Reporting quality and evidence support in randomized controlled trials of Bupleurum-based herbal medicine formulas...

Bupleurum chinense DC (Chaihu) has been used extensively in traditional Chinese medicine (TCM), where it is traditionally prescribed to harmonize "Shaoyang" disorders and regulate liver function-concepts historically associated with neurobehavioral imbalance and convulsive symptoms. Classical formulas containing Bupleurum chinense, such as Xiao Chaihu Tang and Chaihu Shugan San, have long been applied in the management of epilepsy. This longstanding use provides an important ethnopharmacological context and supports the continued clinical evaluation of Bupleurum-based herbal formulations for epilepsy.

This study aimed to systematically assess the clinical outcomes of Bupleurum-based herbal formulas for epilepsy through meta-analysis, followed by an evaluation of the reporting quality and evidence base of the associated randomized controlled trials (RCTs).

A comprehensive search was conducted across four databases (PubMed, Sinomed, Wanfang and CNKI) from their inception to July 2025 to identify RCTs investigating Bupleurum-based CHM formulas in epilepsy. Reporting quality was evaluated using the CONSORT 2010 and CONSORT-CHM Formula 2017 checklists. A meta-analysis was performed on four primary/secondary outcomes: seizure frequency, seizure duration, overall clinical efficacy, and adverse events. The risk of bias was assessed using the Cochrane risk of bias tool.

A total of 33 RCTs involving 2464 patients with epilepsy met the inclusion criteria. The overall reporting quality was suboptimal (mean CONSORT score: 14.90 ± 3.37; CONSORT-CHM score: 14.42 ± 3.36). Key deficiencies included inadequate descriptions of randomization methods (allocation concealment reported in only 6%) and lack of blinding (no mention of blinding for participants/personnel). Meta-analysis of the 33 RCTs demonstrated that Bupleurum-based formulas significantly outperformed antiepileptic monotherapy on all key efficacy outcomes: overall clinical efficacy was more than tripled (OR = 3.49, (95% CI: 2.73-4.47), p < 0.001; I² = 0%), driven by a substantial reduction in seizure frequency (SMD = -2.70, (95% CI: -3.66 to -1.75), p < 0.001, I² = 97.1%, p < 0.001) and a parallel reduction in seizure duration (SMD = -3.56, (95% CI: -5.02 to -2.10), p < 0.001, I² = 95.5%, p < 0.001). Safety profiles were comparable between the herbal and control groups, with no significant difference in adverse event rates (OR = 0.82, (95% CI: 0.55-1.22), p = 0.33), although only 11 of the 33 trials systematically reported adverse events.

Bupleurum-based herbal formulas demonstrate clinically significant benefits in reducing seizure frequency and duration while improving overall treatment response. However, the evidence is undermined by poor reporting quality and methodological flaws. Future trials must rigorously adhere to the CONSORT and CONSORT-CHM guidelines to enhance the validity and translational potential of the findings.