Baseline factors associated with complete root coverage at 6 months after matrix-assisted treatment of multiple gingival recessions: a post Hoc exploratory analysis of randomized trial data

Complete root coverage (CRC) after treatment of multiple gingival recession defects may vary among treated sites, even when the same surgical protocol is used. This post hoc exploratory secondary analysis of randomized trial data aimed to evaluate baseline clinical factors associated with complete root coverage at 6 months after modified coronally advanced flap (MCAF) plus volume-stable collagen matrix (VCMX), with or without injectable platelet-rich fibrin (i-PRF). Patients with multiple gingival recession defects were treated with MCAF plus VCMX, with or without i-PRF. CRC was analyzed at the site level at 6 months. Because multiple treated sites were nested within patients, logistic generalized estimating equations with patient-level clustering were used. Univariable clustered models were followed by a parsimonious adjusted model including intervention group, baseline gingival recession depth, baseline gingival thickness, and age. Baseline probing pocket depth was evaluated in an additional exploratory adjusted model. The analysis included 79 recession sites from 24 patients, with 37 sites treated using MCAF + VCMX and 42 using MCAF + VCMX + i-PRF. Complete root coverage was achieved in 20 sites overall, including 7 sites in the MCAF + VCMX group and 13 in the MCAF + VCMX + i-PRF group. In the univariable clustered analysis, greater baseline gingival recession depth was associated with lower odds of complete root coverage, whereas baseline probing pocket depth was positively associated. In the adjusted model, baseline gingival recession depth remained associated with lower odds of CRC (adjusted OR 0.269, 95% CI 0.099–0.730; p = 0.010), while the association with baseline probing pocket depth became borderline and imprecise. Greater baseline gingival recession depth was associated with lower odds of complete root coverage at 6 months. Because only 20 CRC events were observed, the adjusted GEE estimates were imprecise and should be interpreted as exploratory. The findings should not be considered predictive or as evidence of superiority of either treatment protocol.