FitProven
Evidence Breakdown
Based on 8 studies
THE BIGGER QUESTION
Whether this one statement holds is settled above. What to actually do about it is a wider question, weighed across every claim that bears on it.
Evidence map
For & against, at a glance
Meta-Analysis
Pro
Asbaghi O et al. · 2020Clinical Nutrition ESPENSystematic review and meta-analysis of 12 RCTs of oral berberine supplementation (search through July 2019), pooling anthropometric, inflammatory and liver-enzyme outcomes. It is the most favourable of the berberine anthropometry meta-analyses: body weight fell by 2.07 kg (95% CI −3.09 to −1.05, p < 0.001), BMI by 0.47 kg/m² (95% CI −0.70 to −0.23), waist circumference by 1.08 cm (95% CI −1.97 to −0.19, p = 0.018) and CRP by 0.42 mg/L. Liver enzymes did not change significantly (ALT −1.66 IU, 95% CI −3.98 to 0.65; AST −0.87 IU, 95% CI −2.56 to 0.82). Participants were overwhelmingly people with metabolic disease (type 2 diabetes, metabolic syndrome, NAFLD, PCOS) rather than healthy adults seeking weight loss, and the authors called for higher-quality trials.
0.55
Systematic review and meta-analysis of 12 RCTs of oral berberine supplementation (search through July 2019), pooling anthropometric, inflammatory and liver-enzyme outcomes. It is the most favourable of the berberine anthropometry meta-analyses: body weight fell by 2.07 kg (95% CI −3.09 to −1.05, p < 0.001), BMI by 0.47 kg/m² (95% CI −0.70 to −0.23), waist circumference by 1.08 cm (95% CI −1.97 to −0.19, p = 0.018) and CRP by 0.42 mg/L. Liver enzymes did not change significantly (ALT −1.66 IU, 95% CI −3.98 to 0.65; AST −0.87 IU, 95% CI −2.56 to 0.82). Participants were overwhelmingly people with metabolic disease (type 2 diabetes, metabolic syndrome, NAFLD, PCOS) rather than healthy adults seeking weight loss, and the authors called for higher-quality trials.
Design Meta-Analysis (1.0) × quality 0.55 = impact 0.55
View sourceMeta-Analysis
Neutral
Xie W et al. · 2022Frontiers in PharmacologySystematic review and GRADE-rated meta-analysis of 37 RCTs (3,048 patients with type 2 diabetes) on berberine and glycaemic control. Berberine lowered fasting plasma glucose by 0.82 mmol/L (95% CI −0.95 to −0.70) and HbA1c by 0.63 percentage points (95% CI −0.72 to −0.53), both p < 0.001 — a real but sub-metformin effect. Two structural weaknesses dominate: **34 of the 37 trials were conducted in China** (one in Iran), and the GRADE certainty was only *moderate* for FPG and HbA1c (downgraded for risk of bias), *low* for 2-hour postprandial glucose (risk of bias plus publication bias) and *very low* for hypoglycaemia. Body weight was not the outcome of interest here.
0.60
Systematic review and GRADE-rated meta-analysis of 37 RCTs (3,048 patients with type 2 diabetes) on berberine and glycaemic control. Berberine lowered fasting plasma glucose by 0.82 mmol/L (95% CI −0.95 to −0.70) and HbA1c by 0.63 percentage points (95% CI −0.72 to −0.53), both p < 0.001 — a real but sub-metformin effect. Two structural weaknesses dominate: **34 of the 37 trials were conducted in China** (one in Iran), and the GRADE certainty was only *moderate* for FPG and HbA1c (downgraded for risk of bias), *low* for 2-hour postprandial glucose (risk of bias plus publication bias) and *very low* for hypoglycaemia. Body weight was not the outcome of interest here.
