A

CON
T

IO N

N
U

IN G

ED

ARTICLE 4

ADVERSE DRUG INTERACTIONS IN DENTAL PRACTICE:

INTERACTIONS INVOLVING ANTIBIOTICS
PART II OF A SERIES
ELLIOT V. HERSH, D.M.D., M.S., PH.D.

Background. The prudent use of antibiotics is an integral part of dental practice. While these agents generally are considered safe in the dental setting, their use can result in interactions that can lead to serious morbidity in dental patients. Methods. The faculty of a symposium entitled Adverse Drug Interactions in Dentistry: Separating the Myths From the Facts did an extensive literature review on drug interactions. Through this, they were able to establish a significance rating of alleged adverse drug interactions as they relate to dentistry, based on their scientific documentation and severity of effect. The author of this article focused on antibiotics. Results. Most of the reported drug interactions discussed in this article are well-documented by clinical studies. It is particularly important that dentists be aware of the potentially serious and life-threatening interactions of the antibiotics erythromycin, clar-

ithromycin and metronidazole, and of the antifungal agents ketoconazole and itraconazole, with a host of other drugs whose metabolism is impaired by these antimicrobial agents. In contrast, the alleged ability of commonly employed antibiotics to reduce the effectiveness of oral contraceptive agents is not adequately supported by clinical research. It still is recommended, however, that clinicians discuss this possible interaction with their patients, as it might represent a relatively rare event that cannot be discerned in clinical trials. Conclusions. Potentially serious adverse drug interactions can occur between antimicrobial agents used in dental practice and other drugs patients are taking for a variety of medical conditions. Clinical Implications. It is important that dentists stay abreast of potential drug interactions involving antibiotics to avoid serious morbidity among their patients.

This article, focusing on antibiotics, is the sec-

ond in a five-part series discussing the clinical significance of reported drug interactions as they relate to dentistry. While there are numerous antibiotic preparations available for the treatment of localized and systemic infections, relatively few antibiotic preparations are routinely employed in dentistry (Box, Antibiotics Commonly Used in Dentistry). This factor alone limits the quantity of adverse drug interactions seen with antibiotics in our profession. However, antibiotics are prescribed for longer durations than are other agents routinely admin-

istered in dentistry. For example, local anesthetics and sedatives usually are given in a single course of therapy,1 whereas in most patients analgesics are taken for a few days on an as-needed basis. The typical antibiotic regimen for an odontogenic infection is of a five- to 10-day duration on an around-the-clock schedule. This more chronic dosing sets the stage for some rather serious and potentially life-threatening drug interactions involving antimicrobial agents (such as erythromycin, clarithromycin, ketoconazole and itraconazole), which inhibit the gut wall and liver cytochrome P-450 system with a host of other

236

JADA, Vol. 130, February 1999 Copyright 1998-2001 American Dental Association. All rights reserved.

PHARMACOLOGY
drugs that use the same metabolic pathway. The ability of certain antimicrobial agents to increase blood levels of other drugs with low therapeutic indexes is, in fact, the cause of the most potentially hazardous interactions discussed in this article. The purpose of this article and the others in this series is not simply to list and discuss the theoretical basis of published adverse drug interactions. Our major goal is to explore the scientific documentation and the seriousness of each interaction as it relates to the practice of dentistry. Many of the interactions that will be presented in this article are supported by an abundance of scientific data. For example, there are numerous well-designed pharmacokinetic studies that have demonstrated clearly that the simultaneous ingestion of agents containing divalent or trivalent cations and tetracycline antibiotics greatly impair the absorption of this class of antibiotic agents. On the other hand, although case reports have implicated the administration of penicillins, tetracyclines and other antibiotics used in dentistry with reduced systemic absorption and ultimate failure of oral contraceptive agents, well-controlled clinical trials have failed to demonstrate this interaction.
THE SIGNIFICANCE RATING SCALE

ANTIBIOTICS COMMONLY USED IN DENTISTRY.

BACTERICIDAL Generic Name Trade Name* BACTERIOSTATIC Generic Name Trade Name

Penicillin V Amoxicillin Cephalexin Cefadroxil Metronidazole

Pen-Vee K Amoxil Keflex Duricef Flagyl

Erythromycin Clarithromycin Azithromycin Clindamycin Tetracycline Doxycycline

Eryc Biaxin Zithromax Cleocin Achromycin Vibramycin

* The trade-named bactericidal antibiotics are listed only as examples. Manufacturers are as follows: Pen-Vee K, Wyeth-Ayerst Laboratories; Amoxil, SmithKline Beecham Consumer Health Care; Keflex, Dista Products and Eli Lilly and Company; Duricef, Bristol-Myers Squibb Company; Flagyl, G.D. Searle and Company. The trade-named bacteriostatic antibiotics are listed only as examples. Manufacturers are as follows: Eryc, Warner Chilcott Laboratories; Biaxin, Abbott Laboratories; Zithromax, Pfizer Incorporated Consumer Health Care Group; Cleocin, Pharmacia and Upjohn Company; Achromycin, Lederle Laboratories; Vibramycin, Pfizer Incorporated Consumer Health Care Group.

the practicing dentist, those given a rating of 4 should not be dismissed completely. A rating of 4 typically implies that a few published reports have suggested an interaction that is of major or moderate severity, but that scientific documentation either is lacking or does not support an interaction. Further research is needed to either further establish or refute the interaction. Table 2 summarizes the specific interactions that are associated with antibiotic therapy in dentistry and provides the significance rating for each.
BACTERICIDAL ANTIBIOTICS WITH BACTERIOSTATIC ANTIBIOTICS

The significance rating scale that is assigned to each drug interaction within this article was presented in detail in the first article of this series1 and is summarized in Table 1. Although adverse drug interactions assigned a rating of 1 or 2 clearly should be of concern to

Most teachers of pharmacology to predoctoral and postgraduate dental students preach the central dogma that bactericidal antibiotics should never be combined with bacteriostatic

antibiotics (Box, Antibiotics Commonly Used in Dentistry), because the latter will antagonize the action of the former. In reality, there are relatively few well-authenticated clinical examples of this phenomenon.2 In patients with pneumococcal meningitis, penicillin produced significantly higher recovery rates and fewer mortalities when used alone than when combined with a tetracycline to treat the same infection.3,4 A higher rate of spontaneous reinfection also has been reported in patients with scarlet fever who were treated with a penicillin-tetracycline combination than in those receiving only penicillin.2 In other studies involving scarlet fever, combined penicillin and erythromycin therapy was slightly less effective than monotherapy with penicillin,5 and antagonism between these two antibiotics has been demon237

JADA, Vol. 130, February 1999 Copyright 1998-2001 American Dental Association. All rights reserved.

