The purpose of this study was to investigate the dose-response relationship of alendronate-impregnated allograft with respect to the amount of graft bone and the amount of new bone or total bone after 12 weeks in a goat bone chamber model. Also, the effects of adding DBM and the local application of antibiotics (cefazolin) were tested.
We showed a dose-response relation for local application of alendronate with respect to the amount of new bone as well as the amount of graft bone present in the bone chamber after 12 weeks. Decreased implant fixation has been observed with higher doses of bisphophonates (by blocking bone metabolism completely) [16
]. Therefore, an optimum dose regarding bone resorption is essential, since it yields a positive balance between allograft resorption and the net amount of newly formed bone which results in improved fixation.
Our results show that with increasing alendronate concentration, graft resorption decreased and the amount of necrotic graft bone, left after 12 weeks, increases. This is in accordance with the study of Jakobsen et al. [20
], in which another bisphosphonate, zoledronate, was used. They also studied different bisphosphonate concentrations, however their volume fractions of allograft bone were much higher compared to our study. Our volume fractions of allograft bone were 10.9% (± 8.8) for the lowest alendronate group to 32.8% (± 8.7) for the highest alendronate group, while in the study of Jakobsen, graft bone volume fractions started with mean values above 30%. These differences in results might also be explained by the rinsing method used (3 min vs 10 min) or the fact that Jakobsen had a mechanically loaded situation compared to our unloaded situation. Agholme et al. [17
] also used alendronate locally (2 mg/mL) and found volume fractions of allograft bone compared to ours. In contrast to our study, they did not find any differences between the amount of new bone between a regular dose (with rinsing after impregnation) of bisphosphonate and an overdose amount (without rinsing the graft after impregnation). The volume fractions of new bone were comparable to the mean volume fractions of the 4 different doses in our study, although we did find differences between the doses. They stated that their results may be explained by the four-week study period in their experiments compared to the 6 weeks period in other studies (and thus the 12 week period in our study), that there was not enough time to fully resorb all allograft bone behind the bone ingrowth frontier.
Zoledronate is used in several studies for local application of bisphosphonate and is a more potent bisphosphonate compared to alendronate [38
]. Differences in graft volume fractions between local application of alendronate and zoledronate might be explained by this. Alendronate was chosen in this study as bisphosphonate since it has been used as osteoporosis prophylaxis treatment for almost two decades [39
The amount of new bone in our experiment showed an optimum at a dose of 1 mg/mL alendronate. The amount of new bone was significantly higher compared to control and also when compared to lower and higher alendronate concentrations.
A possible explanation for the difference in newly formed bone between control and BIS groups is that in the control groups the newly formed bone is rapidly remodeled by osteoclasts. This yields a net low amount of new bone, which is entirely the result of osteoclastic activity and might explain the differences observed. Jakobsen et al. stated that this effect could be explained by the preserving effect of the bisphosphonate on the allograft, thereby prolonging the osteoconductive effect [20
An enhancing effect of application of bisphosphonate locally on the amount of new bone has also been observed in other studies [9
] and are confirmed by some in vitro
]. However, it has not been shown in vivo
. The declination in the amount of newly formed bone at the highest alendronate-group,
compared to lower alendronate groups, might be explained by the toxicity of bisphosphonates on bone tissue. At a high dose, bisphosphonates have been shown to be toxic to osteoblasts [44
]. Another explanation might be the lack of space in the bone conduction chamber. The necrotic bone is not resorbed and therefore, no room is left for the formation of new bone.
We chose cefazolin as antibiotic. This is the antibiotic of first choice for orthopaedic surgery in The Netherlands. According to Edin et al. [36
] local levels of cefazolin 200 μg/mL decrease cell replication, but levels of 100 μg/mL do not affect the replication of osteoblasts. The amount of cefazolin used in this study is well above the MIC for S. Epidermidis
, and did not affect bone remodelling since no differences between the control group and the allograft group impregnated with cefazolin could be observed between for % graft bone volume or % new bone volume. However, since our study was powered to show differences, not equality, this result should be interpreted cautiously. Taking into account the half-life of cefazolin, it will stay above MIC - and be effective- for at least 8 to 10 h which is enough for prophylaxis. Cefazolin is completely eluted from the bone chips after three days. No subinhibitory amount of the drug is left behind which can induce resistancies and therefore, cefazolin is an attractive choice for local prophylaxis [29
No effect of the addition of DMB to allograft bone has been found in this study. In addition, histology shows similar resorption characteristics for DBM and allograft groups. The release of growth factors by osteoclastic resorption of allograft bone or DBM is probably quite similar. In addition, when combining DBM, alendronate and allograft bone, no additional bone was formed. Our results confirm several studies [46
], although other studies did find an enhancing effect of DBM on bone formation [48
]. Bae et al. [50
] studied different DBM products and found a higher variability in concentration of bone morphogenic proteins among three different lots of the same DBM than among the different DBM products of different companies. Although our study pooled DBM of two goats (and therefore pooled two different lots), we did not find any effect of the addition of DBM to allograft bone.
Before implantation of the bone chamber we impacted the allograft bone to simulate the clinical situation as much as possible. However, caution should be taken when extrapolating the results. This bone chamber model in goats is un-loaded and therefore quite different from the loaded conditions as in a clinical situation. Also, no cement has been applied, as is done in a cemented revision hip arthroplasty with bone impaction grafting. In addition, a limitation of this study is the lack of biomechanical testing that would determine whether the groups treated with alendronate would be structurally stronger than the control group.
Basic principle of bone impaction grafting is that osteoclasts will resorb bone and osteoblasts will form new woven bone. Micromotions during normal gait cycles induce a rapid osteoclastic response. This too fast bone resorption might lead to intial instability of the implant, periprosthetic osteolysis and later implant migration [6
]. Local application of bisphosphonates has a clear influence on the osteoclastic activity, but little is known of the effect on implant migration and the occurrence of micromotions surrounding the implant.