Drug interaction management PDF

Preview

Summary

Drug interaction management Commentary • Philip D. Hansten the HMG-CoA reductase inhibitors (‘statins’). Statins Pharm World Sci 2003; 25(3): 94–97. © 2003 Kluwer Academic Publishers. Printed in the Netherlands. are commonly used, and have interactive properties that provide good examples to demonst...

Description

Accelerat ing t he world's research.

Drug interaction management Philip Hansten International Journal of Clinical Pharmacy

Cite this paper

Downloaded from Academia.edu 

Get the citation in MLA, APA, or Chicago styles

Related papers

Download a PDF Pack of t he best relat ed papers 

DDIs bet ween dihydropyridine calcium channel blockers and st at ins Dove Medical Press An Overview of Psychot ropic Drug-Drug Int eract ions Kelly Cozza Xenosensors CAR and PXR at Work: Impact on St at in Met abolism Maria Marino

Commentary

Drug interaction management • Philip D. Hansten

Pharm World Sci 2003; 25(3): 94–97. © 2003 Kluwer Academic Publishers. Printed in the Netherlands. P.D. Hansten: University of Washington, 1959 NE Pacific Street, Seattle, WA 98196-7630, USA (e-mail: [email protected]) Key words Computerized screening Drug interactions Drug interaction management Pharmacogenomics Prescribing errors Risk assessment Abstract Although thousands of articles on drug interactions have been published and numerous computerized screening systems have been developed, patients continue to suffer from adverse drug interactions. Possible methods for reducing the risk of drug interactions include improving the knowledge of health care providers, improving computerized screening systems, providing information on patient risk factors, increased use of pharmacogenetic information, more attention to drug administration risk factors, and improving patient education on drug interactions. Accepted March 2003

Upon this gifted age, In its dark hour Rains from the sky A meteoric shower of facts. Wisdom enough to leach us of our ill Is daily spun. But there exists no loom To weave it into fabric. – Edna St. Vincent Millay

Introduction

94

Most health care providers feel overwhelmed by the amount of information published on drug–drug interactions. Dealing with the massive amount of data on drug interactions has been aptly likened to ‘drinking from a fire hose.’ There is simply too much information for any one individual to absorb and digest. And, as the above poem by Millay makes clear, the spinners (who generate the information) vastly outnumber the weavers (who put the information into a usable form), leaving many people awash in a sea of data. There are many sources of drug–drug interaction information available to health care providers. Some of these sources provide extensive information on pharmacokinetic studies of the interaction, discussions of case reports and the like. However simply knowing that two drugs may interact does not provide enough information for the health care provider to devise a plan to reduce the risk of an adverse outcome. Health care providers are often left on their own to come up with a plan to manage the interaction. In this article I will make some recommendations for improvement in the current system of providing drug interaction information to health care providers. To illustrate the points, I will use examples involving

the HMG-CoA reductase inhibitors (‘statins’). Statins are commonly used, and have interactive properties that provide good examples to demonstrate the principles discussed.

Improving the drug interaction knowledge of health care providers In the past 40 years, more than 15,000 articles on drug interactions have been published in the medical and pharmaceutical literature. Since the discovery of the cytochrome P450 isozymes and the ATP-binding cassette (ABC) transporters, the number of drug interaction publications per year has increased substantially. Clearly, it is not possible for individual health care providers to remember all drug interactions. They do, however, need to understand how to use the drug interaction information that is provided by computers, books and other sources. The overwhelming nature of the information notwithstanding, it is both possible and necessary for health care providers to understand some of the general principles and concepts of drug interactions. For example, understanding the mechanisms by which drugs interact with each other can be very useful in the clinical assessment and management of drug interactions. The mechanism of an interaction can be important in predicting the time course of the interaction, and may also provide information leading to a way to minimize the risk of an adverse outcome. It is also possible for the health care providers to remember the interactive properties of drugs, at least the drugs that he or she uses frequently. If one knows that a certain drug is an enzyme inducer, or inhibits a particular cytochrome P450 isozyme, it is possible to predict how this drug is likely to interact with other medications the patient takes.1–3

