After implant insertion, cover screws were secured and the surgical wounds were closed by a resorbable periosteal-muscular inner suture, followed by an external cutaneous 2-0 silk suture.

Each animal underwent systemic antibiotic therapy for 5 days with 8 ml long-acting Clamoxil (Pfizer Limited, Sandwich, USA). After surgery, animals received appropriate veterinary care and were allowed free access to water and standard laboratory nutritional support throughout the trial period.

The sheep were sacrificed 2 months after implantation by an overdose of sodium thiopental (Thiopental, Höchst, Austria).

Bone blocks containing the implants were retrieved from each side of the iliac crest. Each implant was fitted with a one-piece 11-mm straight abutment.

The bone blocks were fixed on a customized loading device to measure implant secondary stability according to a previously described technique [19]. A digital force gauge (Akku Force Cadet, Ametek, Largo, USA) and, on the opposite side, a digital micrometer (Mitutoyo Digimatic Micrometer, Kawasaki, Japan) were used to measure implant micromotion during load application. Horizontal forces of 25 N/cm were applied onto the abutment of the implant perpendicularly to the major axis, and the lateral displacement was measured by the digital micrometer 10 mm above the crest. This parameter represents the “value of the actual micromotion” (VAM) as previously published [20] and validated [21].

Resonance frequency analysis was assessed at the time of animal sacrifice (after 2 months of healing) with the latest Ostell device (Osstell AB, Göteborg, Sweden). The implant stability quotient (ISQ), which ranged between 0 and 100, was recorded.

Removal torque value (RTV) was measured at the time of animal sacrifice (2 months after implantation) after VAM measuring procedures. The RTV was evaluated and recorded for each implant using a digital hand-operated torque wrench (Tonichi STC400CN) by unscrewing the implants until interfacial failure occurred. The digital torque wrench automatically registered the peak removal torque value on the digital display. After the initial interface detachment, the implants were repositioned back in their initial position as accurately as possible and processed for histologic analysis. Although the interfacial detachment created an artifact at the interface, its analysis would still be reliable according to Sennerby et al. [22], who used a similar procedure to study the morphology of the bone-metal rupture.