PHARMACOLOGY
TABLE 1

THE DRUG INTERACTION SIGNIFICANCE RATING SCALE.

SIGNIFICANCE RATING 1 2 3 4 5 SEVERITY RATING DOCUMENTATION RATING Established, probable or suspected Established, probable or suspected Established, probable or suspected Possible Possible Unlikely Major Moderate Minor Major or moderate Minor All

strated in vitro.6 There are several instances, however, in which bacteriostatic antibiotics have been successfully combined with bactericidal antibiotics to prevent the emergence of resistant strains. Take, for example, the triple antibiotic combination of bismuth subsalicylate (Pepto-Bismol, Procter & Gamble), tetracycline (a static antibiotic) and metronidazole (a cidal antibiotic), which has been widely employed for the eradication of Helicobacter pylori in patients with gastrointestinal ulcers.7 In dental medicine, there is no rationale for combining bacteriostatic antibiotics with bactericidal antibiotics to treat odontogenic infections. These combinations are likely to lead to greater toxicity than that of single-drug therapy and, at least with penicillins, a diminution of antibiotic effectiveness.
TETRACYCLINES WITH PRODUCTS CONTAINING DIVALENT AND TRIVALENT CATIONS

The ability of divalent and trivalent cations such as calcium++, magnesium++, bismuth++, iron++, zinc++ and aluminum++ (found in dairy products, antacids and vitamin
238

preparations) to markedly impair the absorption of tetracycline molecules from the gastrointestinal tract is probably the adverse drug interaction most widely known among dental clinicians. These interactions are well-documented and established through numerous case reports and well-controlled clinical studies. 8-15 While the individual tetracycline moieties differ with respect to certain cations (for example, doxycycline and minocycline are not affected as much as other tetracycline moieties with respect to calcium++ and zinc++10,11,15), reductions in serum tetracycline concentrations of 20 to 100 percent in the presence of these cations often are so large that antibiotic levels can fall below those needed to inhibit bacterial growth).2 The significance rating of 2 given to this interaction reflects a likely increase in patient morbidity. The simultaneous ingestion of tetracycline molecules and multicationic products should be avoided absolutely.
METRONIDAZOLE WITH ALCOHOL, LITHIUM

Metronidazole with alcohol. Metronidazole, like disulfiram

(Antabuse, Wyeth-Ayerst Laboratories), has been shown in the laboratory to inhibit the activity of acetaldehyde dehydrogenase, theoretically causing an accumulation of acetaldehyde in patients who ingest alcohol concomitantly.16 The clinical ability of metronidazole to produce disulfiramlike reactions in alcohol consumers (Table 2) is supported by several studies that have shown an incidence ranging from 2 percent to 100 percent of the study population.2,17-19 While the reaction normally is more unpleasant and frightening than it is serious, patients should be advised not to consume alcohol during metronidazole therapy and for at least three days afterward.20 Metronidazole with lithium. Lithium, a monovalent cation, is indicated for the treatment of bipolar (manic-depressive) disorder, most frequently during the manic phase. It is given under close supervision with regular monitoring of blood concentrations because of its low therapeutic index. Therapeutic blood levels are in the range of 0.8 to 1.5 milliequivalent per liter during an acute manic attack and 0.6 to 1.2 mEq/L for maintenance therapy.2,21 Early signs of lithium intoxication can occur at blood concentrations that are just above therapeutic (in the 1.5-2 mEq/L range) and include lethargy, muscle weakness and fine hand tremors. More serious toxicity consisting of confusion, nystagmus and ataxia typically are seen at lithium blood concentrations of 2.0 to 2.5 mEq/L. Life-threatening toxicity (consisting of seizures, coma and circulatory collapse) can occur with lithium concentrations

JADA, Vol. 130, February 1999 Copyright 1998-2001 American Dental Association. All rights reserved.

PHARMACOLOGY

TABLE 2

ADVERSE DRUG INTERACTIONS IN DENTISTRY: ANTIBIOTICS.

POSSIBLE DRUG INTERACTION Bactericidal antibiotics with bacteriostatic antibiotics SIGNIFICANCE RATING 2 MECHANISM AND CLINICAL PRESENTATION Theoretically, bactericidal drugs work best when microbes are actively growing. By inhibiting cell growth, bacteriostatic agents may antagonize the bactericidal agent. This interaction is not consistently demonstrated clinically. Tetracycline molecules chelate divalent and trivalent cations, impairing antibiotic absorption. In addition, antacids can raise the gastrointestinal pH, also impairing absorption. Serum tetracycline levels can be reduced 20 to 100 percent, leading to poor antibiotic efficacy. Metronidazole produces a disulfiram effect by inhibiting the enzyme acetaldehyde dehydrogenase. Acetaldehyde accumulation can lead to facial flushing, headache, palpitation and nausea. Metronidazole inhibits the renal excretion of lithium, leading to elevated lithium blood concentrations. Lithium toxicityas manifested by confusion, ataxia and kidney damagecan result. Metronidazole inhibits the renal excretion of lithium, leading to elevated lithium blood concentrations. Lithium toxicityas manifested by confusion, ataxia and kidney damagecan result. A single case report describes elevated lithium blood concentrations and lithium toxicity with concomitant tetracycline administration.25 Another report26 and a clinical research study27 do not support the interaction. These antibiotics can reduce the gut flora, in particular Eubacterium lentum, which metabolize a significant amount of oral digoxin in 10 percent of patients taking the drug. Elevated concentrations of digoxin can occur in such patients, leading to digitalis toxicity, as manifested by salivation, visual disturbances and arrhythmias. Broad-spectrum antibiotics are hypothesized to reduce the gut flora that normally synthesize vitamin K, a necessary cofactor for clotting. Since warfarins and anisindiones anticoagulant mechanism involves an antagonism of vitamin K-dependent clotting factors, an increased risk of bleeding may occur. Interaction appears significant only in patients with poor vitamin K intake. These antibiotics decrease the metabolism of warfarin and anisindione and can significantly increase prothrombin times and increase the risk of serious bleeding in anticoagulated patients. Signs of this adverse interaction include hematuria, bruising and hematoma formation.

Tetracyclines with products that contain divalent or trivalent cations

Metronidazole with alcohol

Metronidazole with lithium

Tetracyclines with lithium 4

Erythromycin or tetracyclines with digoxin

Tetracyclines or other broad-spectrum antibiotics with warfarin or anisindione

Erythromycin, clarithromycin or metronidazole with warfarin or anisindione

above 2.5 mEq/L.2 In addition, elevated blood lithium concentrations can lead to renal dysfunction, often in the form of nephrogenic diabetes insipidus,

resulting in excessive excretion of dilute urine.2,21 Three cases have been reported in the literature in which the administration of metro-

continued on page 240 nidazole for one week (5001,000 milligrams daily) apparently induced elevations in lithium blood concentrations with concomitant toxicity.22,23 In
239

JADA, Vol. 130, February 1999 Copyright 1998-2001 American Dental Association. All rights reserved.