Design Meta-Analysis (1.0) × quality 0.60 = impact 0.60
View sourceDouble-Blind RCT
Con
Wilding JPH et al. · 2021New England Journal of MedicineSTEP 1: 68-week double-blind RCT randomizing 1,961 adults with obesity (BMI ≥30, or ≥27 with a weight-related condition) and without diabetes to once-weekly subcutaneous semaglutide 2.4 mg or placebo, both with lifestyle intervention. Mean body-weight change was **−14.9% with semaglutide vs −2.4% with placebo** (difference −12.4 percentage points, 95% CI −13.4 to −11.5; p < 0.001), i.e. −15.3 kg vs −2.6 kg (difference −12.7 kg, 95% CI −13.7 to −11.7). 86.4% of the semaglutide group lost ≥5% of body weight, 69.1% lost ≥10% and 50.5% lost ≥15%, versus 31.5%/12.0%/4.9% on placebo. Nausea and diarrhoea were the commonest adverse events, usually transient and mild-to-moderate; 4.5% discontinued for GI events vs 0.8% on placebo.
0.72
STEP 1: 68-week double-blind RCT randomizing 1,961 adults with obesity (BMI ≥30, or ≥27 with a weight-related condition) and without diabetes to once-weekly subcutaneous semaglutide 2.4 mg or placebo, both with lifestyle intervention. Mean body-weight change was **−14.9% with semaglutide vs −2.4% with placebo** (difference −12.4 percentage points, 95% CI −13.4 to −11.5; p < 0.001), i.e. −15.3 kg vs −2.6 kg (difference −12.7 kg, 95% CI −13.7 to −11.7). 86.4% of the semaglutide group lost ≥5% of body weight, 69.1% lost ≥10% and 50.5% lost ≥15%, versus 31.5%/12.0%/4.9% on placebo. Nausea and diarrhoea were the commonest adverse events, usually transient and mild-to-moderate; 4.5% discontinued for GI events vs 0.8% on placebo.
Design Double-Blind RCT (0.85) × quality 0.85 = impact 0.72
View sourceMeta-Analysis
Con
Amini MR et al. · 2020Complementary Therapies in MedicineSystematic review and meta-analysis of 12 RCTs (849 participants) of berberine and barberry (Berberis vulgaris) supplementation on anthropometric outcomes. None of the three headline measures moved: body weight (WMD −0.11 kg, 95% CI −0.13 to 0.91, p = 0.83), BMI (WMD −0.16 kg/m², 95% CI −0.43 to 0.11, p = 0.247) and waist circumference (WMD −0.58 cm, 95% CI −1.89 to 0.72, p = 0.379) were all null. The only significant finding was a reduction in waist-hip ratio (WMD −0.03, 95% CI −0.04 to −0.01). The authors concluded that further high-quality clinical trials are needed before these supplements can be recommended.
0.65
Systematic review and meta-analysis of 12 RCTs (849 participants) of berberine and barberry (Berberis vulgaris) supplementation on anthropometric outcomes. None of the three headline measures moved: body weight (WMD −0.11 kg, 95% CI −0.13 to 0.91, p = 0.83), BMI (WMD −0.16 kg/m², 95% CI −0.43 to 0.11, p = 0.247) and waist circumference (WMD −0.58 cm, 95% CI −1.89 to 0.72, p = 0.379) were all null. The only significant finding was a reduction in waist-hip ratio (WMD −0.03, 95% CI −0.04 to −0.01). The authors concluded that further high-quality clinical trials are needed before these supplements can be recommended.
Design Meta-Analysis (1.0) × quality 0.65 = impact 0.65
View sourceMeta-Analysis
Pro
Asbaghi O et al. · 2020Clinical Nutrition ESPENSystematic review and meta-analysis of 12 RCTs of oral berberine supplementation (search through July 2019), pooling anthropometric, inflammatory and liver-enzyme outcomes. It is the most favourable of the berberine anthropometry meta-analyses: body weight fell by 2.07 kg (95% CI −3.09 to −1.05, p < 0.001), BMI by 0.47 kg/m² (95% CI −0.70 to −0.23), waist circumference by 1.08 cm (95% CI −1.97 to −0.19, p = 0.018) and CRP by 0.42 mg/L. Liver enzymes did not change significantly (ALT −1.66 IU, 95% CI −3.98 to 0.65; AST −0.87 IU, 95% CI −2.56 to 0.82). Participants were overwhelmingly people with metabolic disease (type 2 diabetes, metabolic syndrome, NAFLD, PCOS) rather than healthy adults seeking weight loss, and the authors called for higher-quality trials.
