Preventing Toxic Drug Interactions and Exposures

Donald A. Moffa Jr.

Published: August 2010

As practicing physicians, we are charged with caring for our patients with compassion and rigor. Our prescribing privilege is a powerful tool to help heal our patients and a toxic means to harm them if we are careless, lack knowledge of potential drug interactions, or disregard this knowledge.

More than 3200 prescription drugs, 300 dietary supplements, and 600 herbal products occupy pharmacy shelves in the United States,1 and more than one half of our patients cannot recite an accurate list of their medications.2 This creates a regrettable scenario when we write prescriptions for patients with whom we have little history, we see in our partners’ absence, or we rush to treat in a busy emergency department with the goal of easing disease while assuring safety in our treatment methods.

We do our best to avoid adverse drug interactions. However, for patients taking two medications, the risk of a drug interaction is 15%. This risk rises to 40% for those taking five medications and to an alarming 80% for patients taking seven or more.3 The risk of a toxic medication interaction is real considering that more than one half of noninstitutionalized adults older than 65 years take five or more different medications, and 12% use 10 or more.4 In hospitalized patients, adverse drug interactions are estimated to be as high as the fourth leading cause of death.5


The most dangerous drug combinations in the nursing home population involve warfarin interactions with nonsteroidal anti-inflammatory drugs (NSAIDs), sulfonamides, macrolides, or quinolones; angiotensin-converting enzyme (ACE) inhibitor interactions with potassium supplements or spironolactone; digoxin interactions with amiodarone; and theophylline interactions with quinolones.6 Toxic drug interactions can occur any time two or more medications compete is such a way that their pharmacologic interaction causes a detrimental physiologic response, when a medication is prescribed in excessive amounts, or when one medication produces untoward consequences although it is prescribed according to established guidelines. This last effect is often the most difficult to predict because drug absorption and metabolism can vary with age, concomitant illness, gastric motility, pH of the gastrointestinal milieu, genetic variation, smoking, or some other obscure physiologic parameter.

Back to Top

The Cytochrome p-450 Monooxygenase System

The system responsible for most drug metabolism in the human body is the cytochrome P-450 monooxygenase system (CYP-450). Composed of more than 30 heme-containing isoenzymes, the CYP-450 is embedded in the lipid bilayer of the endoplasmic reticulum of hepatocytes, enterocytes of the small intestines, and, in lesser quantities, in other tissues including the kidneys, lungs, and brain.2 The major isoenzymes responsible for drug metabolism in humans are CYP3A4, CYP2D6, CYP1A2, and CYP2C; CYP3A4 and CYP2D6 account for the bulk of drug metabolism. CYP3A4 is the most abundant CYP-450 isoenzyme in the small intestine (responsible for much first-pass drug metabolism) and in the liver7 and accounts for more than 50% of drug metabolism.2

Some drugs are metabolized by more than one CYP isoenzyme, so that inhibition in its metabolism by one pathway can lead to a compensatory increase in its metabolism by another. Moreover, a drug that is metabolized by one CYP pathway can, at the same time, inhibit another CYP isoenzyme such that the inhibition of the second pathway leads to the toxic accumulation of a drug that is normally metabolized by the second pathway.2

Genetic polymorphism affects drug metabolism such that some persons are extensive metabolizers and others are poor metabolizers of medications. From 5% to 10% of whites and 1% to 3% of Asians and African Americans are poor metabolizers by the CYP2D6 isoenzyme, and poor metabolizers by this pathway are at risk for tricyclic antidepressant (TCA)-induced cardiotoxicity and neuroleptic-induced side effects.8 Prodrugs that are normally converted to their active metabolites (e.g., codeine requires biotransformation to morphine) may be ineffective in poor metabolizers. CYP2C9, CYP2C10, and CYP2C19 make up the metabolically active CYP2C isoenzyme subfamily, and the CYP2C10 isoenzyme exhibits genetic polymorphism, and 20% of African Americans and Asians and up to 5% of whites are poor metabolizers. Of course, whether or not your patient is a poor metabolizer is not apparent, nor may it be clinically relevant. Measuring the O-demethylated metabolite of dextromethorphan in the urine will help determine who is an extensive metabolizer and who is a poor metabolizer.7

Cytochrome P-450 Isoenzyme Induction and Inhibition

Drug metabolism by the CYP-450 system is important to prevent accumulation of medications and substances toxic to the body, but the ability to induce the CYP-450 system can decrease with age or with organ dysfunction (e.g., cirrhosis, hepatitis).7 Isoenzyme inhibition by compounds that compete with the primary drug for the isoenzyme binding site can last several days, and the rate-limiting step in reversal of the inhibition can depend on isoenzyme turnover itself. The CYP-450 isoenzyme half-life ranges from 1 to 6 days7 and may be inhibited or induced by secondary medications.