PHARMACOLOGY
TABLE 2 CONTINUED

ADVERSE DRUG INTERACTIONS IN DENTISTRY: ANTIBIOTICS.

POSSIBLE DRUG INTERACTION Erythromycin, clarithromycin, ketoconazole or itraconazole with a host of other drugs that are metabolized by the CYP3A4 and CYP1A2 system in the gut wall and liver Astemizole, terfenadine or cisapride SIGNIFICANCE RATING Depends on interacting drug (see below) MECHANISM AND CLINICAL PRESENTATION These antimicrobial agents block the metabolism and thus increase blood levels of a number of drugs. Severity of the reaction depends on the therapeutic index of the interacting drug.

Life-threatening ventricular arrhythmias (torsades de pointes). Metronidazole also increases cisapride blood levels. Enhanced and/or prolonged respiratory depression. Ketoconazole not implicated. Increased risk of adverse central nervous system effects, dyskinesias and hypotension. Increased risk of ataxia, vertigo, drowsiness and confusion. Cardiac arrest reported in one child taking erythromycin.68 Enhanced immunosuppression and nephrotoxicity. Increased risk of hypotension, tachycardia and edema. Increased risk of Cushings syndrome and immunosuppression. Increased risk of tachycardia, cardiac arrhythmias, tremors and seizures. Ketoconazole not implicated in this interaction. Muscle pain and rhabdomyolysis (skeletal muscle lysis). Pharmacokinetic interaction demonstrated for azole antifungal drugs. Marked increases in blood levels of both benzodiazepines when taken by mouth, leading to increases in sedative depth and duration. A few clinical reports of arrhythmias or heart block with concurrent erythromycin use. Definitive pharmacokinetic studies are needed. Sporadic case reports have implicated the concomitant ingestion of antibiotics and oral contraceptives with unwanted pregnancies. It is theorized that antibiotics, by decimating the normal gut flora, can interfere with the enterohepatic recycling of the estrogen component of oral contraceptives, thereby leading to subtherapeutic blood levels and ovulation. With the exception of the antituberculosis drug rifampin, clinical studies have failed to demonstrate an interaction.

Alfentanil

1 1 1

Bromocriptine

Carbamazepine

Cyclosporine

1 1

Felodipine and possibly other calcium-channel blockers Methylprednisolone or prednisone Theophylline

1 1

Lovastatin and possibly other -statins

Triazolam or oral midazolam

Disopyramide

Penicillins, cephalosporins, erythromycin, clarithromycin, tetracyclines, metronidazole with combined estrogen and progestin oral contraceptives

one woman, lithium blood concentrations increased from 1.3 mEq/L to 1.9 mEq/L. She developed signs (such as ataxia, mental clouding and muscle weakness) of lithium toxicity.22 In
240

another woman, lithium blood concentrations were elevated to a lesser degree (from 1.1 mEq/L to 1.3 mEq/L), but she developed persistent polyuria and nocturia, indicative of kidney

damage.23 In a third female patient, lithium blood concentrations increased dramatically from 0.9 mEq/L to 2.0 mEq/L. She was admitted to the hospital because of severe confusion

JADA, Vol. 130, February 1999 Copyright 1998-2001 American Dental Association. All rights reserved.

PHARMACOLOGY
and ataxia. She also developed hypernatremia and abnormally dilute urine, consistent with a diagnosis of nephrogenic diabetes insipidus, conditions that persisted for six months.23 While a more definitive clinical trial exploring the likelihood of the metronidazole-lithium interaction has not been performed, these three case reports are well-documented with supporting pharmacokinetic data. A significance rating of 1 is assigned to this adverse drug interaction because the documentation indicates that the interaction is at least suspected and the severity of the interaction is major. Practitioners should avoid the use of metronidazole in patients receiving lithium therapy.
TETRACYCLINES WITH LITHIUM

addition of tetracycline (1 gram per day) for one week.27 While the severity of this potential adverse drug interaction is severe, its scientific documentation puts it at most in the possible category. We therefore assign it a significance rating of 4.
ERYTHROMYCIN, CLARITHROMYCIN OR TETRACYCLINES WITH DIGOXIN

Digoxin (Lanoxin, Glaxo Wellcome Inc.) and digitoxin (Crystodigin, Eli Lilly and Company) are cardiac glycosides indicated for the treatment of congestive heart failure and certain types of atrial arrhythmias. They both have low therapeutic indexes, so that significant changes in their plasma

serum digoxin levels.28,29 A doubling of serum digoxin levels and signs of digitalis toxicity have been reported in patients taking erythromycin simultaneously for three to six days.29-31 A 30 percent rise in serum digoxin levels and a fall in digoxin metabolites have been reported with simultaneous tetracycline administration.29,32 Although these case reports do not represent well-controlled clinical studies, they are, in fact, supported by some rather convincing pharmacokinetic data. A significance rating of 1 is thus assigned to this adverse drug interaction because the documentation indicates that the interaction is at least suspected and the clinical outcome of digitalis toxicity is quite serious and potentially life-threatening.
TETRACYCLINES OR OTHER BROAD-SPECTRUM ANTIBIOTICS WITH ORAL ANTICOAGULANTS

A warning to completely avoid the concomitant administration of tetracycline with lithium also exists in the literature.24 It stems from a single case report of a 30-year-old woman whose lithium blood levels had ranged from 0.50 mEq/L to 0.84 mEq/L over a three-year period.25 After a one-week course of tetracycline hydrochloride (250 mg three times per day), her lithium blood concentration increased to 2.74 mEq/L. She also developed clinical signs of lithium toxicity: drowsiness, slurred speech, fine hand tremors and thirst.25 However, another author has reported using these drugs in combination in his patient population with no observed adverse drug interactions.26 A clinical trial in 14 healthy volunteers demonstrated that lithium blood levels actually decreased slightly (from 0.51 to 0.47 mEq/L) after the

Practitioners should avoid the use of metronidazole in patients receiving lithium therapy.
levels are likely to lead to severe reactions.1 In addition, patients who are taking these agents are more fragile than relatively healthy dental patients because of their cardiac disease. Digoxin is fairly well-absorbed (about 75 percent) after oral administration.28 However, about 10 percent of people harbor enteric bacteria that inactivate digoxin in the gut, greatly reducing the drugs bioavailability and necessitating that they receive greater than average maintenance doses (Table 2).2 In these patients, the administration of certain antibiotics can decimate the enteric bacteria, leading to a marked rise in

Probably the largest number of reported drug interactions involve the oral anticoagulants warfarin, dicumarol and anisindione. These competitive antagonists of vitamin K-dependent clotting factors have a low therapeutic index; administration of other drugs that can increase these antagonists activity can lead to serious and potentially life-threatening bleeding.2 Several antibiotics routinely used in dentistry have been implicated in producing this effect. It has been hypothesized that broad-spectrum antibiotics such as tetracyclines, amoxicillin and ampicillin can reduce endogenous vitamin K levels and enhance the effects of oral anticoagulants by decimating the normal gut flora that produce vitamin K. There have been several case
241

JADA, Vol. 130, February 1999 Copyright 1998-2001 American Dental Association. All rights reserved.

