Comparative efficacy of plant extracts as adjuncts to scaling and root planing for periodontitis: a systematic review and network meta-analysis of randomized controlled trials

Plant extracts may be useful adjuncts to scaling and root planing (SRP) in the non-surgical treatment of periodontitis. However, evidence is largely limited to pairwise comparisons. We conducted a frequentist network meta-analysis (NMA) to compare plant-extract adjuncts across major periodontal outcomes. We searched major electronic databases and trial registries from inception to December 30, 2025, for randomized controlled trials (RCTs) evaluating plant extracts as adjuncts to SRP in adults with periodontitis. Primary outcomes were probing pocket depth (PPD) and clinical attachment level (CAL); secondary outcomes were gingival index (GI) and plaque index (PI). A frequentist random-effects NMA using Stata 18.0 estimated mean differences (MDs) with 95% confidence intervals (CIs). The surface under the cumulative ranking curve (SUCRA) was used for ranking. Positive SRP-referenced MDs favoured the plant-extract adjunct. Thirty-eight RCTs were included, and all contributed to at least one quantitative synthesis. For PPD, CLE (MD 1.13 mm, 95% CI 0.44 to 1.82; moderate certainty; 1 RCT) and Garcinia mangostana (MD 1.00 mm, 95% CI 0.43 to 1.56; moderate certainty; 2 RCTs) showed the largest estimated additional reductions. Lycopene, tea extract, and curcumin also showed statistically detectable reductions, although the effects for tea extract and curcumin were below 0.5 mm and were supported by low-certainty evidence. No adjunct showed a statistically detectable benefit for CAL. Emblica officinalis, curcumin, and tea extract showed reductions in GI, while curcumin and tea extract showed reductions in PI. SUCRA rankings were considered descriptive and were interpreted alongside effect estimates, certainty, and the number of contributing studies. Trial-level safety information was available in 21 of 38 studies and was insufficient for quantitative comparison. No intervention-related serious adverse event was explicitly reported, although adverse-event and withdrawal reporting was generally incomplete. Some plant-extract adjuncts provided additional short-term improvements in periodontal outcomes. However, no intervention can currently be considered optimal because rankings were largely based on sparse, indirect evidence of low-to-moderate certainty. Larger head-to-head RCTs using standardised formulations, clinically meaningful endpoints, comprehensive safety reporting, and longer follow-up are required. PROSPERO CRD420261336531.