Isoenzymes can also be affected by the consequences of other isoenzyme interference. For example, warfarin is a compound of R- and S-enantiomers, and the S-warfarin enantiomer has a significantly greater anticoagulant effect. CYP1A2 metabolizes R-warfarin and CYP2C9 metabolizes S-warfarin, and R-warfarin inhibits CYP2C9. It is not a long stretch to imagine that any drug inhibiting CYP1A2, the isoenzyme that hydrolyzes the R-warfarin enantiomer, can secondarily inhibit S-warfarin metabolism by the accumulation of R-warfarin, increasing the anticoagulation effect of a seemingly stable warfarin dose. In this example, a medication such as ciprofloxacin enhances warfarin’s own ability to anticoagulate.

CYP-450 enzyme synthesis may be stimulated by increases in hepatic blood flow (such as with hepatic enlargement by phenytoin therapy), and isoenzyme function may be enhanced for more rapid drug metabolism by induction from other substances (such as CYP1A2 induction by cigarette smoke or charcoal-broiled foods).7Notwithstanding isoenzyme inhibition or induction, many commonly prescribed medications pose challenges for the physician to avoid causing toxicity when given in combination with other medications.

Back to Top


Warfarin combined with other medications accounts for the most common dangerous drug interactions.6 Warfarin is prescribed with increasing frequency in patients with vascular stagnation states (e.g., atrial fibrillation, ventricular aneurysm), peripheral vascular disease leading to limb ischemia and stroke, and hypercoagulable conditions causing venous thromboembolic events (e.g., deep venous thrombosis, pulmonary embolism). Warfarin toxicity can lead to life-threatening intracranial and gastrointestinal hemorrhages, and the physician should pause to think about medication interactions when prescribing any drug, especially antibiotics and anti-inflammatories, to a patient taking warfarin.

Among the list of medications that inhibit warfarin metabolism and enhance its anticoagulation effects are cimetidine, selective serotonin reuptake inhibitors (SSRIs), antifungals (fluconazole, itraconazole, ketoconazole), erythromycin, (possibly clarithromycin), omeprazole, ciprofloxacin, norfloxacin, trimethoprim-sulfamethoxazole (TMP-SMX), and amiodarone.2,7 Cimetidine is a potent CYP-450 inhibitor, confounding the metabolism of many medications including warfarin by CYP1A2 inhibition of R-warfarin metabolism. Omeprazole inhibits R-warfarin metabolism and can increase anticoagulation, but lansoprazole does not and may be considered an alternative treatment in the appropriate patient needing a proton pump inhibitor.7

Of the antibiotics that enhance warfarin’s anticoagulation effect, erythromycin, ciprofloxacin, and TMP-SMX appear to have the most predictable interaction. Concomitant erythromycin use can cause a twofold increase in the international normalized ratio (INR) after 7 days of treatment. Azithromycin does not appear to increase warfarin’s anticoagulant effect, but clarithromycin should be prescribed with caution because it can increase the risk of bleeding.7 Quinolones can interact with warfarin after 2 to 16 days,7 and ciprofloxacin can cause life-threatening hemorrhage in patients formerly stable on warfarin therapy.9 Ofloxacin is less likely than ciprofloxacin to cause a dangerous interaction with warfarin, and levofloxacin and lomefloxacin are alternative quinolones with little or no effect on warfarin metabolism.7 TMP-SMX enhances anticoagulation by inhibiting CYP2C9, which metabolizes S-warfarin and can cause serious bleeding in combination with warfarin.2 Nonetheless, consider checking the INR more frequently, perhaps every other day, in patients for whom these antibiotic combinations with warfarin cannot be avoided.