PHARMACOLOGY
reports of increased prothrombin times and bleeding in patients taking anticoagulants and a variety of tetracyclines.2,33-35 However, one study demonstrated that tetracycline had no effect on prothrombin times in a group of patients receiving chronic warfarin therapy.36 Case reports implicating amoxicillin or ampicillin with enhanced anticoagulant activity also have appeared in the literature.2,37 Interestingly, there has been at least one report demonstrating that amoxicillin slightly lowered prothrombin times in five patients.2 It appears that the ability of these antibiotics to enhance anticoagulant activity is relatively rare and unpredictable, and that it is of most concern in patients whose dietary intake of vitamin K is poor.38,39 Concurrent use of these antibiotics with oral anticoagulants (in patients with normal vitamin K intake) need not be avoided, but patients need to be monitored for signs (such as bruising, melena, bleeding) of increased anticoagulant activity.
ERYTHROMYCIN, CLARITHROMYCIN OR METRONIDAZOLE WITH ORAL ANTICOAGULANTS

most one-third and virtually doubled the anticoagulant effect of the more active (S-) isomer of warfarin.48 While the enhancement of anticoagulant activity appears in only some people, this potentially serious interaction is supported by pharmacokinetic data. A full course of therapy with these antibiotics in a patient receiving anticoagulation therapy requires consultation with the patients physician before being initiated.49 Warfarin is just one of many drugs whose metabolism can be inhibited by antimicrobial agents that block the cytochrome P-450 system.50,51
ERYTHROMYCIN, CLARITHROMYCIN, KETOCONAZOLE OR ITRACONAZOLE WITH A HOST OF OTHER DRUGS

A marked increase in the effects of warfarin with bleeding has been reported in some patients simultaneously taking a five- to eight-day course of erythromycin,2,30,35,40-45 clarithromycin2 or metronidazole.2,46 The results of a pharmacokinetic study revealed that, in some volunteers, the ingestion of erythromycin (1 g per day for eight days), resulted in a 30 percent reduction of warfarin clearance.47 Another clinical investigation demonstrated that metronidazole (750 mg daily for one week) increased the half-life of warfarin by al242

The cytochrome P-450 system consists of a large set of similar enzymes (isoenzymes) that are responsible for metabolizing a wide range of drugs.50,51 The CYP3A4 isoenzyme accounts for 60 percent of the cytochrome enzymes in the liver and 70 percent of those in the enterocytes found in the gut wall.51 Erythromycin, clarithromycin, ketoconazole (Nizoral, Janssen Pharmaceutica Inc.) and itraconazole (Sporanox, Janssen Pharmaceutica Inc.) are potent inhibitors of CYP3A4 and thus can significantly increase blood concentrations and toxicity of other drugs that use this system for detoxification. The toxicity that one encounters with these interactions is simply an extension of the interacting drugs pharmacological effects, as if an overdose of the interacting agent had been administered. For example, erythromycin or ketoconazole, when taken concomitantly with

the nonsedating antihistamines terfenadine or astemizole, can induce prolongation of the Q-T interval on the electrocardiogram and a life-threatening form of ventricular arrhythmias known as torsades de pointes. These effects are identical to what one would see with an overdose of these nonsedating antihistamines.52,53 The inhibition of other cytochrome isoenzymes such as CYP1A2 accounts for erythromycins ability to greatly enhance theophylline blood concentrations and toxicity.51 In addition to case reports, all of the interactions assigned a significance rating of 1 or 2 in Table 2 are supported by pharmacokinetic data consisting of increased half-lives or plasma concentrations of the interacting drug.2,50,51,54-94 In some instances, the actual adverse pharmacodynamic events published in case reports have been reproduced in clinical studies. Typically, the CYP3A4 inhibitor has to be taken for at least three to five days before the interaction occurs, although there have been reports of exposure times as short as one day. Concomitant use of these antimicrobial drugs with the interacting drug should be avoided. Terfenadine and astemizole. The interaction of this type that has received the most attention over the last six years involves the nonsedating antihistamines terfenadine (Seldane, Hoescht Marion Roussel) and astemizole (Hismanal, Jannsen Pharmaceutica Inc.).2,54-59 Terfenadine is actually a prodrug; that is, the parent compound undergoes almost complete metabolism by CYP3A4 in the gut and liver to an active

JADA, Vol. 130, February 1999 Copyright 1998-2001 American Dental Association. All rights reserved.

PHARMACOLOGY
acid metabolite fexofenadine (Allegra, Hoescht Marion Roussel), as shown in Figure 1. In fact, it is highly unusual to detect any terfenadine in the bloodstream of a patient who has ingested the drug.54,58 The active metabolite fexofenadine provides the desired antihistaminic action.50,51,54 The parent compound terfenadine is potentially cardiotoxic, and if its metabolism by CYP3A4 is impaired, terfenadine can accumulate in the body and result in serious cardiac toxicity, including torsades de pointes.54,55,58,59 Like terfenadine, astemizole also undergoes extensive firstpass metabolism by CYP3A4, and it appears that the parent astemizole molecule also is cardiotoxic.51,58 A number of wellcontrolled studies have demonstrated that erythromycin, clarithromycin and the azole antifungal drugs ketoconazole and itraconazole cause an accumulation of both the parent terfenadine molecule and its active metabolite fexofenadine in healthy volunteers.2,54,55,58,59 In addition, electrocardiographic changes (prolonged Q-T intervals) have been detected in some of the study population.54,59 As terfenadine is the cardiotoxic entity, it has been removed from the market and replaced by fexofenadine, its noncardiotoxic active metabolite. Like fexofenadine, loratadine (Claritin, Schering Corporation) does not appear to be cardiotoxic and appears to be free of this adverse drug interaction with erythromycin, clarithromycin or ketoconazole.51,95 It also appears that the related macrolide antibiotic azithromycin does not interact with terfenadine or astemizole.50,51,55 Cisapride with CYP3A4

Terfenadine

OH HO-C N-CH2CH2CH2CH

CH3 C-CH3 CH3

Fexofenadine

OH HO-C N-CH2CH2CH2CH

CH3 C-COOH CH3

Figure 1. Conversion of potentially cardiotoxic terfenadine to fexofenadine by CYP3A4. Erythromycin, clarithromycin, ketoconazole and itraconazole block this process.

