Drug interactions

How does a supplement–drug interaction arise?

Four mechanisms stand out in the sources:

• Absorption: Minerals such as calcium, magnesium, zinc and iron can form insoluble compounds with some medicines; this is described by NIH for tetracycline and quinolone antibiotics. ^[9]
• Metabolism and transporters: For berberine, a review lists CYP3A4 inhibition and induction and effects on transporters such as P-glycoprotein; it notes that few pairs are documented in humans and that most are mechanism-based. ^[16] According to NCCIH, St. John’s wort can weaken the effect of many medicines. ^[13]
• Pharmacodynamic summation: Two substances acting in the same direction reinforce each other: bleeding (an anticoagulant with some supplements), serotonin (an antidepressant with St. John’s wort), sedation (a benzodiazepine with melatonin), lowered blood glucose, raised potassium (hyperkalemia).
• Reverse direction: The medicine changes the need for the supplement; for example, long-term PPI and metformin use can affect B12 status, and PPI use can affect magnesium status.

Warfarin: vitamin K, omega-3, coenzyme Q10, ginkgo and curcumin

NIH states that warfarin and some anticoagulants used in Europe block the action of vitamin K and that people using these medicines need to keep vitamin K intake consistent. ^[1] For omega-3, NIH reports that high doses of fish oil may have an antiplatelet effect and may raise the INR with warfarin, but that in most studies 3–6 g/day of fish oil did not significantly change anticoagulant status in warfarin users; the labels of FDA-approved omega-3 drugs say INR should be monitored at regular intervals when used with anticoagulants. ^[2]

For coenzyme Q10 and ginkgo the picture is more mixed. NCCIH writes that ginkgo may increase bleeding risk in people using anticoagulants. ^[4] In the randomized, double-blind crossover trial of Engelsen and colleagues, 24 outpatients on long-term stable warfarin received 100 mg/day of coenzyme Q10, 100 mg/day of ginkgo and placebo in 4-week periods; the INR remained stable and the weekly warfarin dose did not change (ginkgo 36.7 mg, coenzyme Q10 36.5 mg, placebo 36.0 mg). ^[3] Because the trial was small and short, it does not definitively exclude an interaction; the same abstract also reports the existence of case reports of reduced response with coenzyme Q10 and increased response with ginkgo. For curcumin, the finding in the source is in animals: in Wistar rats, high-dose curcumin given for 7 days increased the area under the curve of warfarin and of the carboxylic acid metabolite of clopidogrel by about 1.6-fold but did not affect coagulation or platelet aggregation; this result cannot be transferred directly to humans. ^[5]

Levothyroxine: calcium, iron and the laboratory effect of biotin

MedlinePlus writes that calcium carbonate or ferrous sulfate should be taken at least 4 hours before or 4 hours after levothyroxine; it notes that proton pump inhibitors and antacids are also among the products that can interact with levothyroxine. ^[6] The calcium and iron guides give the same 4-hour interval through the levothyroxine label cited by NIH (“not within 4 hours”); the two sources show the same duration.

Biotin, on the other hand, is a problem not for the medicine but for the tests used in medicine monitoring. In the nonrandomized crossover study of Li and colleagues with 6 healthy adults, taking 10 mg/day of biotin for 7 days was associated with interference in 9 of 23 tests (39%) using biotin-streptavidin technology and in none of the 14 tests not using biotin; 5 of 8 competitive immunoassays gave falsely high results and 4 of 15 sandwich immunoassays gave falsely low results. The study measured 11 analytes, including thyroid-stimulating hormone and thyroxine. ^[7] Because these tests are used in dose monitoring in people on thyroid medicine, the interference is meaningful; this is a mechanism-based inference.

Quinolones, tetracyclines and bone medicines: mineral intervals

NIH states that tetracycline and quinolone antibiotics should be taken at least 2 hours before or 4–6 hours after supplements and medicines containing magnesium, and that at least 2 hours should be left between oral bisphosphonates (for example alendronate) and magnesium. ^[9] For ciprofloxacin, the medicine’s own information at MedlinePlus is as follows: antacids and multivitamins containing calcium, magnesium hydroxide, aluminum hydroxide, zinc or iron are taken at least 6 hours before or 2 hours after ciprofloxacin. ^[8] The wording of the two sources differs in form (depending on which medicine is referenced); for other antibiotics in the quinolone class, the interval may vary according to the medicine’s leaflet. For the timing of iron, zinc and magnesium, the page when to take supplements can be consulted.

PPIs, metformin and ACE inhibitors: when the medicine changes supplement status

In the case-control study of Lam and colleagues at Kaiser Permanente with 25,956 cases of B12 deficiency and 184,199 controls, PPI use for at least 2 years (OR 1.65; 95% CI 1.58–1.73) and H2 receptor blocker use for at least 2 years (OR 1.25; 95% CI 1.17–1.34) were associated with a diagnosis of B12 deficiency; this is an association, not evidence of cause and effect. ^[10] For metformin, in the Diabetes Prevention Program Outcomes Study, low B12 (≤203 pg/mL) was found at year 5 in 4.3% of the metformin group and 2.3% of the placebo group; at year 13 this difference was not significant (7.4% versus 5.4%), but when low and borderline-low B12 (≤298 pg/mL) were considered together, it was still more frequent in the metformin group at year 13 (20.3% versus 15.6%), and each year of metformin use was associated with deficiency risk (OR 1.13; 95% CI 1.06–1.20). ^[11] NIH also reports that long-term (generally more than one year) PPI use can lead to hypomagnesemia. ^[9]

For potassium there is a reverse relationship: according to NIH, ACE inhibitors and angiotensin receptor blockers reduce urinary potassium excretion; potassium-sparing diuretics such as amiloride and spironolactone can also lead to hyperkalemia, and monitoring of potassium status is recommended in people using these medicines, especially if kidney function is impaired. ^[12]

Antidepressants and sedatives: serotonin and sedation

NCCIH writes that St. John’s wort, when taken together with some antidepressants or other medicines that affect serotonin, can increase serotonin-related side effects, which can be serious. ^[13] 5-HTP is a serotonin precursor; a review addresses serotonin syndrome among its safety topics. ^[14] Evaluating its use together with serotonergic medicines is therefore a mechanism-based matter.

