Premature Bone Resorption in Vertical Ridge Augmentation: A Systematic Review and Network Meta-Analysis of Randomised Clinical Trials

OBJECTIVES: To systematically appraise the evidence on premature ridge resorption following various vertical ridge augmentation (VRA) techniques in healthy adult patients undergoing staged VRA procedures. The study aimed to identify VRA techniques resulting in the least premature bone resorption and to rank them using Bayesian Network Meta-Analysis (NMA).

MATERIAL AND METHODS: Searches were conducted in six databases to identify randomised clinical trials (RCT) comparing staged VRA techniques with a minimum of 3 months follow-up. Relative premature bone resorption (PBR%) overall (primary) and in sites with uneventful versus complicated healing and need for additional bone grafting (NAG) (secondary) were chosen as outcomes. The risk of bias and certainty in evidence were assessed using Cochrane RoB 2.0 and GRADE tools. Bayesian models estimated treatment effects and rankings.

RESULTS: Ten RCTs, involving 220 participants and 236 defects, were included. Nine RCTs reported mean PBR%, with a range from 6% to 44%, averaging 26%. Seven treatment groups were evaluated: onlay, onlay + barrier, inlay, guided bone regeneration, distraction osteogenesis (DO), tissue expansion + tunnelling (TET), and cortical tenting. Eight RCTs, involving 160 participants and 176 defects, contributed to the NMA. Compared to onlay, all groups had lower mean PBR%. Inlay had the highest probability of being ranked first (Pr = 0.55), followed by DO (Pr = 0.27) and TET (Pr = 0.15). Healing complications significantly increased PBR% (MD 10%, 95% CrI 4.4-15.7).

CONCLUSION: VRA techniques preserving the periosteum (inlay, DO, and TET) exhibit less PBR compared with other techniques. When techniques involve full flap elevation, clinicians should anticipate volume loss at re-entry and consider greater grafting volumes to offset PBR. PROTOCOL REGISTRATION: PROSPERO ID: CRD42023394396; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=394396.