inhibitors. Although the interactions with nonsedating antihistamines are fairly wellknown among dental practitioners, serious and potentially fatal ventricular arrhythmias, including torsades de pointes, also have been reported in patients concomitantly taking cisapride (Propulsid, Janssen Pharmaceutica Inc.) and inhibitors of CYP3A4 (including azole antifungal agents, erythromycin, clarithromycin and metronidazole).50,51,60-62,96 Cisapride is a widely prescribed gastrointestinal prokinetic agent, indicated for the treatment of gastroesophageal reflux

disease. As with the nonsedating antihistamine interactions, deaths have been attributed to this interaction.51,60 Studies in healthy volunteers have demonstrated substantial increases in cisapride blood concentrations and prolongation of the Q-T interval on the electrocardiogram when cisapride and ketoconazole were taken concomitantly.62 As with astemizole and terfenadine, the significance rating assigned to the interaction between cisapride and the antimicrobial agents in Table 2 is 1. Triazolam with midazolam. Other adverse drug inter243

JADA, Vol. 130, February 1999 Copyright 1998-2001 American Dental Association. All rights reserved.

PHARMACOLOGY
netic changes.93,94 Fortunately, when taken by mouth, benzodiazepines such as midazolam and triazolam have enormous therapeutic indexes when compared to other classes of antianxiety-sedative drugs. Overdoses of almost 70-fold have not resulted in fatalities.98 Thus, the interaction between macrolide antibiotics or azole antifungal drugs and midazolam or triazolam is given a significance rating of 2. Oversedation with prolonged and intense psychomotor impairment is a likely, although not life-threatening, outcome.
ANTIBIOTICS WITH ORAL CONTRACEPTIVES

Ethinyl estradiol (picogram/milliliter)

2,000

N orethisterone (nanogram/milliliter)

40

1,500 P < .01

30 P < .01

1,000

20

500

10

0 Rifampin Control

0 Rifampin Control

Figure 2. Blood concentrations (mean + standard error) of ethinyl estradiol, or EE, and norethisterone in women taking oral contraceptives in both the presence and absence of rifampin. Blood levels of both EE and norhisterone are significantly reduced in the presence of rifampin. Data from Back and colleagues.107,108

actions involving these antimicrobial inhibitors of the cytochrome P-450 system are summarized in Table 2. They all involve the accumulation of the interacting drug. The adverse drug interactions with the oral antianxiety-sedative agents triazolam and midazolam are especially relevant to dental practitioners who use oral sedation techniques. In one case report, an 8-year-old child who was premedicated with oral midazolam lost consciousness during an erythromycin infusion.89 His peak midazolam concentrations were double the normal values. The child awakened spontaneously 45 minutes later. In a placebo-controlled study of 12 healthy volunteers, erythromycin (500 mg three times a day for six days) increased peak blood levels of a single dose of oral midazolam almost threefold.90 In addition, drowsiness and other indicators of psychomotor impairment were far more intense in subjects taking erythromycin compared to a placebo. The effects of eryth244

romycin on blood levels and sedative effects of intravenous midazolam were far less profound.90 Another study found that even a single 750-mg dose of erythromycin increased the sedative effects of a single 10mg oral dose of midazolam.91 A study of 16 healthy subjects found that 333 mg of erythromycin daily for three days decreased the clearance of a single 0.5 mg dose of triazolam by about half, and approximately doubled total triazolam blood levels (levels in the area under the concentration curve).92 Azithromycin seems free of interactions with midazolam or triazolam.97 Clinical studies have revealed that ketoconazole and itraconazole have even more dramatic effects on these sedative hypnotics, increasing total blood concentrations of oral midazolam and triazolam 15- and 27-fold, respectively, and peak blood levels three- to fourfold. Intense and prolonged (as long as 17 hours) psychomotor impairment accompanied these pharmacoki-

Among the most contentiously debated adverse drug interactions is the alleged ability of a variety of commonly prescribed antibiotics to reduce blood concentrations and the ultimate effectiveness of oral contraceptive agents. Considering the relatively high usage of these agents, the actual scientific documentation for this interaction is far from overwhelming. Before a discussion of the data or lack of data supporting this interaction, a brief review of oral contraceptive pharmacology is in order. The pharmacology of oral contraceptives. The combination pill is made up of two components: semisynthetic estrogens, typically ethinyl estradiol, or EE, or mestranol, and semisynthetic progesterones known as progestins (for example, norethisterone and levonorgestrel). The estrogen component gains its contraceptive efficacy by blocking ovulation by inhibiting the release of follicle-stimulating hormone, or FSH, and luteinizing hormone,

JADA, Vol. 130, February 1999 Copyright 1998-2001 American Dental Association. All rights reserved.

PHARMACOLOGY

A
EE EE EE
Glucuronic Acid

B
EE SO4 EE EE EE
Glucuronic Acid

EE

SO4
Antibiotic

Gut Flora

EE
Intestine Blood

EE

EE

EE

EE

EE
Intestine Blood

EE

EE

EE

EE

EE

EE

Figure 3. The proposed ability of antibiotics to inhibit the enterohepatic recirculation of ethinyl estradiol, or EE. In the absence of antibiotics (A), enteric bacteria hydrolyze glucuronic acid and sulfate groups from EE molecules liberating the lipid-soluble and active parent compound, which can be readily reabsorbed from the intestine into the bloodstream. In the presence of antibiotics (B), the enteric bacteria are markedly diminished, leading to a significant reduction in EE reabsorbed into the bloodstream.

or LH. The mechanisms of action for the progestin component include an enhancement of the viscosity of cervical fluid, a change in the endometrial lining that makes it unsuitable for egg implantation, and some antiovulatory action.99 The components of the pill are not without potential side effects. The most critical side effect of the estrogen component is an increased risk of venous thromboembolytic disease. The progestin component has been associated with increases in blood pressure, blood glucose and blood lipid levels in some women. An increased risk of myocardial infarction and stroke have been reported in oral contraceptive users older than the age of 35 years who are heavy smokers.99 These deleterious events have led to the development of pills containing reduced dosages of both components. Some authors have

speculated that the widespread use of these so-called minipills, which have reduced amounts of both estrogen and progestin components, increase the risk of drug interactions and contraceptive failure.100 Like any drug, oral contraceptives are not 100 percent effective. When taken correctly, they reduce the chance of becoming pregnant to less than 1 percent. However, the typical failure rate among American populations is around 3 percent per year.101 In the teenage population, the failure rate is reported to be as high as 8 percent, probably from teens missing an average of three pills per cycle.102 Interactions with rifampin. The antituberculosis drug rifampin was the first antibiotic implicated in reducing the effectiveness of oral contraceptives. In 1971, Reimers and Jezek103 reported that 38 of 51

women (75 percent) taking rifampin and oral contraceptives concomitantly experienced breakthrough bleeding, an indicator of ovulation. Two years later, another report stated that in 88 women, five pregnancies and 66 instances of breakthrough bleeding were associated with concomitant use of rifampin and oral contraceptives.104 Other reports of pregnancy have been associated with this interaction.105 In fact, 76 percent of all alleged antibiotic-oral contraceptive interactions involve rifampin.106 Clinical studies clearly demonstrate that rifampin significantly reduces blood levels of both the estrogen and progestin components of oral contraceptives (Figure 2).107,108 Rifampin is a potent inducer of the liver cytochrome P-450 system and increases the metabolism of a number of drugs, including oral contracep245

JADA, Vol. 130, February 1999 Copyright 1998-2001 American Dental Association. All rights reserved.

