Purpose: To evaluate the effects of vertical compressive cyclic load on abutment screws and the stability of screw-retained splinted fixed dental prostheses supported by bone-level nonengaging and partially engaging abutments.

Materials and methods: Ten standardized two-unit prostheses integrated with nonengaging and partially engaging abutments were 3D-printed and cast in nonprecious metal and attached to two implant analogs with conical connections angled 15 degrees toward the vertical axis in a resin bock. The common path of insertion for the prosthesis with partially engaging abutments to angled implant analogs was created by selectively reducing the abutment surface that obstructed the implant housing. A 500-N average load was applied to each sample while oscillating 30 degrees from the vertical axis at 60 Hz to failure. Prosthesis stability was measured by deflection from the initial position using a linear displacement measuring device. Sample groups were assessed using the independent-samples t test and one-way repeated-measures analysis of variance (ANOVA) with post hoc tests.

Results: The mean cycles to failure for nonengaging and partially engaging abutment groups were 27,180 ± 29,420 and 457,890 ± 265,734, respectively. Failed nonengaging samples had 9 out of 10 abutment screws fracture inside the implant housing but with minimal wear to the top of the implant analog. For partially engaging samples, 8 out of 10 implant housings and 7 out of 10 abutment screws fractured at test endpoints. There was no difference in initial deflections between groups. Prostheses supported by nonengaging abutments failed before deflection measurements could be taken after loading. Prosthesis displacement of partially engaged abutments decreased significantly from the initial position after 300,000 load cycles.

Conclusion: Under the experimental conditions, screw-retained splinted fixed dental prostheses supported by partially engaging abutments are 17 times more stable than prostheses supported by nonengaging abutments. Abutment screw fractures are the most prevalent mode of failure for nonengaging abutments at significantly lower cycles with minimal wear on implant analogs. Partially engaging abutment groups failed from implant housing and abutment screw fractures at higher cycles. Loading appears to concentrate preferentially on the medial side on all angled implant components. Selective removal of the abutment obstruction allows a common path of insertion for multiple implants and partial engagement to implant housings.