Prescribing warfarin with other medications that inhibit coagulation or platelet function is risky and should only proceed when a well-defined end point supersedes the chance of causing hemorrhage, such as when low-molecular-weight heparin is given concurrently as a bridge to full anticoagulation therapy with warfarin. Synergy from the antiplatelet effect of NSAIDs and the anticoagulant effect of warfarin increases the risk of bleeding and usually occurs from the accidental combination of the two classes of medications, and although the risk of bleeding is less with cyclooxygenase (COX)-2 inhibitors than with traditional NSAIDs, the risk remains significant. Celecoxib competitively inhibits CYP2C9, the isoenzyme that metabolizes S-warfarin, and significantly increases INR and bleeding risk.10,11

Amiodarone, which is metabolized by CYP2C9 and decreases the total body clearance of both R- and S-warfarin, can increase bleeding. The increased anticoagulation effect begins 1 to 2 weeks after starting amiodarone and can last up to 3 weeks after discontinuing the antiarrhythmic. Consider reducing the warfarin dose by 25% when giving it in combination with amiodarone.7

Some medications can induce warfarin metabolism, thereby reducing the risk of bleeding. Failed anticoagulation has been observed when rifampin, carbamazepine, phenobarbital, or phenytoin is given in combination with warfarin. These medications induce S-warfarin metabolism.7

Back to Top


Digoxin serum concentration has a narrow therapeutic range. Its absorption is affected by gastrointestinal pH, intestinal contents, motility, and blood flow. The kidneys excrete digoxin. Any medication with anticholinergic properties that slows gastrointestinal motility or any condition that impairs digoxin excretion can increase digoxin’s concentration and the risk of digoxin toxicity. Patients at highest risk for digoxin toxicity are those who have renal insufficiency, congestive heart failure, and dehydration. Antibiotics such as clarithromycin, erythromycin, and tetracycline can alter the gut flora and increase digoxin levels, as can other medications that reduce renal clearance, such as quinidine, amiodarone, and verapamil.2 Consider reducing digoxin dosing by 50% when giving it with amiodarone, and monitor levels.

Back to Top


Medications that inhibit phenytoin metabolism (CYP2C) and can favor phenytoin toxicity include cimetidine, omeprazole (impairs phenytoin elimination after 8 days), fluconazole, isoniazid, topiramate, and fluvoxamine.7 In a similar fashion of isoenzyme induction with warfarin, rifampin induces phenytoin metabolism by the CYP2C isoenzyme and can cause therapeutic failure. It sometimes seems, however, that the most likely cause for phenytoin failure is not its interactions with other medications but our patients’ failure to take the anticonvulsant in the first place.

Back to Top


Theophylline is metabolized by CYP1A2. Cimetidine decreases theophylline clearance by 30%, and it takes approximately 2 days for theophylline to reach its new steady state when given in combination with cimetidine.12 Other medications that interfere with theophylline metabolism and increase its serum concentration are erythromycin and clarithromycin (which decrease theophylline clearance by approximately 25% after 7 days), the quinolones ciprofloxacin and norfloxacin, isoniazid (after at least 6 days of coadministration), fluvoxamine (increases theophylline concentrations two- to threefold), and oral contraceptives (which decrease theophylline clearance by 30%). As an alternative to reducing theophylline dosing and diligently measuring its serum concentration, consider prescribing other antibiotics (azithromycin, dirithromycin, oflaxacin, levofloxacin, lomefloxacin, other tuberculosis therapy), other psychiatric medications, and alternative means of contraception.7 Rifampin, carbamazepine, phenobarbital, phenytoin, and cigarette smoke induce theophylline metabolism, thereby decreasing theophylline’s serum concentration. Cigarette smokers might need twice the usual theophylline dose to achieve a therapeutic concentration; the enzyme induction effect can last for several months after smoking cessation.7

Back to Top

Antidepressants, Antipsychotics, Benzodiazepines, and Narcotics

Although little is known about the true mechanisms by which antidepressant and antipsychotic medications work, many interactions among them and other drugs are characterized, and toxicity of one or another substance occurs by affecting the CYP-450 system. Fluoxetine, paroxetine, and sertraline, to a lesser extent, inhibit the CYP2D6 isoenzyme. Coadministration of any of these can increase the plasma concentrations of each, perhaps to toxic levels.7 Fluvoxamine inhibits CYP1A2 and increases the plasma concentrations (with accompanying clinical symptoms) of amitriptyline, clomipramine, clozapine, desipramine, imipramine, and haloperidol.7Fluoxetine can cause delirium when given in combination with clarithromycin.7 Avoid giving SSRIs to patients taking class Ic antiarrhythmics.7

Interactions with benzodiazepines are often unexpected but may be predictable, because benzodiazepines have a significant inhibitory effect on CYP-450 isoenzymes, and other substances that induce the isoenzymes can diminish the benzodiazepine’s intended effect. Smokers experience less drowsiness than nonsmokers when taking chlordiazepoxide or diazepam due to CYP1A2 isoenzyme induction from cigarette smoke. Rifampin induces benzodiazepine metabolism (CYP3A4) and can decrease its effect. Omeprazole inhibits diazepam metabolism (CYP2C) and increases its elimination half-life by 130%.7 Fluoxetine and fluvoxamine increase the serum concentration of alprazolam, midazolam, and triazolam (CYP3A4 inhibition) and can potentiate psychomotor effects.7 Grapefruit juice increases midazolam peak serum concentrations by 50%,13 whereas nefazodone increases alprazolam and triazolam serum concentrations.7 Consider decreasing dosages by as much as 75% of the usual dose when giving benzodiazepines in these combinations.

