Drug and Herb Interactions — Current Study Findings

Paul Reller, L.Ac.

An enormous amount of effort has gone into the publication and distribution of information concerning negative herb and drug interactions and the dangers presented, yet actual cases of harm have been nearly nonexistent. This cry wolf strategy has led to many physicians wondering what is actually going on here, as clinical problems have not emerged. The widespread publication and distribution of alarming in vitro study evidence has not been supported by in vivo studies or translation into clinical cases of harm. Faced with these facts, the organizations sponsoring these in vivo studies suggest that poor quality of herbal products and poor absorption and bioavailability must be the reason why there are no actual clinical cases of harm from herb-drug interactions. Considering the significant proof that is accumulating of not only clinical anecdotal evidence of herbal success in treatment, but also the evidence in large randomized controlled stage three trials of the effectiveness of herbal therapies, these are amazingly vacant explanations. Here is a small amount of scientific data to clarify this situation and help both the M.D. and the herbalist interact with more confidence in Complementary Medicine.

A 2008 joint study of the six most widely reported herbal culprits in potential drug-herb interactions, St. John's Wort, Kava-kava, milk thistle, black cohosh, echinacea and goldenseal, by the Colleges of Medicine and Pharmacology at the Universities of Arkansas and Mississippi, found that evidence of inhibition of the P450 pathway of drug catabolism was found only in goldenseal. This study dealt with the CYP2D6 genetic isoform that expresses the enzymes involved in about 30% of all medication metabolism. The study involved three separate studies on humans, each with 18 participants that were randomized to receive standardized botanical extracts for 14 days on two separate occasions, with a 30-day period between the courses of herbal extract. The CYP expression was evaluated using administeration of a CYP2D6 substrate, debrisoquine 5mg, and measuring the catabolic rate with urinary analysis. The fact that the other five herbs, which have been widely reported over the years, unfairly, and without substantiation, to alter the circulating levels of key drugs prescribed, have been finally cleared of risk, is significant. Click here to see this study: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2562884

The CYP2D6 pathway is both representative of P450 pathways of catabolism in general, and also the path of catabolism of many antidepressants, antipsychotics, beta-receptor antagonists, analgesics, and antiarrhythmia agents, making it important in the subject of clinical problems with negative interactions. Most studies have dealt with the CYP3A4 substrates, which is a somewhat more widely used path of pharmaceutical biotransformation, but which is more easily inducible and modulated, and concerns pharmaceuticals less widely prescribed and with less danger of clinical problems with temporary changes in circulating levels, except for warfarin (coumadin), which utilizes the biotransformation pathway of CYP3A4 and CYP1A2 primarily, and which we be dealt with in more detail below. The CYP3A4 pathway, while associated with a greater number of pharmaceutical catabolisms, is also generally a shared enzymatic pathway with other P450 enzyme expressions, more so than the CYP2D6. One interesting aspect of the plethora of studies concerning drug-herb interaction is the amount of information uncovered that provides the physician with a more complete perspective on pharmaceutical biotransformation and the problems with variance of effective circulating levels and drug-drug interactions. Once again, research concerning Complementary Medicine that is directed at discouraging its use is ultimately supporting the practice.

The CYP2D6 pathway is not easily inducible, and so herbal intake would probably either inhibit the enzyme or have no effect at all. This study confirms the potential of a standardized extract of goldenseal to affect the circulating level of debrisoquine by decreasing the rate of catabolism and temporarily increasing circulating levels. Debrisoquine, a derivative of guanidine, is not a pharmaceutical drug, but has been used extensively in studies to investigate variance in biotransformative processes, either due to genetic polymorphisms in the general population, or due to such factors as blood levels of hematocrit, MCHC, plasma proteins, and ponderal index, or body mass. Two key points of interest are seen in this study cited above, the fact that 5 of the 18 study subjects showed clinically insignificant drops in the rates of urinary excretion of the drug, and two, that a standardized extract of the herb was used, rather than the whole herb extract, which has been shown to be less problematic over the years in scientific studies due to the array of modulatory chemicals in the plant. Research in Europe around 2000 discovered that standardized extracts of various herbs were presenting mild to moderate side effects, while the whole, or natural extract index were not presenting these side effects. The theory was presented that the plant had evolved an array of modulating chemicles to protect itself from unwanted metabolic processes. Another parameter that might be considered in the final analysis, is that the 2D6 genetic expression is been shown to be polymorphic in a significant portion of the population, which results in deficiency of the expression of this enzyme. In actual clinical practice, we might ask what would happen if the whole herb extract was used, at the usual low dosage for a short period of time, and why did the standardized extract of considerable dosage not significantly affect the catabolic pathway in all of the subjects.