0.55
Systematic review and meta-analysis of 12 RCTs of oral berberine supplementation (search through July 2019), pooling anthropometric, inflammatory and liver-enzyme outcomes. It is the most favourable of the berberine anthropometry meta-analyses: body weight fell by 2.07 kg (95% CI −3.09 to −1.05, p < 0.001), BMI by 0.47 kg/m² (95% CI −0.70 to −0.23), waist circumference by 1.08 cm (95% CI −1.97 to −0.19, p = 0.018) and CRP by 0.42 mg/L. Liver enzymes did not change significantly (ALT −1.66 IU, 95% CI −3.98 to 0.65; AST −0.87 IU, 95% CI −2.56 to 0.82). Participants were overwhelmingly people with metabolic disease (type 2 diabetes, metabolic syndrome, NAFLD, PCOS) rather than healthy adults seeking weight loss, and the authors called for higher-quality trials.
Design Meta-Analysis (1.0) × quality 0.55 = impact 0.55
View sourceMeta-Analysis
Neutral
Xie W et al. · 2022Frontiers in PharmacologySystematic review and GRADE-rated meta-analysis of 37 RCTs (3,048 patients with type 2 diabetes) on berberine and glycaemic control. Berberine lowered fasting plasma glucose by 0.82 mmol/L (95% CI −0.95 to −0.70) and HbA1c by 0.63 percentage points (95% CI −0.72 to −0.53), both p < 0.001 — a real but sub-metformin effect. Two structural weaknesses dominate: **34 of the 37 trials were conducted in China** (one in Iran), and the GRADE certainty was only *moderate* for FPG and HbA1c (downgraded for risk of bias), *low* for 2-hour postprandial glucose (risk of bias plus publication bias) and *very low* for hypoglycaemia. Body weight was not the outcome of interest here.
0.60
Systematic review and GRADE-rated meta-analysis of 37 RCTs (3,048 patients with type 2 diabetes) on berberine and glycaemic control. Berberine lowered fasting plasma glucose by 0.82 mmol/L (95% CI −0.95 to −0.70) and HbA1c by 0.63 percentage points (95% CI −0.72 to −0.53), both p < 0.001 — a real but sub-metformin effect. Two structural weaknesses dominate: **34 of the 37 trials were conducted in China** (one in Iran), and the GRADE certainty was only *moderate* for FPG and HbA1c (downgraded for risk of bias), *low* for 2-hour postprandial glucose (risk of bias plus publication bias) and *very low* for hypoglycaemia. Body weight was not the outcome of interest here.
Design Meta-Analysis (1.0) × quality 0.60 = impact 0.60
View sourceDouble-Blind RCT
Con
Wilding JPH et al. · 2021New England Journal of MedicineSTEP 1: 68-week double-blind RCT randomizing 1,961 adults with obesity (BMI ≥30, or ≥27 with a weight-related condition) and without diabetes to once-weekly subcutaneous semaglutide 2.4 mg or placebo, both with lifestyle intervention. Mean body-weight change was **−14.9% with semaglutide vs −2.4% with placebo** (difference −12.4 percentage points, 95% CI −13.4 to −11.5; p < 0.001), i.e. −15.3 kg vs −2.6 kg (difference −12.7 kg, 95% CI −13.7 to −11.7). 86.4% of the semaglutide group lost ≥5% of body weight, 69.1% lost ≥10% and 50.5% lost ≥15%, versus 31.5%/12.0%/4.9% on placebo. Nausea and diarrhoea were the commonest adverse events, usually transient and mild-to-moderate; 4.5% discontinued for GI events vs 0.8% on placebo.
