A Comparative Assessment of Implant Site Viability in Humans and Rats

Abstract

Our long-term objective is to devise methods to improve osteotomy site preparation and, in doing so, facilitate implant osseointegration. As a first step in this process, we developed a standardized oral osteotomy model in ovariectomized rats. There were 2 unique features to this model: first, the rats exhibited an osteopenic phenotype, reminiscent of the bone health that has been reported for the average dental implant patient population. Second, osteotomies were produced in healed tooth extraction sites and therefore represented the placement of most implants in patients. Commercially available drills were then used to produce osteotomies in a patient cohort and in the rat model. Molecular, cellular, and histologic analyses demonstrated a close alignment between the responses of human and rodent alveolar bone to osteotomy site preparation. Most notably in both patients and rats, all drilling tools created a zone of dead and dying osteocytes around the osteotomy. In rat tissues, which could be collected at multiple time points after osteotomy, the fate of the dead alveolar bone was followed. Over the course of a week, osteoclast activity was responsible for resorbing the necrotic bone, which in turn stimulated the deposition of a new bone matrix by osteoblasts. Collectively, these analyses support the use of an ovariectomy surgery rat model to gain insights into the response of human bone to osteotomy site preparation. The data also suggest that reducing the zone of osteocyte death will improve osteotomy site viability, leading to faster new bone formation around implants.

Keywords: bone; computer model; drill; osteogenesis; osteotomy; ovariectomy.

Figure 1.

A model in ovariectomized rats that mimics a patient population undergoing osteotomy in preparation for dental implant placement. Representative sagittal micro–computed tomography (µCT) sections of distal femur of (A) young (19 wk old), (B) young + ovariectomy surgery (OVX) (19 wk old; 8 wk post-OVX), and (C) aged (>12 mo old) rats. Representative transverse µCT sections of distal femur of (A′) young (19 wk old), (B′) young + OVX (19 wk old; 8 wk post-OVX), and (C′) aged (>12 mo old) rats. (D) Quantification of distal femur bone volume/total volume (BV/TV). (E) Scatterdot plot of mean alveolar bone density (HU) in the indicated anatomic regions and proposed osteotomy site. Representative aniline blue–stained (F) transverse and (F′) sagittal tissue sections showing the extraction socket in young + OVX rats, harvested on postsurgery day 0 (PSD0). On PSD7, aniline blue staining of representative (G) transverse and (G′) sagittal tissue sections from the healing extraction socket of young + OVX rats (arrow indicates new bone formation). On PSD21, aniline blue staining of representative (H) transverse and (H′) sagittal tissue sections from the healed extraction site of young + OVX rats. (I) Quantification of aniline blue^+ve pixels/total pixels in the region of interest (ROI) (“—” refers to healed mxM1 extraction site; “+” refers to adjacent pristine bone). (J) The KLS, Twist, and Osteomed drills and (K) their corresponding protocols used for osteotomy site preparation at the healed mxM1 extraction sites. ab, alveolar bone; es, extraction socket; HU, Houndsfield units; pdl, periodontal ligament; RPM, revolutions per minute. Sample sizes (N) as indicated. Scale bars = 300 µm.

Figure 2.

KLS, Twist, and Osteomed drills produce comparable osteotomies. On postosteotomy day 0.5 (POD0.5), aniline blue staining of representative transverse tissue sections from osteotomies produced in young + ovariectomy surgery (OVX) rats using (A) KLS, (B) Twist, and (C) Osteomed drills; arrows indicate damaged bone at interface between the cutting tool and alveolar bone. Adjacent transverse tissue sections stained using picrosirius red, an indicator of collagen organization, then visualized under polarized light from osteotomies produced by (D) KLS, (E) Twist, and (F) Osteomed drills. Near-adjacent tissue sections immunostained with Runx2 when osteotomy sites were prepared using (G) KLS, (H) Twist, and (I) Osteomed drills. Near-adjacent tissue sections stained using terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL), an indicator of apoptotic osteocyte distribution, in osteotomies produced by (J) KLS, (K) Twist, and (L) Osteomed drills; arrows indicate the approximate extent of the radial zones of cell death. os, osteotomy site. Scale bars = 100 µm.

Figure 3.

Computational modeling and biological analyses demonstrate alignment in the response of human and rodent tissues to implant site preparation. On postosteotomy day 0.5 (POD0.5), terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL)–stained representative transverse tissue sections across osteotomies produced by (A) KLS, (B) Twist, and (C) Osteomed drills in ovariectomy surgery (OVX)/mxM1 rats. (D) Quantification of TUNEL^+ve apoptotic osteocytes, expressed as a function of distance from cut edge of osteotomy. (E) Computational model maps the radial heat distribution from the osteotomy edge. (F) Bone temperature expressed as a function of radial distance from the drill; shaded bars correspond temperature ranges to the distribution of dying osteocytes within specified distances from osteotomy edge. Bone biopsies from 2 patients (G, H) costained with TUNEL and 4′,6-diamidino-2-phenylindole (DAPI) to identify dying and viable osteocytes at the periphery of osteotomy sites; merged images show the distribution of TUNEL^+ve and DAPI^+ve cells at the osteotomy edge. (I) Quantification of TUNEL^+ve apoptotic osteocytes, expressed as a function of distance from osteotomy edge. ab, alveolar bone; os, osteotomy site. Scale bar = 100 µm.

Figure 4.

Implant site preparation creates a radial zone of dead and dying osteocytes, the fate of which can be followed in a rodent model but not in patients. Aniline blue staining of tissue sections from representative osteotomies produced in (A) a patient and (B) an ovariectomy surgery (OVX) rat, collected on postosteotomy day (POD) 0.5. Higher magnification images show areas of micro-damage in (A′) patient and (B′) rat samples. When visualized under polarized light, picrosirius red–stained tissues from (C) a patient and (D) rat illustrate a similar organization to the collagenous-rich mineralized matrix at the osteotomy edge. Tartrate-resistant acid phosphatase (TRAP) staining for osteoclasts is similar between (E) a 67-year-old patient and (F) an OVX rat. (G) In rat tissues collected on POD7, analyses for osteocyte viability using 4′,6-diamidino-2-phenylindole (DAPI); (H) osteoclast activity using cathepsin K immunohistochemistry and (I) TRAP activity; (J) mitotic activity using proliferating cell nuclear antigen (PCNA) immunohistochemistry; and (K) osteogenesis using Osterix immunohistochemistry, (L) picrosirius red staining, (M) alkaline phosphatase (ALP) activity, and (N) aniline blue staining. Abbreviations as noted previously. Scale bars in A, B = 200 µm; E–N = 100 µm.