Hierarchical Variance in Early Marginal Bone Change in Monolithic Zirconia Full-Arch Implant Prostheses: A Multilevel Prospective Analysis

A clinically unresolved question in full-arch implant rehabilitation is whether early peri-implant bone adaptation is primarily a local (implant-level) phenomenon, a shared within-arch response, or a patient-level characteristic-information is unavailable from conventional mean-based outcome reporting. This prospective secondary analysis quantified the hierarchical distribution of peri-implant marginal bone level change (ΔMBL) in monolithic zirconia full-arch prostheses directly connected to multi-unit abutments during the first year of function. Unlike mean-based endpoints, this approach identifies whether early bone adaptation variability originates primarily at the implant, prosthetic-arch, or patient level. Of 308 implants placed in 40 completely edentulous patients rehabilitated with 49 prostheses, 2 implants failed before delivery of the definitive prosthesis and were excluded from radiographic analysis, leaving 306 implants available for multilevel analysis. Implant-level ΔMBL was analyzed using an unconditional three-level linear mixed-effects model. Implant survival was 99.2%, prosthetic survival was 100%, and mean ΔMBL at 12 months was 0.38 ± 0.27 mm. Variance partitioning showed that 60.8% of total variability occurred at the implant level, 28.4% occurred at the patient level, and 10.8% occurred at the prosthetic-arch level, providing a framework for identifying at which level of the restorative system future clinical interventions may have the greatest impact. Findings should be interpreted within the context of a secondary analysis with a 12-month follow-up and no explanatory covariates; longer-term comparative studies with explanatory multilevel designs are required to confirm and extend these observations.