0.72
STEP 1: 68-week double-blind RCT randomizing 1,961 adults with obesity (BMI ≥30, or ≥27 with a weight-related condition) and without diabetes to once-weekly subcutaneous semaglutide 2.4 mg or placebo, both with lifestyle intervention. Mean body-weight change was **−14.9% with semaglutide vs −2.4% with placebo** (difference −12.4 percentage points, 95% CI −13.4 to −11.5; p < 0.001), i.e. −15.3 kg vs −2.6 kg (difference −12.7 kg, 95% CI −13.7 to −11.7). 86.4% of the semaglutide group lost ≥5% of body weight, 69.1% lost ≥10% and 50.5% lost ≥15%, versus 31.5%/12.0%/4.9% on placebo. Nausea and diarrhoea were the commonest adverse events, usually transient and mild-to-moderate; 4.5% discontinued for GI events vs 0.8% on placebo.
Design Double-Blind RCT (0.85) × quality 0.85 = impact 0.72
View sourceMeta-Analysis
Con
Amini MR et al. · 2020Complementary Therapies in MedicineSystematic review and meta-analysis of 12 RCTs (849 participants) of berberine and barberry (Berberis vulgaris) supplementation on anthropometric outcomes. None of the three headline measures moved: body weight (WMD −0.11 kg, 95% CI −0.13 to 0.91, p = 0.83), BMI (WMD −0.16 kg/m², 95% CI −0.43 to 0.11, p = 0.247) and waist circumference (WMD −0.58 cm, 95% CI −1.89 to 0.72, p = 0.379) were all null. The only significant finding was a reduction in waist-hip ratio (WMD −0.03, 95% CI −0.04 to −0.01). The authors concluded that further high-quality clinical trials are needed before these supplements can be recommended.
0.65
Systematic review and meta-analysis of 12 RCTs (849 participants) of berberine and barberry (Berberis vulgaris) supplementation on anthropometric outcomes. None of the three headline measures moved: body weight (WMD −0.11 kg, 95% CI −0.13 to 0.91, p = 0.83), BMI (WMD −0.16 kg/m², 95% CI −0.43 to 0.11, p = 0.247) and waist circumference (WMD −0.58 cm, 95% CI −1.89 to 0.72, p = 0.379) were all null. The only significant finding was a reduction in waist-hip ratio (WMD −0.03, 95% CI −0.04 to −0.01). The authors concluded that further high-quality clinical trials are needed before these supplements can be recommended.
Design Meta-Analysis (1.0) × quality 0.65 = impact 0.65
View sourceShowing the 4 strongest of 8 studies. Tap any node to expand its detail.
Evidence
PRO (1)
PRO Meta-Analysis0.55 Asbaghi O, Ghanbari N et al. (2020)
Body weight -2.07 kg (95% CI -3.09 to -1.05, p < 0.001), BMI -0.47 kg/m2 (95% CI -0.70 to -0.23), waist circumference -1.08 cm (95% CI -1.97 to -0.19, p = 0.018), CRP -0.42 mg/L; liver enzymes unchanged (ALT -1.66 IU, AST -0.87 IU, both non-significant)
Systematic review and meta-analysis of 12 RCTs of oral berberine supplementation (search through July 2019), pooling anthropometric, inflammatory and liver-enzyme outcomes. It is the most favourable of the berberine anthropometry meta-analyses: body weight fell by 2.07 kg (95% CI −3.09 to −1.05, p < 0.001), BMI by 0.47 kg/m² (95% CI −0.70 to −0.23), waist circumference by 1.08 cm (95% CI −1.97 to −0.19, p = 0.018) and CRP by 0.42 mg/L. Liver enzymes did not change significantly (ALT −1.66 IU, 95% CI −3.98 to 0.65; AST −0.87 IU, 95% CI −2.56 to 0.82). Participants were overwhelmingly people with metabolic disease (type 2 diabetes, metabolic syndrome, NAFLD, PCOS) rather than healthy adults seeking weight loss, and the authors called for higher-quality trials.