Improve computerized drug interaction screening systems It is clear that computerized drug interaction screening systems have not been as successful as one would hope. Currently available computerized systems in the United States have been shown to have significant deficiencies and there is insufficient information to determine whether computerized drug interaction screening is actually helpful in reducing drug interaction morbidity and mortality in the community pharmacy setting3,4. Numerous deficiencies and limitations of computerized drug interaction screening systems have been identified, and some of these are discussed below. Excessive number of drug interactions on the systems Many pharmacists find that computerized drug interaction screening systems detect a large number of drug–drug interactions of questionable clinical significance. Indeed, informal polls of pharmacists by the author (P.D.H.) around the United States suggest that

pharmacists ignore most drug interaction alerts provided in ambulatory patient settings. To test a popular Internet drug interaction-checking program the author entered the medications for one ambulatory patient into the system. This Internet program detected 69 drug–drug interactions for the patient. Clearly, those who create the databases for these programs need to be more selective in choosing the drug interactions to be included.

that most patients who take interacting drug combinations do not manifest adverse consequences9. For example, verapamil, a CYP3A4 inhibitor, has been associated with a 10-fold increase in the risk of simvastatin myopathy16. But the overall risk of myopathy in patients receiving simvastatin plus verapamil is reportedly less than 1%. So the relative risk of simvastatin myopathy is substantially increased by verapamil, but the absolute risk of using the combination is very small. Substantial evidence suggests that the risk of statin-induced myopathy increases with increasing serum concentrations of the statin. Accordingly, it has been recommended that simvastatin dose not exceed 20 mg daily in patients receiving verapamil concurrently10. But the fact that such a small percentage of patients on verapamil and simvastatin manifest clinical evidence of myopathy suggests that other – largely unknown – risk factors also play a role in determining which patients develop the muscle toxicity. Several factors reportedly increase the risk of statininduced myopathy (e.g., advanced age, low body weight, renal insufficiency) but the exact role such risk factors play in any given patient is usually unclear6.

Drug class differences not handled correctly Almost all drug classes interact heterogeneously, because individual members of a drug class are often not metabolized by the same cytochrome P450 isozymes or ABC transporters as other members of the class. The statins are a good example, because simvastatin and lovastatin are extensively metabolized by CYP3A4, atorvastatin is moderately metabolized by CYP3A4, fluvastatin is metabolized by CYP2C9, and pravastatin and rosuvastatin are not metabolized by cytochrome P450 isozymes5. Thus, combining all members of this drug class together is rarely justified when considering drug interactions. Nonetheless, it is common for reviews and computer systems to include all statins together as interacting with CYP3A4 inhibitors, even though the risk is primarily limited to lovastatin, simvastatin, and – to a Incorporate pharmacogenetic information lesser extent – atorvastatin6.

into risk assessment Management guidelines are often inadequate It is not enough for the health care provider to be informed that two drugs may interact. It is also important to have information on measures that can be taken to reduce the likelihood of an adverse outcome7,8. Management options include: • Avoiding the combination entirely. For some drug interactions, the risk always outweighs the risk, and the combination should be avoided. Because drug classes are usually heterogeneous with regard to drug interactions (as described above) one can often select a non-interacting alternative for either the object drug or the precipitant drug. • Adjusting the dose of the object drug. Sometimes, it is possible to give the two interacting drugs safely as long as the dose of the object drug is adjusted. • Spacing dosing times to avoid the interaction. For some drug interactions involving binding in the gastrointestinal tract, to avoid the interaction one can give the object drug at least two hours before or four hours after the precipitant drug. In this way the object drug can be absorbed into the circulation before the precipitant drug appears. • Monitoring for early detection. In some cases, when it is necessary to administer interacting drug combinations, the interaction can be managed through close laboratory or clinical monitoring for evidence of the interaction. In this way the appropriate dosage changes can be made, or the drugs discontinued if necessary.

Provide information on patient risk factors that increase the chance of an adverse outcome It is clear from both the clinical experience of physicians and pharmacists as well as published studies

