As I write this, the American news cycle is firmly focused on the issue of drug harms. It’s in the headlines not because of the thousands of cases of drug toxicity, hospitalizations, and even deaths that are documented each year, but because of the untimely death of singer Whitney Houston. While the cause of Houston’s death has not yet been identified,prescription drugs and alcohol are suspected to have played a role. If that’s the case, she’ll join a long list of celebrities whose deaths have been attributed to the abuse of prescription drugs. Over at Natural News, Mike Adams has already added her name to the list of “celebrities killed by Big Pharma“. He elaborated on drug-related deaths back in 2009 when actor Brittany Murphy died, deeming her death to be due to “Acute Pharmaceutical Toxicity“:

As you already guessed, there’s a fatal flaw in this pharmaceutical approach to sick care: Pharmaceuticals have never been tested in combination with other drugs. So all the so-called “gold standard science” is absolutely worthless at knowing what might happen when half a dozen pharmaceutical drugs are combined in a patient’s body. Brittany Murphy may have been on as many as TEN drugs!

Despite the fact that no combination testing has ever been done on pharmaceuticals, they are regularly prescribed in combination. Obviously, this creates a whole new realm of unknown risk based on the way multiple drugs might chemically interact in the human body.

The more pharmaceuticals you take, the more dangerous they become. While one pharmaceutical chemical may at first seem harmless (even though just one drug can actually kill you), when you start adding a second, third, fourth and fifth prescription on top of that, you’re dealing with Acute Pharmaceutical Toxicity (APT) that’s never even been tested in clinical trials.

Pharmacists are trained to help people avoid the most toxic two-drug combinations, but they rarely have any real knowledge about what happens when you combine three, four, five or more drugs. No one does. The science has simply never been done on that question. It’s no wonder: With all the possible combinations and permutations of pharmaceutical toxicity, it would take literally trillions of clinical trials to test them all.

Adams’ ignorance of medicine is obvious here. Combinations of drugs are studied in clinical trials all the time. You can start with the HIV treatments, move on to cancer drugs, and then chronic illnesses to see studies examining two, three and more in combination. But if a particular combination hasn’t been studied, are we still in the realm of science-based medicine? Alternative health proponents, sensitive to the lack of evidence supporting their preferred treatments, see drug combinations as just one example ofSBM hypocrisy. We’re told that “only 10–35% of medical practice is based on randomized controlled trials” as a justification for unproven or disproven treatment strategies.While this particular statistic has been repeatedly examined and debunked, and the risks of polypharmacy have been discussed at SBM, a science-based approach to combining drugs, even in situations when they haven’t been directly studied in clinical trials, hasn’t gotten as much attention.

It’s important to acknowledge that adverse reactions from prescription drugs are a major cause of harms and death. In 2008 poisonings caused more deaths than car accidents, and many of these poisonings were from prescription, not illegal, drugs. It’s been estimated that adverse reactions may cause 2-6% of hospital admissions, and that proportion may be even higher in specific age groups. Each case of drug-related harm has its own set of contributors, which may include health professional culpability, a lack of proper education, and patient factors (including situations of deliberate abuse). The tragedy is not just their absolute numbers, but the fact that many of these events are both predictable and avoidable — particularly those that result from combining prescription drugs, or mixing prescription drugs and alcohol. As Harriet Hall has noted before, it’s misleading to say that combining drugs can’t be evidence-based. How different drugs interact when combined in the body isn’t a scientific black box: An understanding of drug kinetics and of molecular biology allows us to predict with fair accuracy how drugs will behave when combined. So while we cannot anticipate all idiosyncratic reactions to drugs, there already exists the knowledge and tools to be doing a much better job preventing drug-related harms.