Similar to concurrent benzodiazepine use, rifampin increases the rate of opioid metabolism (CYP2D6) and may induce narcotic withdrawal symptoms.7 Codeine is a prodrug that requires demethylation to its active form. Demethylation by the CYP-450 system is impaired in poor metabolizers and may be inhibited in extensive metabolizers who are taking drugs that compete for the same metabolic pathway, thereby reducing codeine’s analgesic effect.7 However, inhibition of the CYP3A4 metabolic pathway can enhance the effects of some narcotics. For example, surgical patients taking erythromycin who are also given alfentanil might experience prolonged respiratory depression.14 Cimetidine doubles fentanyl’s elimination half-life (CYP3A4 inhibition) and decreases meperidine clearance (CYP2D6 inhibition).7Meperidine’s metabolite, normeperidine, is renally excreted and lowers the seizure threshold in patients with seizures. This can prove troublesome for some patients with renal insufficiency who are given meperidine for analgesia.

Back to Top

Cardiovascular Drugs and Statins

Erythromycin, itraconazole, and cimetidine inhibit the metabolism of many cardiovascular medications and can increase serum concentrations to toxic levels, causing clinically significant effects. Itraconazole and erythromycin increase felodipine levels (CYP3A4 inhibition) and cause clinically significant changes in systolic and diastolic blood pressures and heart rate. Rifampin given with verapamil causes a tremendous increase in verapamil clearance by inducing the CYP3A4 isoenzyme responsible for its metabolism. Other calcium channel blockers may be similarly affected. Quinidine inhibits CYP2D6 (affecting flecainide, mexilitine, propafenone, and propranolol metabolism) and is metabolized by CYP3A4. CYP3A4 interactions are well documented with cimetidine, phenytoin, phenobarbital, rifampin, metronidazole, and ciprofloxacin. Consider alternative therapy. HMG-CoA reductase inhibitors can cause a diffuse myopathy, and the greatest risk is when they are prescribed with other medications such as cyclosporine, gemfibrozil, niacin, itraconazole, and erythromycin that compete with the CYP3A4 isoenzyme pathway.7

Back to Top

Chemotherapeutic Medications

CYP3A4 metabolizes several chemotherapeutic medications including epipodophylotoxins, tamoxifen, paclitaxel, vinca alkaloids, and ifosfamide,7 the last of which requires biotransformation to its cytoxic metabolite and an inactive metabolite that, incidentally, is neurotoxic. Drugs that induce CYP3A4 (rifampin, carbamazepine, phenobarbital, phenytoin) can promote ifosfamide metabolism and, hence, neurotoxicity.7 CYP3A4 inhibitors including ketoconazole, itraconazole, diltiazem, verapamil, and cyclosporine can interfere with ifosfamide metabolism, but the clinical significance is not known.7

Vinblastine metabolism is inhibited by doxorubicin, etoposide, ketoconazole, and erythromycin, although the clinical significance is not known.7 Nifedipine increases vincristine’s elimination half-life fourfold, and toxicity is possible. Itraconazole (but not fluconazole) reduces busulfan clearance, making fluconazole an attractive choice for treating fungal infections in patients taking busulfan.7 Phenobarbital and phenytoin increase etoposide’s metabolism, and its metabolism is inhibited by cyclosporine.7Tamoxifen metabolism is inhibited by erythromycin, cyclosporine, nifedipine, and diltiazem. Cyclosporine decreases doxorubicin clearance by one half, increasing the likelihood of toxicity with signs including nausea, vomiting, and myelosuppression.7

Cyclosporine undergoes premetabolism in the small intestine, with further metabolism in the liver by CYP3A4 isoenzymes. Drugs (e.g., ketoconazole, verapamil, nicardipine, fluconazole, itraconazole, erythromycin, clarithromycin, tacrolimus) that inhibit CYP3A4 can increase cyclosporine serum concentrations. Cyclosporine degradation is increased by the CYP3A4 inducers rifampin, phenytoin, carbamazepine, and phenobarbital.7