Weighted 0.55 — k=12 RCTs; the paper does not state a pooled participant total, so sample_size is omitted rather than guessed. This is the most favourable of the berberine anthropometry meta-analyses and it disagrees with the others, which is itself a warning sign. Participants were overwhelmingly people with type 2 diabetes, metabolic syndrome, NAFLD or PCOS rather than healthy adults seeking weight loss (serious indirectness), the included trials are small, and the authors call for higher-quality trials. Funding could not be verified, so the field is omitted.
Clinical Nutrition ESPEN
AGAINST (3)
AGAINST Double-Blind RCTn=1961CONFLICTED0.85 Wilding JPH, Batterham RL et al. (2021)
-14.9% body weight vs -2.4% on placebo at 68 weeks (difference -12.4 percentage points, 95% CI -13.4 to -11.5); 50.5% lost >=15% of body weight vs 4.9% on placebo
STEP 1: 68-week double-blind RCT randomizing 1,961 adults with obesity (BMI ≥30, or ≥27 with a weight-related condition) and without diabetes to once-weekly subcutaneous semaglutide 2.4 mg or placebo, both with lifestyle intervention. Mean body-weight change was −14.9% with semaglutide vs −2.4% with placebo (difference −12.4 percentage points, 95% CI −13.4 to −11.5; p < 0.001), i.e. −15.3 kg vs −2.6 kg (difference −12.7 kg, 95% CI −13.7 to −11.7). 86.4% of the semaglutide group lost ≥5% of body weight, 69.1% lost ≥10% and 50.5% lost ≥15%, versus 31.5%/12.0%/4.9% on placebo. Nausea and diarrhoea were the commonest adverse events, usually transient and mild-to-moderate; 4.5% discontinued for GI events vs 0.8% on placebo.
Weighted 0.85 — 1,961 randomized, 94.3% completing, 68 weeks, double-blind and placebo-controlled, with a very large and precisely estimated effect — as strong as obesity pharmacotherapy trials get. Held below the top band for independence rather than rigour: the manufacturer designed the trial, oversaw its conduct and performed data collation and analysis, making this an industry-funded positive on the sponsor's own drug, the direction in which sponsor bias is most dangerous.
Funding: Novo Nordisk — the funder profits from this result.
New England Journal of Medicine
AGAINST Meta-Analysisn=8490.65 Amini MR, Sheikhhossein F et al. (2020)
Null across the headline anthropometrics: body weight WMD -0.11 kg (p = 0.83), BMI WMD -0.16 kg/m2 (p = 0.247), waist circumference WMD -0.58 cm (p = 0.379); only waist-hip ratio moved (WMD -0.03, 95% CI -0.04 to -0.01)
Systematic review and meta-analysis of 12 RCTs (849 participants) of berberine and barberry (Berberis vulgaris) supplementation on anthropometric outcomes. None of the three headline measures moved: body weight (WMD −0.11 kg, 95% CI −0.13 to 0.91, p = 0.83), BMI (WMD −0.16 kg/m², 95% CI −0.43 to 0.11, p = 0.247) and waist circumference (WMD −0.58 cm, 95% CI −1.89 to 0.72, p = 0.379) were all null. The only significant finding was a reduction in waist-hip ratio (WMD −0.03, 95% CI −0.04 to −0.01). The authors concluded that further high-quality clinical trials are needed before these supplements can be recommended.
Weighted 0.65 — 12 RCTs, 849 pooled participants. A real meta-analysis with a clean null on the outcomes that matter, but the included trials are individually small, mostly in metabolic-disease populations rather than healthy adults, and the authors themselves say higher-quality trials are needed. No funding statement could be verified, so the field is omitted.