This brings us to the subject of biotransformative variance in the general population. As stated above, common differences between individuals may have a significant effect on the rate of metabolism of drugs. A 1991 study at the Laboratoire du Centre de Medecine Preventive et Centre du Medicament, in Vandoeuvre-les-Nancy, France, and published in the journal Clinical Chemistry, found that metabolic rates in the random study population varied considerably. Extensive metabolizers excreted an average of 17.4% of debrisoquine in the urine, while poor metabolizers secreted only 0.5%. To review this study, click here: http://www.clinchem.org/cgi/content/abstract/37/3/327. This information on drastic biotransformation variance with debrisoquine was available in 1991, yet was not mentioned in the study on herb-drug interactions cited above. What does this information actually tell us? It tells us that the general health of the individual patient has a lot to do with the effectiveness of the pharmaceutical drugs administered, because a holistic array of health factors can vary the effective circulating dosage considerably. Prescribing doctors should be asking themselves if their patients could benefit from Complementary Medicine in improvement in these general health factors, such as blood quality and quantity, immune health, health of the digestive system, hepatoprotection and decrase in genetic polymorphic expression, etc. Optimal health of the patient apparently has a lot to do with the metabolic variance and effectiveness of the drugs prescribed.

Current studies of these herbs mentioned above on the CYP3A pathway have also shown that these herbs do not significantly affect the P450 pathway. One 2007 study at the same medical schools cited above shows that clarithromycin and rifampin, frequently prescribed for acute infections, including pneumonia, meningitis, and for the now prevalent TB, significantly modulate the CYP3A catabolism, yet milk thistle and black cohosh have no effect. Click here to see this study: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1865122. This study finds that a search of the medical databases still finds only a handful of herbs with any potential to affect the CYP3A catabolism, and the St. John's Wort (hypericum perforatum) is the most noteworthy, yet chronic ingestion of the active ingredient hyperforin would be necessary to cause up regulation of intestinal CYP3A4, producing reduction in oral bioavailability of medications using this intestinal catabolism. St. John's Wort is not usually prescribed on a chronic basis by a Licensed Acupuncturist and herbalist, but rather to effect changes with short term use that leads to a resolution of health problems. It is also prescribed in small dosage. Here too, is a mistake that many M.D.'s and researchers make, applying paradigms of chronic long term use of herbs in the way pharmaceuticals are generally prescribed.

In the case of reviews of drug-herb interactions, we note that there is seldom mention of very many herbs, and the same herbs continue to be mentioned in every review of the subject, usually St. John's Wort is one of two herbs mentioned. Click here to see a typical citing of a medical text on this subject cited in Current Phamacogenomics in 2004: http://www.ingentaconnect.com/content/ben/cpg/2004/00000002/00000002/art00009. The second herb mentioned, evodia fruit (wu zhu yu), is seldom used in standard herbal TCM prescription, and only for short courses if at all, because it is a very spicy, bitter and hot herb with slight toxicity, and is useful only for treatment of hiatal hernia, liver or stomach organ pain, and certain pathologies of vomiting or diarrhea associated with liver and stomach organ pathology. The professional herbalist is aware of the problems of long term use and interactions of this herb from study in medical school and from current herbal clinical texts, and prescibes appropriately. TCM also utilizes knowledge of other herbs which moderate its mild potential adverse effects and utilizes these in formulas with evodia. This is the standard of practice in Traditional Chinese Medicine, and the Licensed Acupuncturist. The potential for drug-herb interactions is handled with overcaution in TCM medical schools for the most part, and few herbs come up with potential for such interactions, the same as in these current medical reviews.