The magnitude of drug interactions is known to differ markedly from person to person, and some of this variability is due to pharmacogenetic influences. The cytochrome P450 isozymes are known to be under genetic control, and a dozens of genetic variants have been identified, particularly for CYP2D6 and CYP2C191. For example, patients who are deficient in CYP2D6 may have an increased risk of intolerance to simvastatin compared with patients with normal CYP2D6 activity11. For example, suppose that in people with ‘normal’ cytochrome P450 isozymes, a particular drug (drug A) is metabolized primarily by CYP2C19, with CYP3A4 as minor alternative pathway. In such patients, a potent CYP3A4 inhibitor such as ketoconazole would tend to have a relatively small effect on the pharmacokinetics of drug A. On the other hand, if a CYP2C19 deficient patient takes the drug, the alternative CYP3A4 pathway may become the predominant route of drug A metabolism, and the addition of a CYP3A4 inhibitor would substantially increase its serum concentrations. Under other circumstances a patient who is deficient in a cytochrome P450 isozyme may be less likely to manifest a drug interaction. For example, a CYP2D6 deficient patient who is taking a drug metabolized by CYP2D6 (drug B) may have an adequate therapeutic response with a very low dose of the drug. Then if a potent CYP2D6 inhibitor is added to the patient’s regimen, there will be no interaction with drug B because the patient does not have any CYP2D6. On the other hand, in a patient with a normal amount of CYP2D6 who is receiving a typical dose of drug B, the addition of a potent CYP2D6 inhibitor is likely to substantially increase the serum concentrations of drug B, possibly leading to toxicity. Deficiency of cytochrome P450 isozymes is not the only genetic variation observed. Some people have

95

multiplication of genes for the CYP2D6 isozyme, and have much higher CYP2D6 activity than normal. Such patients may not respond well to drugs that are metabolized by CYP2D6. Simvastatin appears to have a CYP2D6 pathway, and a patient with high CYP2D6 activity failed to achieve any cholesterol reduction when given simvastatin11. But supernormal CYP2D6 activity may also affect the outcome of drug–drug interactions. For example, if such individuals take a precipitant drug that is metabolized by CYP2D6, one would not expect them to achieve sufficient serum concentration of this drug to interact with other drugs in a clinically important way. It is possible to determine a person’s genotype or phenotype for many of the cytochrome P450 isozymes, but this is used primarily in research rather than as a clinical tool for predicting drug response. As these procedures become more automated and less expensive, however, it is likely that they will become more widely used for clinical management, at least for selected patients.

the outcome of the interaction. This is particularly true when both the object and precipitant drug are given chronically. But dosing times can be important for interactions where one drug binds with another in the gastrointestinal tract. In this case, giving the object drug two hours before or four to six hours after the precipitant drug can usually minimize the magnitude of the interaction.

Provide information on drug administration risk factors that increase the chance of an adverse outcome

Improve patient education on drug interactions

The magnitude of most drug interactions is dependent upon the way the drugs are administered. Of all the factors that affect the outcome of drug interactions, we have more useful information in this area than any other. Yet this valuable information is often not made available to the health care providers who are dealing with the drug interaction in a specific patient. Dose Almost all drug interactions are dose related, with larger doses of the object drug and/or precipitant drug increasing the magnitude of the interaction. For example, the ability of grapefruit juice to inhibit the intestinal presystemic metabolism of lovastatin or simvastatin depends on the amount of grapefruit juice ingested. A single daily glass of regular strength grapefruit juice for three days produces a modest increase in lovastatin serum concentrations, while a glass of double strength grapefruit juice three times daily produces a dramatic increase in lovastatin serum concentrations12,13. Duration The duration of therapy with the precipitant drug can be an important determinant to the outcome of a drug interaction. A single dose of a CYP3A4 inhibitor – no matter how potent – is unlikely to produce a clinically important effect on a drug that is metabolized by CYP3A4. For example, a patient on simvastatin is likely to manifest only a transient increase in simvastatin serum concentrations if itraconazole is given for only a day or two. And given that it usually takes at least two to three weeks of administration of a CYP3A4 inhibitor with simvastatin before symptoms of myopathy develop, the risk of short-term itraconazole is probably quite small.

96

Sequence The sequence of administration of the object and precipitant drug can be important when the object drug is titrated to an optimal therapeutic response (e.g., oral anticoagulants). In that case, the primary danger is when the precipitant drug is added in a patient stabilized on the object drug, particularly in ambulatory patients where the object drug response may not be monitored frequently. On the other hand, when the object drug is initiated and titrated in the presence of a stable dose of the precipitant drug, the risk is usually lower.