Back to Top

Protease Inhibitors

Significant interactions occur with saquinavir, ritonavir, indinavir, and nelfinavir, all of which inhibit CYP3A4 isoenzyme. Ritonavir also inhibits CYP2D6 and reduces the clearance of medications metabolized by that isoenzyme, including benzodiazepines, calcium channel blockers, antidepressants, antiarrhythmics, corticosteroids, anticoagulants, opiates, and clarithromycin.7

Rifampin decreases saquinavir concentrations by isoenzyme induction so that patients concurrently treated for tuberculosis might require alternative antimycobacterial regimens.15 Other CYP3A4 inducers that can decrease protease inhibitor serum concentrations include phenobarbital, phenytoin, carbamazepine, dexamethasone, and tobacco.7

Back to Top

Estrogens and Corticosteroids

Clinically significant interactions with oral contraceptive pills (OCPs) from CYP3A4 isoenzyme induction that reduce OCP effectiveness are reported with carbamazepine, ethosuximide, phenobarbital, phenytoin, primidone, and rifampin. Antibiotic therapy can interfere with OCP effectiveness. In these cases, one may consider alternative contraception, medroxyprogesterone, or higher OCP doses.7 OCPs reduce prednisolone clearance and increase phenytoin concentrations.7 Therefore, consider frequent phenytoin concentration monitoring to avoid toxicity.

Back to Top

Angiotensin-converting Enzyme Inhibitors, Potassium-sparing Diuretics, and Potassium Supplements

Other toxicities can occur aside from involving the CYP-450 system. Life-threatening hyperkalemia can occur when patients take excessive potassium supplementation (especially in renal failure), when they are prescribed the combination of ACE inhibitors and spironolactone (a common regimen in treating congestive heart failure), or from other medications that either increase potassium absorption or decrease potassium clearance. Medications that can cause life-threatening hyperkalemia include amiloride, ACE inhibitors, beta blockers, cyclosporine, digoxin, heparin, NSAIDs, intravenous penicillin G potassium, pentamidine, potassium-sparing diuretics, spironolactone, succinylcholine, triamterene, and trimethaprim.2 Consider alternative medications in any patient who may be at risk for hyperkalemia.

Back to Top

Serotoninergic Agents

Patients taking serotoninergic medications are at risk for the potentially life-threatening complication of serotonin syndrome when taking any other medication that can increase serotoninergic activity. Serotonin syndrome involves excessive serotonin stimulation at the 5-HT1A receptor, both centrally and peripherally. Minor clinical features include neuromuscular, autonomic, and cognitive and behavioral excitation. More-severe cases can include seizure, coma, hyperthermia, rhabdomyolysis, organ failure, and cardiac arrest, and serotonin syndrome may be confused with neuroleptic malignant syndrome (NMS) or overdose.2 Medications that inherently increase serotoninergic activity include SSRIs, monoamine oxidase inhibitors (MAOIs), MDMA (3,4-methylenedioxymethamphetamine), cocaine, and tramadol. Dextromethorphan, SSRIs, and meperidine block serotonin reuptake, MAOIs inhibit serotonin breakdown, and buspirone and lithium act as serotonin precursors.2 All can contribute to serotonin toxicity, especially if given in combination or if taken surreptitiously.

Back to Top

Nonsteroidal Anti-inflammatory Drugs

NSAID use is not benign, and serious toxicity can occur even when taken in commonly advised doses. Adverse gastrointestinal effects, including bleeding from NSAID use, is estimated to account for the 15th most common cause of death in the United States.16 NSAIDs are on the Beers list of drugs to avoid in the elderly. One should prescribe NSAIDs for the elderly with extreme caution and understand the risk of significant hemorrhage when NSAIDs are concurrently taken with warfarin, other NSAIDs or antiplatelet agents, and sulfonylurea agents.2

Back to Top

Medications that Prolong the qt Interval

Medications that prolong the QT interval of the cardiac conduction cycle increase the risk of ventricular dysrythmias such as torsades de pointes. Notwithstanding their intended benefit, several culprit drugs have been withdrawn from the market because of their arrhythmogenic potential. These include terfenadine, astemizole, cisapride, and sparfloxacin. It is fortunate that similar medications within the same drug classes are available in their place. However, other medications can prolong the QT interval, including antiarrhythmics (amiodarone, disopyramide, procainamide, quinidine), calcium channel blockers (diltiazem, verapamil), antibiotics (gatifloxacin, levofloxacin, moxifloxacin, azithromycin, clarithromycin, erythromycin), antifungals (fluconazole, itraconazole, ketoconazole), antidepressants (amitriptyline, desipramine, imipramine, fluvoxamine, nefazodone), antipsychotics (droperidal, haloperidol, pimozide, thioridazine, ziprasidone), and protease inhibitors (delavirdine, indinavir, saquinavir, nelfinavir, ritonavir).2 Although there are no strict warnings against using these medications (except, perhaps, the black box warning for droperidol), it is reasonable to suggest avoiding them in combination to the allowable extent and to check baseline and subsequent electrocardiograms to follow the QT interval for signs of toxicity and the potential for dysrhythmia.