Besides the lack of reported clinical problems with herb-drug interactions, and the paucity of herbs found with potential for negative interaction despite the large number of herbs used in standard practice, there are considerations of both the modulatory array of chemicals in a plant that reduce negative side effects, and the subject of reactive adaption of our own body's genetic expression and rate of response to increased enzymatic need, that need to be considered when looking beyond research theory into actualities of negative herb-drug interactions. Current research indicates that there is considerable individual variance in P450 expression and enzyme activity level, and that a post-transciptional mechanism is responsible for coregulation of P450 enzyme activity. To confirm these findings, click here to access a 2007 study by the University of Kansas Medical Center: http://dmd.aspetjournals.org/cgi/content/full/35/9/1700. While the material here is highly technical, the content suggests that P450 enzyme activity is affected by a number of variables, including genetic coregulation that may be stimulated by other chemicals in an herb, P450 oxidoreductase, constitutive androstane receptor, and hepatic nuclear factor 4alpha, which also could be affected by various chemicals within a single herb, or by hormonal regulation, and immune cytokines. Deficiency of the hormone leptin in study animals produces about a 20% decrease in P450 levels, and the study cited above shows that levels of hemoglobin and plasma transport proteins have a significant effect on clearance of drugs from circulation.

It seems that applying alarm in the subject of drug-herb interactions may have been premature or an expression of bias when so little evidence has been generated and so many variables not accounted for. Hepatic nuclear factor 4alpha (HNF4alpha) was found to be the most significant post-transcriptional regulator of P450 activity rates, and this genetic factor is known to be influenced both by immunoprecipitation affecting binding to CYP promoters, and by chemicals in common Chinese herbs. A study in 2008 revealed that berberine, from the Chinese herb scutellaria baicalensis (Huang qin) upregulates both mRNA and protein expression of HNFalpha. Click here to confirm this finding published in the World Jounal of Gastroenterology: http://www.wjgnet.com/1007-9327/14/6004.asp. This finding shows how a Chinese herbal formula could modify the P450 effects of specific chemicals in specific herbs in the formula. Chinese doctors that prescribe these formulas have been aware for some time of the chemical effects and have analyzed the modulatory effects of formula variations by both observing symptoms, signs and laboratory reports to reduce any harmful effects on liver metabolism and pharmaceutical clearance. Chinese research on the pharmacodynamics of herbal chemicals has been extensive, and prescription of herbs in China is handled professionally within the hospitals by standard medical doctors and pharmacists with daily clinical experience with concommittant prescription of pharmaceuticals and herbal extracts. Indeed, the majority of teachers in TCM medical schools in the United States are medical doctors and pharmacists with training and experience in China.

In addition to these potential regulators and modulators of the P450 expression rates mentioned above, it has been found that pharmaceuticals themselves sigfificantly affect the P450 expression over time and with dosage. In one study, Nelfinavir, an antiviral drug, increased hepatic CYP3A by 85%, and a regulatory intestinal protein Pgp by 83%. Study animals had highly variable expression of hepatic CYP3A levels with different dosage, and clearance changed significantly over time, implying a decrease in effectiveness of the drug in the weeks following administration. Increases in CYP3A expression rose as high as 175% due to administration of the drug. To see this 2001 study by Glaxo SmithKline, click here: http://dmd.aspetjournals.org/cgi/content/full/29/5/754. Acetominaphen is a drug that has presented severe problems in the clinical setting due to competition in the P450 catabolism. Concommitant use of acetaminophen and ibuprofen may cause severe hepatotoxicity, because the phase one P450 enzyme catabolism of acetaminophen produces toxic n-acetyl-p-benzoquinone-imine that must be conjugated with glutathione before it can be excreted, and ibruprofen interferes with the conjugation. Acetaminophen is not only available over the counter, but is also added to almost all narcotic pain medication, despite the stress it puts on the P450 biotransformative pathway. Surely, short courses of herbs present minor problems with variance in circulating levels of medications compared to these types of problems with drug-drug interactions.