There are two main types of drug interactions:

Pharmodynamic interactions change the effect of a drug, without changing the amount of drug in the body. The celebrity overdose is a common example: combining multiple drugs that can depress and impair the central nervous system can lead to significant sedation and even death. These outcomes are not a surprise — they are a direct extension of their pharmacologic action. We can use pharmacodynamic interactions in more positive ways, too: Different types of antihypertensives work in different ways, so a combination of drugs may be effective in lowering blood pressure when raising the dose of a single drug is ineffective or causes unwanted side effects. But not in all cases. Combining drugs like ACE-inhibitors with potassium-sparing diuretics may lower blood pressure, but both drugs also elevate potassium levels, sometimes to life-threatening levels. Another example of pharmacodynamics is the treatment of pain with a narcotic such as codeine, plus an anti-inflammatory,such as naproxen. Both provide pain relief, albeit by different mechanisms of action.

Pharmacokinetic interactions are the result of one drug affecting another drug’s action by modifying its concentration at the site of action. This can be accomplished in four different ways:

1. A change in drug absorption

Unless it’s injected, a drug needs to be absorbed (usually from our gastrointestinal tract). Modifying the environment (say, reducing stomach acidity with a proton pump inhibitor like Prilosec) can modify how extensively some drugs are absorbed, if that absorption is dependent on an acidic environment. Or we can influence absorption by slowing down or speeding up the motility of your gastrointestinal tract: Change transit time, and you can change the extent of absorption. Or kill off some bacteria in your colon antibiotics, and it may reduce the circulation of drugs that are secreted in the bile and then reabsorbed. It doesn’t have to be the stomach, either. Drugs administered via the skin can have their absorption affected by solvents like DMSO.

Cellular pumps allow drugs and other chemicals to cross otherwise impermeable barriers. P-glycoprotein is a cellular pump located in the intestine, kidney, liver and blood-brain and blood-testes barrier. Loperamide (Imodium) is an anti-diarrheal that is structurally similar to narcotics, but lacks the usual narcotic effects on the brain, because p-glycoprotein blocks it from crossing the blood-brain barrier. When the function of one drug depends p-glycoprotein, and another drug modifies its action, unpredictable effects can results.

2. A change in drug distribution

Drug molecules don’t just float along in the bloodstream — they may hitch a ride on proteins, binding to them and effectively decreasing the amount available for activity at the site of action, or in the liver, when the drugs will be transformed (and eliminated). The impact of one drug on the protein binding of another can modify a drug’s action. If one drug displaces another from its protein binding, this can raise the effective or “free” levels in the blood, potentially causing a toxic effect.

3. A change in drug metabolism

Drug metabolism is the process that convert drugs (and other chemicals) into molecules that can more easily be excreted from the body, usually by way of the kidneys. These transformations are catalyzed by enzymes, and some drugs act to inhibit or induce enzyme action. In particular, a group of enzymes called cytochrome P450, or CYP enzymes, are the main metabolic pathway for many drug products. While there are dozens of different CYP enzymes, only a handful act on drugs. Consequently, if we know a specific drug is metabolized by, inhibits, or induces a particular CYP enzyme, it becomes easier to predict the possible effects on other drugs — without the need for direct evidence to verify the interaction.

Many drugs are well absorbed from the gastrointestinal tract, yet fail to appear in high levels in the blood circulation. It’s the liver at work, causing what’s called the “first pass” effect — metabolizing drugs as they first pass through the liver, before they circulate in the body. Inhibition of certain enzymes, particularly CYP3A4, can dramatically suppress this first pass effect, potentially changing usual doses into overdoses.

4. A change in drug elimination

Drugs can induce elimination of other drugs, reducing peak levels and duration of effect, possibly to an extent that efficacy is compromised. Or they can inhibit elimination, raising peak levels and the duration of effect, possibly to toxic levels.

Drug-Food Interactions

We tend to separate drug-drug interactions from drug-food interactions, but from a biochemical perspective, it doesn’t matter: foods are just combinations of chemicals, some of which may interact with drugs. And the mechanisms for these are the same:

  • pharmacodynamic effects: alcohol can increase the sedating effects of narcotics, antihistamines, and sedatives like benzodiazepines
  • absorption issues: calcium and iron can reduce drug absorption, while some drugs are absorbed more extensively in the presence of food.
  • drug kinetic issues: vitamin-K containing foods can antagonize the effects of warfarin; tyramine-containing foods combined with the (now rarely used) monoamineoxidase (MAO) inhibitors can cause a massive increase in blood pressure and even stroke. Grapefruit juice has been identified as a significant cause of drug interactions, through inhibition of CYP enzymes.

