Herbal interventions for the management of type 2 diabetes mellitus: a narrative review of clinical evidence, mechanisms, and translational challenges

Vollständiger Abstract

Worum geht es in dieser Arbeit?

Abstract Background Type 2 diabetes mellitus (T2DM) is a major and growing global health burden, and many people use herbal products alongside conventional therapy. Although numerous clinical trials and evidence syntheses have evaluated herbal interventions, the certainty, reproducibility, safety, and translational relevance of these findings remain uncertain. Methods This critically appraised narrative review synthesised English-language clinical evidence published from January 2020 through 12 August 2026, with emphasis on randomised controlled trials (RCTs), systematic reviews, meta-analyses, and umbrella reviews. PubMed, Scopus, Web of Science, Embase, and the Cochrane Library were searched using predefined combinations of T2DM, individual herbal agents, and evidence-synthesis terms. Reference lists were hand-searched. Because the original review was not prospectively registered as a systematic review and database export counts were not retained, record-level screening numbers are not retrospectively fabricated; the revised manuscript therefore reports the selection process transparently and qualitatively. Results Published meta-analyses reported reductions in HbA1c for several interventions, including berberine (mean difference [MD] − 0.73% across 46 RCTs), whole-plant Berberis preparations (MD − 0.30%), fenugreek (MD − 0.54%), and curcumin (MD − 0.54% across 18 RCTs, n = 1382). Aloe vera leaf gel and psyllium had among the largest point estimates in an overview of meta-analyses, but their confidence intervals were wide and the underlying interventions were heterogeneous. Gymnema sylvestre showed high statistical heterogeneity, whereas evidence for cinnamon and bitter melon was inconsistent. The 2025 umbrella review found moderate-to-high certainty for turmeric and ginger but low or very low certainty for several other interventions. Across the literature, small sample sizes, short follow-up, inadequate blinding in some trials, publication bias concerns, and variability in botanical preparations limit confidence in pooled estimates. Conclusion Herbal interventions may provide adjunctive glycaemic benefit for selected patients, but current evidence does not support replacing established pharmacotherapy. HbA1c improvements should not be interpreted as evidence of reduced cardiovascular events, nephropathy, retinopathy, mortality, or long-term disease progression. Translation into routine care requires standardised products, better herb–drug interaction studies, longer and adequately powered trials, robust product authentication, and clinically relevant outcome assessment. Clinical trial number Not applicable.