The case of St. John's Wort, almost universally cited in the subject of drug-herb interactions

The study cited above confirms a number of scientific studies that showed that St. John's Wort is not a potent modulator of CYP2D6 catabolism in humans. Other studies have shown that St. John's Wort may modulate the CYP3A4 pathway, but a number of these studies used questionable drugs that produced study bias on the part of the designers. The chief drug used in these studies was dextromethorphan/dextrorphan, and the measurement was of dextromethorphan/dextrorphan urinary ratio, which can be altered by normal changes in urinary pH by twenty-fold. Why these previous researchers did not use a more reliable parameter, such as the studies on St. John's Wort cited above did, has not been explained. We see how study design can have a great affect on the outcome of the research, and how it is possible that research funding and design produced numerous studies that contained inherent bias and misinformation concerning the dangers of St. John's Wort.

Researchers into St. John's Wort effect on liver enzymes have also noted that this herb, hypericum, has a great number of modulatory effects from the array of chemicals in the whole plant extract. This means that while some chemicals would exert an inhibitory or competing effect on the P450 catabolism, other chemicals in the herb would soon modulate this effect, thus protecting the organism. In these studies, St. John's Wort did exert changes on the P450 catabolism that changed circulating levels somewhat during the first week in vivo, but had the opposite effect during the second week of taking the herb, due to chemicals that aided overall liver metabolism, resulting in an optimum catabolic clearance over time. Now, when we look at our physician's desk reference, we see many drugs that are competitive for various P450 catabolic pathways, and recommendations that these drugs not be prescribed concurrently, or with caution. These recommendations are so prevalent that often they are overlooked in prescription today where the patient is prescribed a number of pharmaceuticals, often by different M.D.'s, and often with prescription of drugs that have been introduced to the market relatively recently. The problem with drug-drug interactions seems to be handled too often with less than optimal scrutiny, yet the published warnings of drug-herb interactions, almost entirely based on a few studies of St. John's Wort are widely distributed.

What is almost always missing from reviews of St. John's Wort on drug-herb interactions is the manner in which the various chemicals in the whole plant extract is modulatory and protective of the patient. Besides having a balance array of chemicals in the herb, some of which protect the organism from excess inhibition of genetic expression, the same way that these chemicals protect the St. John's Wort plant itself, there are chemicals that are proven to help the body modulate potential ill effects in other ways. One way that St. John's Wort protects the organism is by stimulating an increased immune response to modulate unwanted metabolic side effects. The immune system is very responsive and regulatory for much of our common organ function, and St. John's Wort is proven to be immunostimulatory. Click here to see a 2002 in vitro study by the State University of New York, Upstate Medical University, in Syracuse: http://www.liebertonline.com/doi/abs/10.1089/107555302760253667. St. John's Wort also contains a number of chemicals that modulate hormone metabolism, which itself is protective of unwanted metabolic changes in the body. To see the array of chemicals and proven biological acitivities in a herb or food, go to Dr. Duke's Phytochemical and Ethnobotanical Databases, sponsored by the U.S. government. We see at this database that St. John's Wort contains such chemicals as hyperoside, that is antihepatotoxic, cAMP-inhibitor IC50=0.14 mg/ml, hepatoprotective IC80=30 ug/ml, xanthine oxidase inhibitor, antioxidant, and diuretic, all of which could work to optimize drug biotransformation and clearance. A number of other chemicals are found in St. John's Wort that also could be beneficial to liver metabolism and enzyme balance, including quercetin, a potent antioxidant that is a tyrosinase-inhibitor, P450 inducer and P450 inhibitor, allelochemic, anticariogenic, etc. This plant has evolved a metabolism of high levels of protective chemicals to insure its survival, and these chemicals translate into protection of homeostasis in the human organism.