Drug-Herb Interaction

Herbal products can be a nightmare from a drug interaction perspective. In general, herbs raise the level of therapeutic uncertainty and risk, compared to drug-drug interactions. Compared to the relative straightforward data on drugs with known pharmacokinetics and predictable interactions, herbs can contain many different chemicals, of which the “active” ingredient(s) may not even be known. Combine the lack of standardization, and the possibility of poor quality control standards, and you’ll see most pharmacists wince when you ask about drug-herb interactions: the unknowns make combining herbs and drugs potentially risky, especially in situations where the drugs have a narrow “therapeutic index“, or when the stakes are high, such as cancer chemotherapy.

While we don’t have good estimates of their true prevalence, we know that some herbs are demonstrably problematic: St. John’s Wort (Hypericum perforatum) (SJW) can cause significant drug-drug interactions, including HIV drugs and transplant therapies. There are multiple cases of transpanted organ rejection linked to initiation of SJW. Why? SJW is a powerful inducer of CYP3A4 enzyme, which increases the metabolism of immunosuppressants, decreasing their effects.

What’s most frustrating about drug-herb interactions is that the natural products industry seems determined to keep consumers in the dark about the potential harms, as Mother Jones outlined this week, in a column entitled What the Supplement Industry Isn’t Telling You About St. John’s Wort:

The real problem here lies in transparency to consumers—a problem that goes directly back to the supplement’s manufacturers. In a 2008 study published in BMC Complementary and Alternative Medicine that tested 74 different SJW brands, less than a quarter of the product labels identified possible interactions with antidepressants. Even more disturbing was that only 8 percent identified possible interactions with birth control.

Many groups, like the Center for Science in the Public Interest, have tried to push the FDA to standardize SJW labels to properly reflect possible dangers. But since supplement makers are not required by law to warn consumer about health risks associated with their products, it hasn’t been easy. “These companies fight warning labels like the dickens, and whether they intend it or not, that affirms the belief that natural products are unequivocally good for you,” says Stephen Gardner, litigation director at CSPI.

And that’s the issue. It’s not that drugs are inherently harmful, and herbs are wonderful and safe panaceas. Any product, whether it’s a herb or a synthetic drug, has the potential to harm, and to interact negatively. And unless the potential for interactions is well understood, we need to approach the combination of herbs and drugs with caution. Part of the solution is ensuring that health professionals and consumers alike are asking the right questions about their safety and efficacy.


One of the most interesting areas in drug interactions and drug safety is our evolving knowledge of pharmacogenomics: how genetic factors can influence drug behavior. The vision for pharmacogenomics is to maximize efficacy and to avoid adverse drug reactions. The reality is that we’re not there yet, but the science is progressing. For example, the dosing of warfarin (Coumadin), a drug associated with a significant bleeding risk, can be modified to minimize toxicity based in part on testing for the genes CYP2C9 and VKORC1. While it doesn’t eliminate the dosing uncertainty, and the clinical usefulness of testing remains limited, it’s a promising sign of what may become a more predictable way of selecting and dosing drugs. Over time, our accuracy at drug selection and dosing may become much more personalized than it is today.

Where do we go from here?

Harms associated from combining prescription drugs neither validates alternative medicine, nor invalidates science-based medicine. Celebrity or otherwise, many of the harms attributed to prescription drugs are predictable and avoidable. Every treatment decision boils down to an individual evaluation of risk versus benefit, and we can combine therapies with a great degree of confidence based on our understanding of how they will (or will not) interact. Drug-related injuries and toxicity are a real issues, one that medical systems could be doing a much better job of addressing. But when it comes to understanding how these harms are occurring, and preventing them, it’s not just the drugs we need to look at: we already have the information, technology, and capacity to significantly reduce the occurrence of drug-related harms.


Posted by Scott Gavura