ZrO Summit 2025, Group 1, 'Survival and Clinical Performance of Zirconia Compared to Titanium Implants: A Systematic...

To compare survival and clinical performance of zirconia versus titanium dental implants, with meta-analyses stratified by follow-up period and, where feasible, by study design.

This PRISMA-compliant review was registered in PROSPERO (XXXX). MEDLINE, Embase, Scopus, CENTRAL, and Web of Science (until May 2025), along with citation tracking, were searched for comparative human studies. The primary outcome was marginal bone loss (MBL); secondary outcomes were implant survival, probing depth (PD), keratinized tissue width (KTW), Pink Esthetic Score (PES), and patient-reported outcomes (PROMs). Random-effects models were used (Hartung-Knapp); heterogeneity was assessed with I²/Q. Subgroups contrasted one-piece versus two-piece zirconia implants. Prespecified sensitivity analyses included risk-of-bias restriction and leave-one-out influence checks. Publication bias and meta-regression analyses were not undertaken (<10 studies per outcome).

Nine studies (six randomized controlled trials, two prospective cohorts, and one cross-sectional) involving 323 patients and 435 implants (222 zirconia and 213 titanium) met the inclusion criteria. Outcome-specific meta-analyses utilized only studies reporting the relevant endpoint (typically 2-6 studies at 1 year and 2-4 studies at ≥5 years). At 1 year, RCTs showed no difference in survival (risk ratio [RR]:0.98; 95% CI: 0.94-1.01; I²=0%). In contrast, observational cohorts were imprecise and heterogeneous (RR 0.78; 0.24-2.57; I²=91.9%). Five RCTs indicated slightly greater MBL around zirconia (mean difference [MD] 0.16 mm; 0.02-0.29; I²=0%), whereas two observational studies suggested a non-significant opposite trend (MD 0.11 mm; -0.73 to 0.51). No significant differences were observed for PD, KTW, PES, or PROMs (2-4 studies per outcome). Subgroup analysis suggested a potential difference in 1-year survival for two piece zirconia (RR 0.94; 0.91-0.97; I²=0%), while one-piece showed no difference (RR 0.99; 0.94-1.03; p<0.001 for subgroup difference). At 3 years, survival did not differ (RR 0.93; 0.86-1.01; I²=0%). At ≥5 years, survival (RR 0.99; 0.91-1.08; I²=0%) and MBL (MD -0.03 mm; -0.14 to 0.08; I²=0%) were comparable; PD slightly favored zirconia (MD 0.29 mm; -0.55 to -0.02; I²=0%), while PES showed no difference. Sensitivity analyses confirmed robustness for ≥5-year outcomes, although the 1-year MBL effect was attenuated after risk-of-bias restriction.

In healthy patients, zirconia and titanium implants demonstrate comparable clinical and radiological outcomes over a follow-up period of up to five years. Evidence remains insufficient regarding long-term outcome for zirconia implants, and part of the available data is based on legacy systems that are no longer on the market.