Besides the 2008 in vivo study cited above that found no problems with St. John's Wort in competition at the P450 catabolism, the numerous studies cited are all in vitro studies with flawed study design and inherent bias. One 2008 study at the University of Tokyo Department of Pharmacy tries to correct for study bias and inaccuracy with new methodology, and compares the P450 effects of St. John's Wort to a number of pharmaceuticals in study to see which chemicals have the potential to reduce plasma concentrations of competing pharmaceuticals to levels considered ineffective. Seven chemicals were analyzed statistically from the data collected in 42 DDI studies and only rifampin, phenytoin and carbamazepine reduced the plasma concentrations of other drugs competing in the CYP3A4 pathway to ineffective plasma levels, not St. John's Wort. To see the summary of this study click: http://www.ncbi.nlm.nih.gov/pubmed/18783297?ordinalpos=1&itool=EntrezSystem2.PEntrez.Pubmed.Pubmed_ResultsPanel.Pubmed_DefaultReportPanel.Pubmed_RVDocSum. There is no hesitancy to prescribe Tegretol, one of these altering drugs, for an increasing array of problems, such as neuralgia and bipolor disorder.

Hepatoprotective or inhibiting of liver function?

This brings us to an interesting question. Are the chemicals in total in the herbs studied that effect the P450 catabolism balancing and protective of liver homeostasis, or dangerous and destructive of this homeostasis? We are talking about the whole array of chemistry evolved in the plant, and not a standardized extract that significantly boosts content of one particular chemical. We are also not talking about the pharmaceutical variants of the plant chemistry devised by the current biochemical industry, or products that utilize a patented method of boosting a specific herbal chemical in a plant to unnatural potency.

Numerous phytochemical studies over the decades have demonstrated how herbs and foods contain chemicals that are preserving of optimal homeostasis and are hepatoprotective. One example is the Indian food spice, Trachyspermum ammi, called Carum copticum seed, or Ammi copticum, or simply ajowan. Ayurvedic medicine, like ancient Chinese medicine, has promoted the use of various medicinal herbs in common food preparation for centuries to benefit public health. A study published in the Journal of Ethnopharmacology in 2005 describes the hepatoprotective mechanisms in this food spice. To see the summary, click here: http://www.ncbi.nlm.nih.gov/pubmed/15763373?dopt=Abstract. As we see, this seed extract shifted dose response curves to the right, protecting the patient from excessive dose response, and prevented unhealthy induced rise in liver transanimases that are associated with liver dysfunction. The presence of a calcium channel antagonist also protected against unwanted side effects that could be common in pharmaceutical overdosage.

A number of the herbs that are studied to find evidence of drug-herb interactions are specifically the herbs that became popular because of their dramatic hepatoprotective roles in modern medicine. As more patients in the 1990's took the harmful HIV cocktails with AZT and protease inhibitors, sudden dramatic rises in liver enzymes became common, as did metabolic syndrome, insulin resistance, peripheral neuropathy and a host of serious side effects linked to liver dysfunction. The patients found that taking a small dose of milk thistle for a short course would usually bring the liver enzymes to normal. Similar effects of hepatoprotection with concommittant use of pharmaceuticals was observed with kava-kava and St. John's Wort. Use of goldenseal seemed to prevent gallstones and liver cirrhosis, with content of berberine and chlorogenic acid, an allelochemic, antihepatoxic, cholagogue, choleretic, anti-inflammatory immunostimulant. Rise in popularity and reports of significant clinical benefit seemed to stimulate pharmaceutical investment in research to undermine these herbal therapies. To date, there are almost no cases of actual clinical injury from drug-herb interaction with these herbs, yet there is a vast amount of information distributed to medical doctors, pharmacists and the general public with dire warnings.