PHARMACOLOGY
cillin combination during dental impaction surgery. At the threemonth follow-up, she was found to be pregnant; 40 weeks after surgery, she gave birth to healthy twins.112 Probably the most comprehensive report of potential antibiotic-oral contraceptive interactions was supplied by Back and colleagues,113 who gathered data from the United Kingdoms Committee on Safety in Medicines between 1968 and 1984. During this period, 63 pregnancies were reported with simultaneous administration of oral contraceptives and antibiotics (excluding rifampin). Penicillins were implicated in 32 of these pregnancies, tetracyclines in 12, cotrimoxazole (sulfamethoxazole and trimethoprim) in five, metronidazole in three, erythromycin in two and antibiotics not commonly used in dentistry in the other nine. The authors tempered these findings by also reporting that in England between 1973 and 1984, there was a total of 159,000,000 prescriptions for penicillins; 68,000,000 for tetracyclines; 40,000,000 for cotrimoxazole; 31,000,000 for erythromycin; and 6,000,000 for metronidazole. In addition, they reported, there were approximately 2,500,000 regular users of oral contraceptives per year during this period.114 Thus, the actual number of reported pregnancies in England alleged to involve oral contraceptive interactions with antibiotics other than rifampin might be, in reality, extremely low. In the United States, 29 reports of unintended pregnancies in oral contraceptive users who received penicillins or tetracyclines were listed in the U.S. Department of Health and

Ethinyl estradiol (picogram/milliliter)

1,200 1,000 800 600 400 200 0

Control

Tetracycline Tetracycline Day 1 Days 5-10

Figure 4. Blood concentrations (mean + standard error) of ethinyl estradiol, or EE, in women taking oral contraceptives in both the absence and the presence of tetracycline. There was no effect on blood levels of EE following one day or five to 10 days of antibiotic therapy. Data from Murphy and colleagues.132

tives.51,105 At present, rifampin remains the only antibiotic that has been scientifically demonstrated to interfere with the effectiveness of oral contraceptive agents. Interactions with other antibiotics. Case reports of more commonly prescribed antibiotics interfering with oral contraceptive effectiveness began to surface in 1975. In that year, Dossetor109 reported three cases in her practice in which patients taking oral contraceptives apparently had become pregnant when given ampicillin. Five years later, another report described a 20year-old student who claimed to be totally compliant with her oral contraceptive regimen, but nevertheless became pregnant after a five-day course of tetracycline.110 In 1982, DeSano and Hurley111 described 16 pregnancies over a two-year period in their private obstetric and gyne246

cologic practices, all in patients who claimed to be totally compliant with their contraceptive regimen but who also had consumed additional medications. Antibiotics had been consumed in 13 of these cases: ampicillin in five, penicillin in three, tetracycline in one and antibiotics not commonly used in dentistry in the other four. However, four of the patients taking ampicillin and two of those taking penicillin also had taken a variety of other medications, including antihistamines, decongestants and analgesics. All of these case reports were, by necessity, uncontrolled, retrospective and subject to recall bias.105 In 1986, a case report of an alleged antibiotic-oral contraceptive interaction appeared in the dental literature. A 19-yearold who had taken an oral contraceptive for 18 months received an intramuscular injection of a long-acting peni-

JADA, Vol. 130, February 1999 Copyright 1998-2001 American Dental Association. All rights reserved.

PHARMACOLOGY
Human Services MEDWATCH Spontaneous Reporting System.115 These numbers again must be tempered with the fact that as of 1994, approximately 11,000,000 women per year were using oral contraceptives in the United States.116 Assumptions in the literature. Although these case reports certainly should not be ignored and theoretically could indicate a rare interaction, a number of reviews have been published implying that the ability of commonly prescribed antibiotics to reduce the effectiveness of oral contraceptives is an established interaction.117-119 For example, one otherwise excellent article in the dental literature discussing the impact of the 50 drugs most frequently dispensed in clinical practice contains a statement that the antibiotics that interfere with the ovulatory inhibiting effects of oral contraceptives are penicillin V potassium, amoxicillin, cephalexin, tetracycline and erythromycins.117 The author cites as his evidence one of the previously described case reports112 and not any clinical trials or pharmacokinetic data. In 1991, the ADA Health Foundation Research Institute published a statement concerning this issue. A small portion of the statement read as follows: Unfortunately, many antibiotics commonly used in dentistry interfere with the action of oral contraceptives, resulting in unexpected pregnancies. Failure to inform a patient using oral contraceptives during antibiotic therapy resulting in a birth could leave the dentist responsible for damagesincluding child support payments.120 While it was very important to warn dentists of the possibility of this interaction, there were not enough scientific data to support the statement that the interaction was indeed established. Even the Physicians Desk Reference121 at best describes the interaction with antibiotics other than rifampin as possible. Other authors have reported one or two successful litigations against dentists.115,122 Unfortunately, these legal proceedings cannot be researched or even substantiated.123 The pharmacological basis of the interaction. Antibiotics ability to inhibit the enterohepatic recirculation of the estrogen component of the oral contraceptive is the theory most often mentioned in the literature to explain the alleged interaction124-126 (Figure 3). In humans, the bioavailability of EE is only about 40 percent to 50 percent because of a large first-pass metabolism in the gut and the liver. These inactive conjugated metabolites (50 percent to 60 percent of the parent compound) then are excreted in the bile, where the drug would be lost if not for the bacteria in the colon, which are thought to split the EE conjugates. This liberates the active parent compound, which can readily be reabsorbed in the intestine and provide the necessary additional blood levels of the drug. In theory, this enterohepatic recirculation can be inhibited by antibiotics that kill the colonic bacteria involved in the deconjugation process. In fact, studies in rats and rabbits do show that pretreatment with ampicillin interferes with the enterohepatic recirculation process.127-128 However, no clinical study has been able to demonstrate that a similar scenario also occurs in humans. Proving the interaction. A study many reference as scientific proof of an interaction between antibiotics and oral contraceptives is one in which treatment with ampicillin reduced endogenous estrogen concentrations in pregnant patients.129 However, this study did not evaluate the effects of antibiotics on the blood levels of the estrogen or progestin component of oral contraceptives, and any attempt to correlate ampicillins ability to reduce endogenous estrogen in pregnant patients with the failure of oral contraceptives is scientifically questionable. A number of studies, however, have looked directly at oral contraceptive blood concentrations both in the absence and the presence of antibiotics. Studies showed no significant decreases in plasma levels of either the EE or the progestin contraceptive component in women treated with ampicillin,130,131 tetracycline,125,132 doxycycline,133 metronidazole,130 erythromycin,125 clarithromycin,134 temafloxacin135 or fluconazole.136 Figure 4 illustrates the results of one such study. The antibiotic cotrimoxazole actually increased EE levels significantly,137 whereas clarithromycin significantly reduced FSH and LH levels134 all of these being possible indicators of increased contraceptive efficacy. In another study of 22 women taking oral contraceptives, none of the subjects ovulated while ingesting roxithromycin (a macrolide related to erythromycin), but 11 (50 percent) ovulated while ingesting the positive control rifampin.138 There is one potential weakness in all these studies: their
247