Back to Top

Medications in the Elderly and in Children

Prescribing medications to patients at the extremes of age is often troublesome for the physician. Children might not have developed the necessary metabolic pathways. The elderly, whose metabolism has slowed, might have confounding medical conditions that prevent drug absorption or elimination.

Considering the most common drug classes of medications dispensed to children (analgesics, antibiotics, antiepileptics, asthma and allergy medications, and psychotropic medications), approximately 15% of medications have a potential dosing error (including overdosing and underdosing with respect to dosing guidelines).17This could have toxic consequences. The most common medication errors for children are in prescribing analgesics; oxycodone is most commonly overdosed 15% of the time. Antiepileptics are the most commonly underdosed medication class (20% of the time). The potential for prescribing error for amoxicillin is 3% and 12% for cephalexin but 33% for azithromycin.17 It is, also, ill-advised to prescribe topical diphenhydramine to children because the large surface area on the skin for absorption and the unchecked distribution of the medication can lead to high serum levels, sedation, and obtundation.18-20

An estimated 30% of hospital admissions for the elderly are linked to drug-related problems or toxic drug effects.21 This might cost $85 billion and 106,000 deaths annually.21 Medication-related deaths, if categorized as such, may be the fifth leading cause of death in the United States.21 According to the National Healthcare Quality Report for 2005, the percentage of elderly Americans who take at least one of the 33 drugs considered to be potentially inappropriate for the elderly according to the Beers criteria21 dropped from 21.3% in 1996 to 18.4% in 2002.22 This shows improvement in physicians’ prescribing practices, but clearly there is more improvement to be made. Box 1 lists drugs at highest risk for causing adverse reactions when given to the elderly.

Box 1: Medications with Potentially High Risk of Adverse Effects in the Elderly
Amphetamines and Anorexic Agents
  • All except methylphenidate hydrochloride
  • Amiodarone
  • Disopyramide
  • Nitrofurantoin
Anticholinergics and Antihistamines
  • Chlorpheniramine
  • Cyproheptadine
  • Dexchlorpheniramine
  • Diphenhydramine
  • Hydroxyzine
  • Promethazine
  • Tirpelennamine
  • Amitriptyline
  • Fluoxetine, daily
  • Tirmethobenzamide
  • All except phenobarbital or when used to control seizures
  • Chlorazepate
  • Chlordiazepoxide
  • Chlordiazepoxide-amitriptyline
  • Clidinium-chlordiazepoxide
  • Diazepam
  • Flurazepam
  • Halazepam
  • Quazepam
  • Short-acting doses greater than:
    • Alprazolam (2 mg)
    • Lorazepam (3 mg)
    • Oxazepam (60 mg)
    • Temazepam (15 mg)
    • Triazolam (0.25 mg)
Gastrointestinal Antispasmotics
  • Belladonna alkaloids
  • Dicyclomine
  • Hyscyamine
  • Clidinium-chlordiazepoxide
  • Propantheline
  • Chlorpropamide
Laxatives (Long-term Use)
  • Bisacodyl
  • Cascara sagrada
  • Neoloid (except in combination with opiods)
  • Guanadrel
  • Guanethidine
  • Meprobamate
  • Mesoridazine
  • Methyldopa
  • Methyltestosterone
  • Mineral oil
  • Nifedipine
  • Thioridazine
  • Thyroid, dessicated
  • Ticlopidine
Muscle Relaxants
  • Carisoprodol
  • Chlorzoxazone
  • Cyclobenzaprine
  • Metaxalone
  • Methocarbamol
  • Orphenadrine
  • Oxybutynin
Narcotic Analgesics
  • Meperidine
  • Pentazocine
Nonsteroidal Anti-inflammatory Drugs
  • Indomethacin
  • Ketorolac
  • Long-term use of:
    • Naproxen
    • Oxaprozin
    • Piroxicam

Adapted Fick DM, Cooper JW, Wade WE, et al: Updating the Beers criteria for potentially inappropriate medication use in older adults: Results of a US consensus panel of experts. Arch Intern Med 2003;163(22):2716-2724 (table 1).

