Material and methods

Thirty-two fresh sheep mandibles were collected for miniscrew insertion. All sheep were adult males about the same size. None had any remarkable diseases. All were obtained from a slaughterhouse in Tehran, Iran. The review board of the School of Dentistry of Shahed University approved this study plan.

The mandibles were denuded from the soft tissues and stored in a solution containing equal amounts of normal saline solution and 70% ethanol at 4°C. This is a suitable method for preserving the mechanical and elastic properties of bones. The tests were done within the first 24 hours after the animals’ deaths. The edentulous segment between the incisors and the premolars was selected for implant insertions. Segments with similar shape and morphology of the buccal surfaces were selected. Then 64 segments of bone were cut with at least a length of 20 mm. One implant was inserted per segment. These segments were embedded in resin (ProBase; Ivoclar Vivadent, Schaan, Liechtenstein) except for the buccal surface. Resin blocks with dimensions of 2 × 5 × 2.8 cm were prepared. To avoid overheating the bone, resin block curing was done under water. Eight blocks were assigned for each insertion angle: half for the pullout tests and the other half for the shear tests. A custom holder was built for grabbing the blocks. The walls of the blocks were trimmed to be parallel and flat. A custom x-y table was mounted on the custom holder. For precise insertion, 4 metal plates were made with 4 guiding-canal angles (30°, 45°, 60°, and 90°). The width of the metal plate guiding canals was constructed almost equal to the width of the handhold screwdriver head, and the length was 10 mm.

Pilot drillings were carried out at a depth of 1 mm in the center of the buccal cortex with the predetermined angle for each group test. The miniscrews underwent only 1 test. A total of 64 miniscrews (16-G2-006, Dual-Top G2; Jeil Medical, Seoul, Korea) with the same size (1.6 × 6 mm) were placed in the prepared bone segments at angles of 30°, 45°, 60°, and 90° to the buccal surface. The design test matrix consisted of 8 groups of 8 implants each: 4 groups for pullout tests and 4 groups for shear tests. Using the metal plates, we placed the implants with a handhold screwdriver (111-010; Jeil Medical) and a driver shaft (113-GD-I0l; Jeil Medical). The miniscrews were driven in until the soft tissue cap reached the bone surface. Upon completion of the insertion process, a conventional dental radiograph was taken to (1) measure the cortical bone thickness and (2) evaluate the insertion angle to the buccal cortex and confirm that the implant did not contact the lingual cortex. The thickest part of the buccal cortical plate on each radiograph was considered an indicator of the quantity of bone. It was necessary to compensate for enlargement of the cortical thickness. This was done by measuring the length of the miniscrew in the radiograph and its actual size (8 mm). A simple proportional relationship was then set up:

The insertion angle was measured at the intersection of 2 lines: the longitudinal axis of the miniscrew and another line tangential to the buccal surface.

The pullout (tensile) and shear tests were done on the inserted implants using a universal testing machine (Z020; Zwick/Roell, Ulm, Germany). To apply the desired load to the screws, 4 custom miniscrew holders were designed (1 for each insertion angle) to grab the miniscrews. These holders comprised 2 parts. The upper parts were attached to the load cell, and the lower parts were integrated with miniscrews ( Fig 1 , A ). The lower part was a cylinder with an internal diameter of 2.91 mm (maximum width of miniscrew head, 2.90 mm). An 0.85-mm hole was drilled in the lower part at 0.5 mm from the base ( Fig 1 , B ). The angle between the upper and lower parts was similar to the predetermined test (ie, when the insertion angle was 30°, the angle between the upper and lower parts was 30°). So the directions of loading were parallel to the alveolar surface in the shear tests and perpendicular in the pullout tests ( Fig 2 ). The implant heads were placed in the custom miniscrew holder and secured by passing an 0.8-mm hard stainless steel wire (Dentarum, Ispringen, Germany) through the implant head hole and the lower part hole. All tests were done with a 2-N preload to minimize any play between the implant and the holder. A custom x-y-z table was mounted on the lower jaw of the testing machine to check the accuracy and to recognize any deviation during testing. All tests were done at room temperature (22°C). During testing, the bone blocks were kept moist with a wet paper wrapped around them.

Custom implant holder: A, upper part is attached to the testing machine; B, upper and lower parts for the 60° pullout test.

A, In the shear tests, the direction of force (anteroposterior) was parallel to the alveolar surface and had a different plane to miniscrew angulation (occlusogingival direction); B, in the pullout tests, the direction of force was 90° to the alveolar surface.

The tests began with a crosshead speed of 0.05 mm per second. The tests were discontinued when the implant failed. Full pullout of a miniscrew was considered the failure point. For each test, the maximum force before the failure point was recorded as the primary stability. The test results were collected digitally for all samples. The second author (A.G.) prepared the samples and performed the tests.