On sedation, the EMA product information states that melatonin can increase the sedative effect when used together with benzodiazepines and with zolpidem, zaleplon and zopiclone; the same document states that fluvoxamine increases melatonin exposure (area under the curve) about 17-fold and that this combination should be avoided. ^[19] Melatonin is a licensed medicine in Türkiye; the interaction warnings in the Turkish SmPC (KÜB) are reported in the melatonin guide. For ashwagandha, a regulatory review reports that the French ANSES (April 2024) recommended that people using sedatives avoid these supplements; this is not a measured drug-interaction study but a risk-group assessment. Among the authors of the review are people from an industry association, a food regulatory consultancy firm and a law firm. ^[18]

Immunosuppressants: cyclosporine, probiotics, ashwagandha, berberine

NCCIH writes that St. John’s wort can weaken the effect of cyclosporine (a medicine that prevents transplant rejection). ^[13] With berberine the direction is reversed: in two studies by Xin and colleagues with 6 healthy men each, a single 0.3 g dose of berberine with 3 mg/kg cyclosporine increased the area under the curve by 19.2%; with 6 mg/kg cyclosporine, 0.3 g of berberine twice a day for 10 days produced no significant change. ^[15] In a randomized controlled trial in kidney transplant recipients (52 + 52 patients, 3 months, 0.2 g of berberine three times a day), the cyclosporine trough level was 29.3% higher than in the control group; in the 6-patient pharmacokinetic part, the area under the curve rose by 34.5%. ^[25] The berberine review cites this trial as the clinical basis of the interaction. ^[16]

For probiotics, Boyle and colleagues state that the overall safety record is very good but that they should be used with caution in groups such as premature infants and people with immune deficiency. ^[17] For ashwagandha, among the concerns of Denmark’s DTU is a possible effect on the immune system; this is a possible effect, not a measured interaction result. ^[18]

Blood glucose, cholesterol and heart medicines: berberine, coenzyme Q10, green tea, red yeast rice

The berberine review lists the risks of hypoglycemia, hypotension and QT prolongation along the axis of pharmacodynamic summation; this is a mechanism-based warning and the pairs documented in humans are limited. ^[16] For people using blood-glucose-lowering medicines, adding berberine therefore calls for evaluation with a physician or pharmacist.

With statins, for coenzyme Q10 the question is need rather than an interaction, and meta-analyses conflict: in the meta-analysis of Banach and colleagues of 6 RCTs (302 patients), coenzyme Q10 showed no significant effect on muscle pain or creatine kinase ^[20]; the meta-analysis of Qu and colleagues of 12 RCTs (575 patients), by contrast, reported reductions in muscle pain, weakness, cramps and fatigue and no change in creatine kinase. ^[26] Details are in the coenzyme Q10 guide. For green tea, NCCIH reports that large amounts of green tea reduced the blood level and effect of the beta blocker nadolol, that green tea extract can reduce atorvastatin levels, and that an interaction with raloxifene has been shown; details are in the green tea guide. ^[27] Red yeast rice is a separate problem: NCCIH states that monacolin K is structurally identical to lovastatin, that the products can cause the same kinds of drug interactions as statins and that the amount of monacolin K in products cannot be known. ^[21] In Türkiye, the use of monacolin K and red yeast rice as active substances in food supplements was not found suitable in the Ministry of Agriculture and Forestry list (Restricted Substances List; list entry 15/06/2022). ^[22]

Chemotherapy: antioxidant supplements

In SWOG S0221 (1,134 breast cancer patients), use of any antioxidant supplement (vitamins A, C, E, carotenoids, coenzyme Q10) before and during treatment was associated with a signal toward an increased recurrence rate (adjusted HR 1.41; 95% CI 0.98–2.04; P = 0.06); use of B12 and iron was associated with significantly worse outcomes, while multivitamin use was not associated with survival outcomes. The authors state that, for supplements other than multivitamins, the results are consistent with cautionary recommendations. ^[23] The NCI PDQ summary writes that research on interactions between supplements and cancer medicines is limited, that for many antioxidants there is no information to determine safety and efficacy, and that it is not an official guideline. ^[24] In people receiving chemotherapy, every supplement is evaluated together with the oncology team.

Known interactions: table

How is interaction information read?

The table places statements from the relevant sources side by side; a source not mentioning a pair does not show that no interaction exists. Product information (leaflet, SmPC (KÜB)) and the instructions of the prescribing physician or pharmacist are not replaced by this summary; medicine times or doses are not changed personally. The relation of a supplement to time and meals is covered in the guide when to take supplements, label information in how to read a label, general safety in are supplements safe, and definition and scope in the pages what are supplements and do you need supplements.

Related supplement guides: melatonin, probiotics, berberine, ashwagandha, curcumin, green tea, coenzyme Q10, magnesium, potassium, vitamin B12, biotin, vitamin E.

This guide does not recommend a personal dose. In people who use regular medicines, are pregnant, have kidney disease or are receiving cancer treatment, supplements are evaluated together with the medicine list by a physician or pharmacist.