Back to Top

Grapefruit Juice

Grapefruit juice contains furanocoumarin derivatives that can inhibit the CYP3A isoenzymes. It appears that only the CYP3A isoenzymes in the mucosal enterocytes of the small intestine are affected and not those of the liver. The isoenzymes are both reversibly and irreversibly inhibited, and it can take up to 72 hours to regenerate them after ingesting only a small amount grapefruit juice. For grapefruit juice to significantly affect a medication’s metabolism, the drug must be a substrate of the CYP3A isoenzyme, have poor oral bioavailability because of extensive presystemic metabolism, and undergo extensive biotransformation by CYP3A in enterocytes. Significant interactions that raise circulating drug levels are unlikely, but the medications with the largest possibility for interaction are lovastatin, simvastatin, buspirone, and amiodarone.23

Back to Top


Knowledge of adverse drug interactions and toxic combinations is the framework for preventing iatrogenic harm. Knowing one’s patient is integral to avoiding prescribing errors. Nutritional state, alcohol consumption, cigarette smoking, herbal medications, even grapefruit juice can affect drug metabolism, leading to the possibility of drug toxicity. Several resources including personal digital assistant (PDA) formularies and online drug formularies are available to help the physician check medication interactions. Using electronic medical records with integral prescription-writing software makes it easier to document a patient’s adverse medication history and can alert the physician about potential medication interactions for future prescriptions. Physician order entry can help reduce medication conflicts for our patients. The most immediate solutions in avoiding medication interactions are to avoid prescribing redundant medications and similar classes of medications, to advise patients against taking competing over-the-counter medications, to control patients’ chronic medical conditions, to counsel against alcoholism, and to assist them with smoking cessation.

Back to Top


  • The risk of a drug interaction is more likely the more medications a patient takes.
  • Warfarin, angiotensin-converting enzyme (ACE) inhibitors, digoxin, and theophylline are often associated with the most dangerous drug interactions.
  • So many drug interactions are possible with warfarin that prescribing any medication to the patient taking warfarin should give the physician pause to consider the potential for drug toxicity or to consult a prescribing formulary.
  • Consider lifestyle choices when prescribing medications that require therapeutic serum concentrations and are metabolized by cytochrome pathways affected by cigarette smoking and diet, for example.
  • Rifampin, phenytoin, carbamazepine, phenobarbital, and tobacco are potent inducers of the cytochrome P-450 system.
  • The risk of myopathy attributed to statins is increased when they are taken in combination with cyclosporine, gemfibrozil, niacin, itraconazole, or erythromycin.
  • Life-threatening hyperkalemia is possible in patients taking amiloride, ACE inhibitors, beta blockers, cyclosporine, digoxin, heparin, nonsteroidal anti-inflammatory drugs, intravenous penicillin G potassium, pentamidine, potassium-sparing diuretics, spironolactone, succinylcholine, triamterene, or trimethoprim.
  • Many commonly prescribed medications from various classes can prolong the QT interval.
  • Children are often underdosed or overdosed with medications, and the elderly can experience untoward effects from medications that younger adults might not.

Back to Top

Suggested Readings

  • Agency for Healthcare Research and Quality. National Healthcare Quality Report 2005. PDF available at (accessed January 23, 2008)
  • Fick DM, Cooper JW, Wade WE, et al: Updating the Beers criteria for potentially inappropriate medication use in older adults: Results of a US consensus panel of experts. Arch Intern Med. 2003, 163: (22): 2716-2724.
  • Gaeta TJ, Fiorini M, Ender K, et al: Potential drug-drug interactions in elderly patients presenting with syncope. J Emerg Med. 2002, 22: (2): 159-162.
  • Greenblatt DJ, Patki KC, von Moltke LL, Shader RI. Drug interactions with grapefruit juice: An update. J Clin Psychopharmacol. 2001, 21: (4): 357-359.
  • Gurwitz JH, Field TS, Avorn J, et al: Incidence and preventability of adverse drug events in nursing homes. Am J Med. 2000, 109: (2): 87-94.
  • Kaufman DW, Kelly FP, Rosenberg L, et al: Recent patterns of medication use in the ambulatory adult population of the United States: The Slone survey. JAMA. 2002, 287: (3): 337-344.
  • McPhillips HA, Stille CJ, Smith D, et al: Potential medication dosing errors in outpatient pediatrics. J Pediatr. 2005, 147: 761-767.
  • Michalets EL. Update: Clinically significant cytochrome P-450 drug interactions. Pharmacotherapy. 1998, 18: (1): 84-112.
  • Prybys KM. Deadly drug interactions in emergency medicine. Emerg Med Clin North Am. 2004, 22: 845-863.
  • Wolfe MM, Lichtenstein DR, Singh G. Gastrointestinal toxicity of anti-inflammatory drugs. N Engl J Med. 1999, 340: 1888-1899.


