Comparison of Push-In versus Pull-Out Tests on Bone-Implant Interfaces of Rabbit Tibia Dental Implant Healing Model

Purpose: This study aimed to investigate whether push-in and pull-out tests measure mechanical properties of the bone-implant interface differently, and which test is more sensitive to changes over the healing period. Materials and Methods: Two identical self-threading dental implants (3.3x8.5mm) were placed in medial surface of the proximal condyles of left and right tibias of 20 rabbits (40 implants total). Five rabbits each were sacrificed after 1, 4, 8, and 12 weeks of healing. Push-in test was performed on one side's tibia implant and pull-out on the other side's implant, at a rate of 6mm/min. Primary and secondary implant stabilities and tibia weight were measured on all implants. Results: The push-in test generated significantly higher failure load (p=.0001; 530N vs 279N), lower displacement at failure (p=.0003; 0.436mm vs 0.680mm), and higher interface stiffness (p<.0001; 1,641N/mm vs 619N/mm) than pull-out test. Failure load, stiffness, and secondary implant stability were significantly higher for longer compared with shorter healing periods, while displacement, tibia weight, and primary stability were not. Failure load and stiffness differed significantly for four healing times for the push-in but not for the pull-out test. Failure load was significantly correlated with secondary implant stability for both push-in (r=0.66) and pull-out (r=0.48) tests, but stiffness was significantly correlated with secondary stability only for the push-in test (r=0.72; pull-out test r=0.40). Conclusion: The push-in test appeared more sensitive than pull-out to changes in mechanical properties at bone-implant interfaces during healing in rabbit tibia model.