Complementary Therapies in Medicine
AGAINST Meta-Analysis0.45 Xiong P, Niu L et al. (2020)
BMI -0.29 kg/m2 (95% CI -0.51 to -0.08) and waist circumference -2.75 cm (-4.88 to -0.62), but body weight itself unchanged (-0.11 kg, -0.99 to 0.76, p = 0.79)
Dose-response meta-analysis and systematic review of 10 RCTs of berberine supplementation on obesity indices. Berberine produced a statistically significant but very small reduction in BMI (WMD −0.29 kg/m², 95% CI −0.51 to −0.08, p = 0.006) and in waist circumference (WMD −2.75 cm, 95% CI −4.88 to −0.62, p = 0.01), but no significant change in body weight itself (WMD −0.11 kg, 95% CI −0.99 to 0.76, p = 0.79). The dose-response analysis found the BMI and waist-circumference effects tracked treatment duration rather than dose.
Weighted 0.45 — No single pooled total is reported, so sample size is omitted: 10 studies from 9 articles, with 983 individuals in the BMI analysis, 841 for waist circumference and 378 for body weight, some arms as small as 12 per group. The result is internally incoherent — BMI and waist fell while body weight did not, which the authors concede is paradoxical — and the published abstract's waist-circumference estimate (-2.75 cm) does not match the accepted manuscript's (-1.78 cm). Small, inconsistent and inconsistently reported.
Complementary Therapies in Clinical Practice
NEUTRAL (4)
NEUTRAL RCTn=1040.65 Wu X, Li Q et al. (2005)
Berberine 0.2 g three times daily raised cyclosporin A exposure. In a separate 6-patient, 12-day pharmacokinetic study: AUC +34.5%, steady-state concentration (Css) +34.5%, trough concentration (Cmin) +88.3%, time to peak (tmax) +1.7 h, half-life +2.7 h, apparent oral clearance (CL/F) -40.4% (all P < 0.05). The paper reports no percentage change in Cmax
Randomized controlled clinical and pharmacokinetic study in renal-transplant recipients on cyclosporin A: "52 renal-transplant recipients were treated with CsA and 0.2 g BBR three times daily for 3 months, while another 52 subjects received CsA without BBR co-administration."
The pharmacokinetic numbers come from a separate, smaller study, not from a subset of those 104: "six renal-transplant recipients were included with a 3-mg/kg dosage of CsA twice daily before and after oral co-administration of 0.2 g BBR three times daily for 12 days." In those six patients berberine markedly raised cyclosporin exposure — AUC +34.5%, steady-state concentration (Css) +34.5%, trough concentration (Cmin) +88.3%, time to peak (tmax) +1.7 h, half-life +2.7 h, and apparent oral clearance (CL/F) −40.4% (all P < 0.05). The paper reports no percentage change for Cmax; an earlier version of this record misattributed the +88.3% trough figure to Cmax.
The authors attribute the interaction to inhibition of CYP3A4 "in the liver and/or small intestine." P-glycoprotein is not mentioned anywhere in this paper — a previous version of this record credited a P-gp contribution that the source does not support. This remains a direct human demonstration that berberine is not pharmacologically inert: it can substantially alter blood levels of co-administered drugs metabolised by CYP3A4.
Weighted 0.65 — 104 renal-transplant recipients over 3 months with a pharmacokinetic subset — a reasonable n and a large, unambiguous drug-interaction signal. Downgraded because no blinding or placebo is described, it is single-centre, and both arms rose from baseline, so the between-group difference is materially smaller than the within-group headline numbers suggest. No funding statement is retrievable.
European Journal of Clinical Pharmacology
NEUTRAL Meta-Analysisn=30480.60 Xie W, Su F et al. (2022)
Berberine lowered fasting plasma glucose by 0.82 mmol/L (95% CI -0.95 to -0.70) and HbA1c by 0.63 percentage points (-0.72 to -0.53), both p < 0.001 — real but sub-metformin
Systematic review and GRADE-rated meta-analysis of 37 RCTs (3,048 patients with type 2 diabetes) on berberine and glycaemic control. Berberine lowered fasting plasma glucose by 0.82 mmol/L (95% CI −0.95 to −0.70) and HbA1c by 0.63 percentage points (95% CI −0.72 to −0.53), both p < 0.001 — a real but sub-metformin effect. Two structural weaknesses dominate: 34 of the 37 trials were conducted in China (one in Iran), and the GRADE certainty was only moderate for FPG and HbA1c (downgraded for risk of bias), low for 2-hour postprandial glucose (risk of bias plus publication bias) and very low for hypoglycaemia. Body weight was not the outcome of interest here.