The effects of herbal chemicals on the circulating level of Coumadin, or warfarin

Coumadin, or warfarin, is biotransformed by the CYP3A4 and CYPa2 catabolic pathways located in red blood cells that produce NADPH-CYP regulatory enzymes. Warfarin places a dramatic stress on the P450 pathway and problems have been significant in clinical cases with comcommittant use of other drugs, such as quinidine derivatives used for patients with cardiovascular pathologies and atrial fibrillation. The problems resulted in deficiency of prothrombins in erythrocyte production and subsequent hemorrhage or bleeding pathologies, such as purpura. Also, long term use of warfarin has resulted in numerous cases of blood dyscrasia, with deficient or excess production of platelets and prothrombins. These cases are so numerous that many medical doctors and researchers looked for alternatives to warfarin prescription, and finally the FDA advised that low dose aspirin was just as effective in clinical study for chronic use to prevent thromboses. Subsequently, many foods and herbs have been mentioned as potential alterers of warfarin circulating effective levels and warnings issued. These warnings include such common foods as broccoli, pineapple, garlic, avocado, onion, brussel sprout, etc. Herbal warnings have been alarming. Less alarming are the serious contraindications with an array of common pharmaceuticals proven to cause problems, which include 6 common antibiotics, 5 cardiac drugs, hypertensive agents, cimetidine, piroxicam, phenylbutazone, TB drugs, 3 sedatives and anxiolytic agents, cholestyramine and sucralfate, as well as alcoho. To confirm these findings, click here: http://www.annals.org/cgi/content/full/121/9/676.

Evidence suggests that there are three definitive problematic interactions, from phylbutazone, sulfinpyrazone, and griseofulvin, and a host of drugs that showed evidence of inhibiting effect, including nafcillin, rifampin, cholestyramine, barbituates, carbamazepine, chlordiaepoxide, sucralfate, and dicloxacillin. There was also a host of drugs that were proven to create potentiating effects of warfarin, including erythromycin, fluconazole, miconazole, metronidazole, cotrimoxazole, and isoniazid, as well as ciprofloxacin, tetracycline, and 6 other antibiotics. Several cardiac drugs showed evidence of potentiating effects, including propranolol, clofibrate, amioodaraone, sulfinpyrazone and clofibrate. Simvastatin, quinidine, and acetylsalicylic acid, or aspirin, also had evidence of harmful potentiation. Now most of these drugs that produce serious drug-drug interactions are routinely prescribed without much apparent consideration concerning drug-drug interactions, yet medical doctors are routinely cautioning patients to avoid herbal remedies at all when taking coumadin, while not discontinuing these pharmaceutical prescriptions. Evidence that foods inhibited warfarin was found only for large amounts of avocado. Now very few people eat large amounts of avocado. The study cited does not confirm evidence clinically of any herb that produced medical problems.

Surely, by now, you have read enough information that at least casts some doubt on all the information given to medical doctors over the last 5-10 years at pharmaceutical luncheons et al, and published prominently in both medical journals and People magazine.

Where do we proceed from here?

There is a great potential in Complementary Medicine to actually use these scientific studies to logically address some changes to treatment protocol that would benefit the practice of the medical doctor and the health of the patients. By working with Complementary physicians, instead of against them, the practice of medicine can be adjusted to arrive at the most sensible and least harmful overall treatment strategy. Harmful drug-drug interactions can be decrased, and side effects diminished, by communicating treatment possibilities between the medical doctor and the Licensed Acupuncturist and herbalist. Everyone can benefit if we work together.

Information Resources

1. To easily see the array of chemicals and proven biological activities in a herb or food, click here to access Dr. Duke's Phytochemical and Ethnobotanical Database. A look at these biological acitivities often shows how the herb or food has evolved an array of modulatory effects to minimize potential harm to the organism: http://www.ars-grin.gov/duke/
2. An example of medication breakdown, or catabolism, in the liver, which produces even more harmful chemicals than the medication itself, is cited in this study of AZT catabolites: http://molpharm.aspetjournals.org/cgi/content/abstract/39/2/258. Often, the patient needs increased stimulation of metabolic pathways and supplementation of potential nutrient deficiencies to optimize the clearance of these harmful metabolites. Herbal medicine, nutrient prescription, and acupuncture can help your patient reduce side effects and preserve general health in these ways.

The information on this website is not intended to be used as a specific medical advice or cure. Please consult with the practitioner or an appropriate physician, such as a licensed acupuncturist, naturopath, or medical doctor, to discuss the proper application of the information contained on this website.