  1. Preskorn SH. Drug approvals and withdrawals over the last 60 years. J Psychiatr Pract. 2002, 8: (1): 41-50.
  2. Prybys KM. Deadly drug interactions in emergency medicine. Emerg Med Clin North Am. 2004, 22: 845-863.
  3. Gaeta TJ, Fiorini M, Ender K, et al: Potential drug-drug interactions in elderly patients presenting with syncope. J Emerg Med. 2002, 22: (2): 159-162.
  4. Kaufman DW, Kelly FP, Rosenberg L, et al: Recent patterns of medication use in the ambulatory adult population of the United States: The Slone survey. JAMA. 2002, 287: (3): 337-344.
  5. Gauthier I, Malone M. Drug-food interactions in hospitalized patients. Methods of prevention. Drug Saf. 1998, 18: (6): 383-393.
  6. Gurwitz JH, Field TS, Avorn J, et al: Incidence and preventability of adverse drug events in nursing homes. Am J Med. 2000, 109: (2): 87-94.
  7. Michalets EL. Update: Clinically significant cytochrome P-450 drug interactions. Pharmacotherapy. 1998, 18: (1): 84-112.
  8. Nemeroff CB, DeVane CL, Pollock BG. Newer antidepressants and the cytochrome P450 system. Am J Psychiatry. 1996, 153: 311-320.
  9. Byrd DC, Gaskins SE, Parrish AM, Freeman LB. Warfarin and ciprofloxin interaction: Case report and controversy. J Am Board Fam Pract. 1999, 12: 486-488.
  10. Stading JA, Skrabal MZ, Faulkner MA. Seven cases of interaction between warfarin and cyclooxygenase-2 inhibitors. Am J Health Syst Pharm. 2001, 58: 2076-2080.
  11. Shaefer MA, Plowman BK, Morreale AP, Egan M. Interaction of rofecoxib and celecoxib with warfarin. Am J Health Syst Pharm. 2003, 60: 1319-1323.
  12. Shinn AF. Clinical relevance of cimetidine drug interactions. Drug Saf. 1992, 7: 245-267.
  13. Hirri HM, Green PJ. Audit of anticoagulant therapy and acute hospital admissions. Clin Lab Haematol. 2002, 24: 43-45.
  14. Maurer PM, Bartkowski RR. Drug interactions of clinical significance with opioid analgesics. Drug Saf. 1993, 8: 30-48.
  15. Piscitelli SC, Flexner C, Minor JR, et al: Drug interactions in patients infected with human immunodeficiency virus. Clin Infect Dis. 1996, 23: 685-693.
  16. Wolfe MM, Lichtenstein DR, Singh G. Gastrointestinal toxicity of anti-inflammatory drugs. N Engl J Med. 1999, 340: 1888-1899.
  17. McPhillips HA, Stille CJ, Smith D, et al: Potential medication dosing errors in outpatient pediatrics. J Pediatr. 2005, 147: 761-767.
  18. Bernhardt DT. Topical diphenhydramine toxicity. Wis Med J. 1991, 90: (8): 469-471.
  19. Chan CY, Wallander KA. Diphenhydramine toxicity in three children with varicella-zoster infection. DICP. 1991, 25: (2): 130-132.
  20. Woodward GA, Baldassano RN. Topical diphenhydramine toxicity in a five year old with varicella. Pediatr Emerg Care. 1988, 4: (1): 18-20.
  21. Fick DM, Cooper JW, Wade WE, et al: Updating the Beers criteria for potentially inappropriate medication use in older adults: Results of a US consensus panel of experts. Arch Intern Med. 2003, 163: (22): 2716-2724.
  22. Agency for Healthcare Research and Quality. National Healthcare Quality Report 2005. PDF available at (accessed March 20, 2009)
  23. Greenblatt DJ, Patki KC, von Moltke LL, Shader RI. Drug interactions with grapefruit juice: An update. J Clin Psychopharmacol. 2001, 21: (4): 357-359.