Weighted 0.60 — 37 RCTs, 3,048 patients with type 2 diabetes, GRADE-rated by the authors — and their own ratings are the reason for the downgrade: only moderate certainty for fasting glucose and HbA1c (risk of bias), low for 2-hour postprandial glucose (risk of bias plus publication bias), very low for hypoglycaemia. 34 of the 37 trials come from a single country, a known publication-bias hotspot for this compound. No commercial relationships declared.
Frontiers in Pharmacology
NEUTRAL Meta-Analysisn=8890.50 Liu D, Zhao H et al. (2025)
triglycerides -0.367 mmol/L; fasting glucose -0.515 mmol/L; waist circumference -3.270 cm; BMI -0.435 kg/m2; no significant effect on blood pressure or HDL
Systematic review and meta-analysis of 12 randomized placebo-controlled trials (889 participants, durations 84–140 days) of berberine on the components of metabolic syndrome, with formal Cochrane RoB 2 assessment. Berberine reduced triglycerides (WMD −0.367 mmol/L, 95% CI −0.560 to −0.175), fasting plasma glucose (WMD −0.515 mmol/L, 95% CI −0.847 to −0.183), waist circumference (WMD −3.270 cm, 95% CI −4.818 to −1.722) and BMI (WMD −0.435 kg/m², 95% CI −0.856 to −0.013, p = 0.043), with no significant effect on blood pressure or HDL. The risk-of-bias picture is the key finding: of 12 trials only 2 were low risk of bias in all domains, 2 were high risk and the remaining 8 raised "some concerns"; the trials came overwhelmingly from China (plus Mexico and India). Adverse events did not differ from placebo and were mostly mild GI symptoms (nausea, constipation, abdominal discomfort), though two trials reported no safety data at all.
Weighted 0.50 — k=12 placebo-controlled RCTs, 889 participants. Only 2 of the 12 trials were low risk of bias in all domains — 2 were high risk and 8 raised "some concerns" — heterogeneity was high (I2 81-90% for the lipid and glucose outcomes), and publication bias was detected for triglycerides (Egger p=0.041). The trials came overwhelmingly from China, and the funders are Chinese traditional-medicine institutes reporting a favourable result for a traditional-medicine compound.
Funding: State Key Laboratory of Dampness Syndrome of Chinese Medicine; Chinese Academy of Traditional Chinese Medicine
Frontiers in Pharmacology
NEUTRAL Narrative Review0.30 Khoshandam A, Imenshahidi M et al. (2022)
Oral berberine bioavailability is under 1% owing to intestinal first-pass metabolism, P-glycoprotein efflux and self-aggregation; 11-23% is excreted in faeces, and it interacts clinically with metformin, cyclosporin A and digoxin
Comprehensive pharmacokinetic review of berberine, the principal alkaloid of Berberis vulgaris. It documents that oral berberine has very poor solubility and bioavailability — under 1% — because of extensive intestinal first-pass metabolism, P-glycoprotein-mediated efflux and self-aggregation, so very little of an oral dose reaches systemic circulation intact. Metabolism proceeds via demethylenation, reduction and dioxymethylene cleavage (phase I) then glucuronidation, sulfation and methylation (phase II); the liver is the main distribution site and 11–23% is excreted in faeces. The review also flags clinically relevant interactions with metformin, cyclosporin A and digoxin, and reviews formulation strategies developed specifically to work around the absorption problem.
Weighted 0.30 — Narrative pharmacokinetic review with no systematic search strategy, no inclusion criteria and no quantitative synthesis, so what gets cited rests on the authors' discretion. The funding and conflict-of-interest statements sit behind a paywall and could not be verified from any free source, so funding is omitted rather than guessed. Carried mainly for the sub-1% bioavailability figure, which is well corroborated elsewhere.
Phytotherapy Research