Patient education can play an important role in reducing the risk of some drug–drug interactions. This can take the form of verbal instructions from the health care provider, patient instruction leaflets given with the prescription, auxiliary warning labels placed on the medication package, books, and, increasingly, the Internet. For example, all patients receiving HMG-CoA reductase inhibitors should be warned about the symptoms of myopathy (muscle pain, muscle weakness, and dark urine). Patients so warned would be expected to seek medical attention more rapidly if they develop myopathy as a result of a statin drug interaction. Early detection and discontinuation of the statin appears to be important for minimizing the severity of statin-induced myopathy6. Patient education is also important when a prescription medication have potential adverse interactions with herbal medications or OTC medications. For example, St. John’s Wort has been shown to substantially reduce serum concentrations of simvastatin, probably as a result of induction of CYP3A4 by St. John’s Wort14. Lovastatin, and to a lesser extent atorvastatin, are also metabolized by CYP3A4 and would also be expected to interact with St. John’s Wort. Thus, when patients who receive prescriptions for one of these statins should be warned to avoid St. John’s Wort.

Conclusion

The past 35 years have seen major advances in our understanding of drug–drug interactions, particularly in the area of the molecular mechanisms by which drug interact. But our ability to appropriately apply this information to specific patients has lagged far behind. As Edna St. Vincent Millay said, we need more weavers – people who take the data published in the medical and pharmaceutical literature and Dosing times For most metabolic drug interactions involving the show how this information can be applied so that the cytochrome P450 system, the specific dosing times of risk of adverse drug interactions can be minimized. the object and precipitant drugs have little effect on

References 1 Levy RH, Thummel KE, Trager WF, Hansten PD, Eichelbaum M, editors. Metabolic drug interactions. Philadelphia: Lippincott Williams & Wilkins, 2000. 2 Hansten PD, Horn JR. Drug interactions analysis and management, St. Louis, Missouri, Facts & Comparisons, 2002. 3 Hazlet TK, Lee TA, Hansten PD, Horn, JR. Performance of community pharmacy drug interaction software. J Am Pharm Assoc 2001; 41: 200–4. 4 Chrischilles EA, Fulda TR, Byrns PJ, Winckler SC, Rupp MT, Chui MA. The role of pharmacy computer systems in preventing medication errors. J Am Pharm Assoc 2002; 42: 439–48. 5 Williams D, Feely J. Pharmacokinetic-pharmacodynamic drug interactions with HMG-CoA reductase inhibitors. Clin Pharmacokinet 2002; 41: 343–70. 6 Pasternak RC, Smith SC, Bairey-Merz CN, Grundy SM, Cleeman JI, Lenfant C. ACC/AHA/NHLBI Clinical advisory on the use and safety of statins. J Am Coll Cardiol 2002; 40: 568–73. 7 Hansten PD, Horn JR, Hazlet TK. ORCA: OpeRational ClassificAtion of drug interactions. J Am Pharm Assoc 2001; 41: 161–165. 8 Peterson JF, Bates DW. Preventable medication errors: identi-

9

10 11

12

13 14

fying and eliminating serious drug interactions. J Am Pharm Assoc 2001; 41: 159–60. Doucet J, Chassange P, Trivalle C, Landrin I, Pauty MD, Kadri N et al. Drug–drug interactions related to hospital admissions in older adults: a prospective study of 1000 patients. J Am Geriatr Soc 1996; 44: 944–8. Orloff DG. Label changes for Simvastatin (Zocor). US Food and Drug Administration, 2002. Mulder AB, van Lijf HJ, Bon MA, van den Bergh FA, Touw DJ, Neef C, Vermes I. Association of polymorphism in the cytochrome CYP2D6 and the efficacy and tolerability of simvastatin. Clin Pharmacol Ther 2001; 70: 546–51. Rogers JD, Zhao J, Liu L, Amin RD, Gagliano KD, Porras AG et al. Grapefruit juice has minimal effects on plasma concentrations of lovastatin-derived 3-hydroxy-3-methylglutaril coenzyme A reductase inhibitors. Clin Pharmacol Ther 1999; 66: 358–66. Kantola T, Kivisto KT, Neuvonen PJ. Grapefruit juice greatly increases serum concentrations of lovastatin and lovastatin acid. Clin Pharmacol Ther 1998; 63:3 97–402. Sugimoto K, Ohmori M, Tsuroka S, Nishiki K, Kawaguchi A, Harada K et al. Different effects of St. John’s Wort on the pharmacokinetics of simvastatin and pravastatin. Clin Pharmacol Ther 2001; 70: 518–24.

97...