JADA, Vol. 130, February 1999 Copyright 1998-2001 American Dental Association. All rights reserved.

PHARMACOLOGY
relatively small sample sizes (ranging from six to 25 women). If the interaction occurs rarely, as in one in 1,000 people, studies of this size may not detect the interaction. It should be noted, however, that these relatively small sample sizes readily display the rifampinoral contraceptive interaction.107,108,138 There are two retrospective studies, with somewhat larger databases, that have examined pregnancy rates in women who consumed antibiotics as dermatologic patients. (The issue of potential antibiotic interactions with oral contraceptives is of great importance to dermatologists, because they often prescribe antibiotics on a chronic basis to women of child-bearing potential.) In the first study, 281 patients were surveyed; 34 were found to have used antibiotics and low-estrogen contraceptives for a combined total of 71 years.139 One woman who concurrently took tetracycline and oral contraceptives for 12 months became pregnant, giving an overall pregnancy rate of 1.4 percent per year among patients taking antibiotics. In a larger study of 356 patients with a history of combined antibiotic-oral contraceptive use and of 425 women taking oral contraceptives and no antibiotics, researchers found a pregnancy rate of 1.6 percent in the antibiotic group and 0.96 percent in the control group.101 There was no significant difference (P = .4) in pregnancy rates between the antibiotic and control groups, and both groups had pregnancy rates below the 3 percent typically found in the United States. Are there any data other than case reports that support
248

the possibility of an interaction between commonly prescribed antibiotics and oral contraceptives? Shenfield and Griffin140 presented a single case study of an oral contraceptive user who had a history of breakthrough

The failure of oral contraceptives in women simultaneously ingesting antibiotics may indicate background noise that is, the normal failure rate of these drugs or a relatively rare interaction that cannot be detected by clinical trials.
bleeding while on antibiotics. While taking minocycline, this woman displayed lower EE plasma concentrations and the abolition of two of three plasma peaks, as compared with when she was not taking the antibiotic.140 It was hypothesized that this conceivably could indicate an antibiotic-induced reduction of the enterohepatic recirculation process in this subject. One other study, while revealing no effect of ampicillin on plasma EE levels of the entire study population, did identify two women in the sample whose EE concentrations were significantly reduced while they were taking ampicillin.131 However, neither of these women experienced breakthrough bleeding or changes in laboratory parameters that would indicate ovulation. Interaction vs. normal contraceptive failure rate.

The failure of oral contraceptives in women simultaneously ingesting antibiotics may indicate background noisethat is, the normal failure rate of these drugs or a relatively rare interaction that cannot be detected by clinical trials. Although the interaction, by definition, cannot be classified as established, probable or even suspected, the philosophy of recommending the use of additional forms of birth control for all oral contraceptive users receiving antibiotic therapy, as a possible protection of a few of them from unwanted pregnancies, still seems justified. Of note, however, are the recently published legal proceedings in which a plaintiff and her husband sued a gynecologist and an oral surgeon for malpractice and wrongful life after she allegedly became pregnant either during or shortly after she was given a prescription for penicillin V.141 Both doctors had failed to warn her of a potential interaction with the oral contraceptive she was taking. She and her husband lost the case for the following reasons: (1) Her experts cited review articles with no data and articles that showed no statistically significant correlation between antibiotic use and oral contraceptive failure; (2) her experts failed to show a scientifically validated interaction between penicillin V and the oral contraceptive she was taking; (3) under California law, physicians do not have to discuss risks of a very low incidence (the failure rate would have to be more than double the normal expected rate); and (4) her experts also failed to prove that she became pregnant when she was taking penicillin.

JADA, Vol. 130, February 1999 Copyright 1998-2001 American Dental Association. All rights reserved.

