The purpose of this study was to investigate the release kinetics of Zoledronic acid (ZOL) incorporated in a poly(D, L-lactide) coating (PDLLA) and its effect on the osseointegration of implants, compared to a systemic ZOL application.
Release kinetics of ZOL out of the coating showed an initial peak of approximately 90% in release of C14-labled ZOL from PDLLA coated K-wires within the first 24 hours. The biomechanical and histological analyses revealed a significant higher strength of fixation in the group with systemic ZOL compared to PDLLA. However, no enhancement of osseointegration in groups treated with ZOL-coated implants in comparison to the controls could be shown.
These results are in contrast to other studies, showing an enhancement of osseo-integration or reduced implant migration after systemic [12
] or local [1
] application of Bisphosphonates (BPs). Especially the BP ZOL, as used in this study, has been shown to be among the most potent of its class [37
]. However, existing studies vary in different aspects, such as additional use of bone compaction [2
], the used BPs [7
], the experimental model [14
], animal species [2
] implants [2
] and application methods [7
]. Among implant coatings, fibrinogen [4
] and especially hydroxyapatite (HA) [20
] have recently been used successfully to improve implant fixation by local application of BPs. Studies using ZOL locally were able to show an enhancement of peri-prosthetic bone quality and osseous integration by ZOL-coated implants with a similar animal model [20
] or a significant enhancement of bone apposition and mean amount of femoral canal filling even after one year compared to controls [21
]. Also a dose depending improvement of histological and biomechanical results by ZOL-application has been proved [14
But there are also some studies where local BPs did not improve the bone-implant integration or even impaired it. After showing a benefit of local Alendronate in the fixation of porous-HA-coated implants [2
] Jakobsen et al. detected a decrease in implant fixation when using soaked morselized allograft with the same dose and BP [29
]. This effect, also shown in other studies with Pamidronate [28
], could be due to the combination of densely compacted bone and BP [7
]. But even in other study designs locally applied BPs did not always lead to significantly increased bone/implant contacts of BP-coated implants [22
] nor did the additional application of BP enhance the biomechanical properties of coated implants [23
Regarding systemic delivery, BPs have been shown to increase peri-implant bone density and implant-bone contact ratio in animal [39
] and clinical [9
] studies. Even though the same i.v. ZOL concentration (0.1 mg/kg) was used as by Yu et al. [39
], no improvement in osseointegration was detected compared to the groups with PDLLA/ZOL or uncoated implants, except for a significant biomechanical enhancement in comparison to the PDLLA group. A possible reason could be the chosen time point of the application, being one week after surgery for Yu et al. vs. immediately during surgery in the presented study. This might have led to a stronger impairment of bone catabolism.
There are different explanations for the positive effects of BP in osseointegration. A reduction of implant migration by BPs, seen by Hilding and Aspenberg [19
] was probably due to the inhibition of the resorption of periprosthetic necrotic bone [18
Also bone formation around screws coated with fibrinogen, Pamidronate and Ibandronate was supposed to be based on reduced bone loss due to BP with the retained bone serving as scaffold for new bone cells [38
]. Thus the enhancement of periprosthetic stabilization by BP appears to depend on the contact to surrounding bone and maybe even a press fit position [28
]. Especially in a press fit situation the bone next to the implant may be necrotic and prone to resorption by osteoclasts. This resorption could be inhibited by BP, leading to an enhanced implant fixation.
In the present study, however, the implant fixation was not press fit in the medullary canal but in the cortex of the insertion point. Therefore neither local nor systemic inhibition of osteoclasts by ZOL might have supported additional implant ingrowth.
Since the time point of BP application seems to be decisive for its local effect on bone cells [39
], data about release kinetics
is important for a better prediction concerning the effect of the locally released BPs on the osseointegration of implants. Few other studies dealing with local release of BPs showed data for the individual specific elution kinetics
]. Regarding the used coating, the present work is the first describing the release kinetics
of ZOL out of the PDLLA coating [15
However, even though the detected release kinetics with an initial peak confirms previous findings for the PDLLA coating [24
] and though comparable kinetics between a phosphate buffered saline solution and cell culture medium have been shown [15
], the here obtained releasing curve of ZOL cannot reflect the true release dynamics of intramedullary implants. Future studies will also have to investigate if the effect of the substance on bone cells could be improved by modification of the coating and variation of the release resulting in a slow sustained or delayed release [25
]. In this context it could be tried to use C14-labled ZOL with autoradiographic analysis for local in vivo detection of released ZOL [16
] as systemic detection would not be promising due to the high affinity of BPs to bones [13
There are some limitations of this study. Among those should be seen the single time point (56 days), as no possible effect over time could be detected, even though other experiments have shown that osseous integration of implants was completed after six weeks [36
]. Further limiting was the fact that the chosen implant model was not weight bearing and thereby effects of direct load transfer were not addressed.
Retrospectively, the use of a micro-CT with its possibility of a 3-D detection of newly formed bone or bone/implant contacts should be seen as best method for this purpose and would have avoided the danger of harming fragile structures like trabeculae by histological preparation. Another shortcoming was the lack of direct determination of the bioactivity of C14-ZOL after coating and release from PDLLA. However, previously published data has shown bioactivity of PDLLA-released ZOL on human bone cells [31