxygen.
Variation in cellular content based on harvest location
Although the iliac crest has long been the gold standard, it is not because the bone harvested from this site has been identified as superior. In fact, there is some variation in the population of stem cells among different bone sites, and alternative bone graft donor sites may offer higher concentrations of various stem cell populations.
Investigators have studied the use of bone marrow aspiration from the vertebral body as an alternative to the iliac crest. In a clinical study of 21 adults undergoing posterior spinal fixation, McLain and colleagues found that transpedicular aspirates of vertebral bone marrow had greater concentrations of progenitor cells compared with matched controls from the iliac crest. In a similar study, investigators studied bone marrow harvested from the vertebral body and iliac crest of 15 adults undergoing both anterior and posterior spinal surgery. Progenitor cells from the vertebral body exhibited an increased level of alkaline phosphatase activity. Higher numbers of proliferating colonies of cells that promote osteogenesis were found in the vertebral body aspirates compared with the iliac crest aspirates. Progentitor cells from both sites expressed comparable levels of CD166, CD105, CD49a, and CD63 cell surface receptors.
Growth factor content of intramedullary reamings
In a direct comparison between iliac crest and intramedullary reamings, Schmidmaier and colleagues determined that reaming debris contains comparable quantities of growth factors as iliac crest. In addition, the aspirated irrigation fluid also contains growth factors. The growth factors (BMP-2, BMP-4, transforming growth factor [TGF]-β1, insulin-like growth factor-I (IGF-I), fibroblastic growth factor-a (FGFa), fibroblastic growth factor-b (FGFb), platelet-derived growth factor-bb (PDGFbb), and vascular endothelial growth factor [VEGF]) were quantified by enzyme-linked immunosorbent assay (ELISA) assay and then normalized by the weight of the harvested bony material. BMP-4 was not detected in material taken from either the iliac crest or the RIA reaming debris. However, five growth factors (BMP-2, TGF-β1, IGF-I, FGFa, and PDGFbb) were found to be higher in the RIA reaming debris compared with iliac crest curettings. Not only were five of the quantified growth factors found to be higher in the RIA reaming debris than in the iliac crest graft material, but the total protein was found to be 2.1 times higher in the RIA reaming debris. Iliac crest had higher levels of VEGF and FGFb than the RIA reaming debris. Although this study did quantify the various growth factors present in RIA reaming debris and in the RIA aspiration fluid, the study did not specifically examine their osteoinductive capacity.
Viability of material from intramedullary reaming
Investigators have raised concern about the cellular viability of bone grafts harvested by intramedullary reaming secondary to the heat that can be generated during the reaming process. Thermally induced osteonecrosis not only injures the cortical bone and cells, but it can also denature critical proteins necessary for stimulating bone growth.
Necrosis of bone can occur at a temperature of 46°C following an exposure time of 5 minutes. The small vessels coursing through bone are even more heat sensitive than bone tissue, and thermal damage to the blood vessels can impair the nutrition of the bone. Using specific investigation methods, others have found that the threshold value of thermal osteonecrosis was 47°C with an exposure time of 1 minute.
The design of intramedullary reamers has evolved from early designs with shallow flutes to contemporary designs with deep flutes and narrow drive shafts. Although this evolution occurred to minimize pressure build up during reaming, it also results in decreased temperature elevation that can be seen with older reamer designs. Significant heat generation is possible in certain circumstances that may lead to thermal bone necrosis. Because of the continuous irrigation, and the sharp single-use reamer cutting head, heat build up with the RIA system is substantially less than with conventional reaming. Higgins and colleagues reported that the maximum temperatures reached in the distal diaphysis in the RIA group (42.0 ± 9.1°C) were significantly lower ( P = .025) than in the standard reaming group (58.7 ± 15.9°C).