Development of an advanced numerical model of heat transfer in a bone-implant system

OBJECTIVES: Implant removal can be required in case of malpositions, biological or mechanical failures. This procedure, performed with trephine burs, piezosurgery or counter-torque ratchets, often results in bone defects requiring corrections with advanced regenerative techniques. Few case reports describe implant removal after weakening the bone-implant interface by applying a localized heating to the fixture: however, this procedure must be standardized to obtain a predictable result, avoiding extensive bone damage. Aim of this preliminary study is to create a high-fidelity numerical model of the bone-implant system, to be validated by in vitro and in vivo tests, for predicting the most appropriate parameters necessary for a minimally invasive implant removal using heating. MATERIAL AND METHODS: A 3-D virtual advanced numerical model was created using ANSYS CFX, in order to simulate the behavior of a bone-implant system under thermal variations. Various parameters can be adjusted to set the model: heating modalities (temperature and/or heat flux, time and source) and system characteristics (implant geometry and material, type of bone). In vitro tests were performed using titanium implants (Premium, Sweden & Martina, Italy - 3.85 x 8.5, 10 and 13 mm). Implants were inserted in bovine bone (stabilized at 37 ± 0.5 °C), and then heated using an endodontic heat carrier (System B, Sybron Endo, USA). Bone temperatures were recorded using thermocouples at different depths (1 mm, middle and apex of implant) and at different distances from the implant [0.5 mm (T1), 1 mm (T2), 1.5 mm (T3) and 2 mm (T4)]. Non parametric data were analyzed using Friedman and Wilcoxon tests. RESULTS: The heat carrier, set at 100 °C, was placed in contact with the inner part of the implants for 140 sec. The threshold temperature of 47 °C was reached in T1 after 65.2 ± 11.7 sec, in T2 after 89.3 ± 29.1 sec, in T3 after 110.9 ± 29.6 sec, in T4 the temperature of 47 °C was never reached within 140 sec. Differences between T1, T2, T3 and T4 resulted statistically significant (p<0.05). Implant length and distance from heat source are factors influencing temperature variations. CONCLUSIONS: From the results of this preliminary study, it seems possible to produce a controlled thermally-induced injury to the bone-implant interface, allowing for a minimally invasive implant removal. Various parameters (implant geometry and material, bone characteristics) must be considered to set temperature and heating duration. In vivo histologic studies are necessary to analyze quality and quantity of bone-implant interface modifications after heat-induced injury, before starting any clinical application.