PHARMACOLOGY
CONCLUSION Surg Gynec Obstet 1962;114(1):9-14. 9. Neuvonen PJ, Gothoni G, Hackman R, Bjorksten K. Interference of iron with the absorption of tetracyclines in man. Br Med J 1970;4(734):532-4. 10. Neuvonen P, Matilla M, Gothoni H, Hackman R. Interference of iron and milk with absorption of tetracycline. Scand J Clin Lab Invest 1971;27(suppl 116):76. 11. Leyden JJ. Absorption of minocycline hydrochloride and tetracycline hydrochloride. effects of food, milk and iron. J Am Acad Dermatol 1985;12(2):308-12. 12. Waisbren BA, Hueckel JS. Reduced absorption of Aureomycin caused by aluminum hydroxide gel (Amphogel). Proc Soc Exp Biol Med NY 1956;73(1):73-4. 13. Harcourt RS, Hamburger M. The effect of magnesium sulfate in lowering tetracycline blood levels. J Lab Clin Med 1957;50(3):464-8. 14. Albert KS, Welch RD, DeSante KA, DiSanto AR. Decreased tetracycline bioavailability caused by a bismuth subsalicylate antidiarrheal mixture. J Pharm Sci 1979;68(5):586-8. 15. Pentilla O, Hurme H, Neuvonen PJ. Effect of zinc sulphate on the absorption of tetracycline and doxycycline in man. Eur J Clin Pharmacol 1975;9(2/3):131-4. 16. Fried R, Fried LW. The effect of Flagyl on xanthine oxidase and alcohol dehydrogenase. Biochem Pharmacol 1966;15(11):1890-3. 17. Ban TA, Lehmann HE, Roy P. Preliminary report of the therapeutic effect of Flagyl in alcoholism. Union Med Can 1966;95(2):147-9. 18. Sansoy OM. Evaluation of metronidazole in the treatment of alcoholism: a comprehensive three-year study comprising 60 cases. Rocky Mt Med J 1970;67(2):43-7. 19. Penick SB, Carrier RN, Sheldon JR. Metronidazole in the treatment of alcoholism. Am J Psychiatry 1969;125(8):1063-6. 20. Flagyl (metronidazole capsules). In: Physicians desk reference. 52nd ed. Montvale, N.J.: Medical Economics Company; 1998:2738-40. 21. Potter WZ, Hollister LE. Antipsychotic agents and lithium. In: Katzung BG, ed. Basic and clinical pharmacology. 7th ed. Stamford, Conn.: Appleton & Lange; 1998:464-82. 22. Ayd FJ Jr. Metronidazole-induced lithium intoxication. Int Drug Ther News 1982;17(1):15-6. 23. Teicher MH, Altesman RI, Cole JO, Schatzberg AF. Possible nephrotoxic interaction of lithium and metronidazole. JAMA 1987;257(24):3365-6. 24. Heng MCY. Cutaneous manifestations of lithium toxicity. Br J Dermatol 1982;106(1):107-9. 25. McGennis AJ. Lithium carbonate and tetracycline interaction. Br Med J 1978;1(6121):1183. 26. Jefferson JW. Lithium and tetracycline. Br J Dermatol 1982;107(3):370. 27. Fankhauser MP, Lindon JL, Conolly B, Healey WJ. Evaluation of lithium-tetracycline interaction. Clin Pharm 1988;7(4):314-7. 28. Katzung BG, Parmley WW. Cardiac glycosides and other drugs used in congestive heart failure. In: Katzung BG, ed. Basic and clinical pharmacology. 7th ed. Stamford, Conn.: Appleton & Lange; 1998:197-215. 29. Lindenbaum J, Rund DG, Butler VP, Tse-Eng D, Saha JR. Inactivation of digoxin by the gut flora: reversal by antibiotic therapy. N Engl J Med 1981;305(14):789-94. 30. Friedman HS, Bonventre MV. Erythromycin-induced digoxin toxicity. Chest 1982;82(2):202. 31. Maxwell DL, Gilmour-White SK, Hall MR. Digoxin toxicity due to interaction of digoxin with erythromycin. Br Med J 1989;298(6673):572. 32. Dobkin JF, Saha JR, Butler VP, Lindenbaum J. Effect of antibiotic therapy on digoxin metabolism. Clin Res 1982;30(2):517A. 33. Westfall LK, Mintzer DL, Wiser TH. Potentiation of warfarin by tetracycline. Am J Hosp Pharm 1980;37(12):1620-5. 34. Caraco Y, Rubinow A. Enhanced anticoagulant effect of coumarin derivatives induced by doxycycline coadministration. Ann Pharmacother 1992;26(9):1084-6. 35. ODonnell D. Antibiotic induced potentiation of oral anticoagulant agents. Med J Aust 1989;150(3):163-4. 36. Messinger WJ, Samet CM. The effect of bowel sterilizing antibiotics on blood coagulation mechanisms. Angiology 1965;16(1):29-36. 37. Soto J, Sacristan JA, Alsar MJ, Fernandez-Viadero C, Verduga R. Probable acenocoumarol-amoxycillin interaction. Acta Haematol 1993;90(4):195-7. 38. Udall JA. Human sources and absorption of vitamin K in relation to anticoagulant stability. JAMA 1965;194(2):127-9. 39. Pineo GF, Gallus AS, Hirsh J. Unexpected vitamin K deficiency in hospitalized patients. Can Med Assoc J 1973;109(9):880-3. 40. Bartle WP. Possible warfarin-erythromycin interaction. Arch Intern Med 1980;140(7):985-7. 41. Husseri FE. Erythromycin-warfarin interaction. Arch Intern Med 1983;143(9):18316. 42. Schwartz JI, Bachmann K. Erythromycin-warfarin interaction. Arch Intern Med 1984;144(10):2094. 43. Sato RI, Gray DR, Brown SE. Warfarin interaction with erythromycin. Arch Intern Med 1984;144(12):2413-4. 44. Hassell D, Utt JK. Suspected interaction: warfarin and erythromycin. South Med J 1985;78(8):1015-6. 45. Bussey HI, Knodel LC, Boyle DA. Warfarin-erythromycin interaction. Arch Intern Med 1985;145(9):1736-7. 46. Kazmier FJ. A significant interaction between metronidazole and warfarin. Mayo Clin Proc 1976;51(12):782-4. 47. Bachmann K, Schwartz JI, Forney R, Frogameni A, Jauregui LE. The effect of erythromycin on the disposition kinetics of warfarin. Pharmacology 1984;28(3):171-6. 48. OReilly RA. The stereoselective interaction of warfarin and metronidazole in man. N Engl J Med 1976;295(7):354-7. 49. Yagiela JA, Neidle EA, Dowd FJ, eds. Pharmacology and therapeutics for dentistry. 4th ed, St Louis: Mosby Year Book; 1998. 50. Cupp MJ, Tracy TS. Cytochrome P450: new nomenclature and clinical implications. Am Fam Physician 1998;57(1):107-16. 51. Michalets EL. Update: clinically significant cytochrome P-450 drug interactions. Pharmacotherapy 1998;18(1):84-112. 52. Woosley RL, Chen Y, Freiman JP. Mechanism of cardiotoxic action of terfenadine. JAMA 1993;269(12):1532-6. 53. Snook J, Boothman-Burell D, Watkins J, Colin-Jones D. Torsades de pointes ventricular tachycardia associated with astemizole overdose. Br J Clin Pract 1988;42(6):257-9. 54. Honig PK, Woosley RL, Zamani K,

Adverse drug interactions involving antimicrobial agents can occur in dental patients. Of particular significance is the ability of certain antimicrobial agents to cause the accumulation of a number of drugs that have low therapeutic indexes. It is important that clinicians stay abreast of potential drug interactions to avoid serious morbidity among their patients. s
Dr. Hersh is an associate professor, University of Pennsylvania, School of Dental Medicine, Department of Oral Surgery and Pharmacology, 4001 Spruce St., Philadelphia, Pa. 19104-6003. Address reprint requests to Dr. Hersh. This article is based on material presented March 4, 1998, in a symposium entitled Adverse Drug Interactions in Dentistry: Separating the Myths From the Facts. Dr. Hersh and Dr. Paul Moore cochaired this symposium, which was presented at the 27th General Session of the American Association for Dental Research in Minneapolis. The symposium was jointly sponsored by the International Association for Dental Research, the American Association of Dental Schools and the American Dental Association and was supported in part by a grant from Warner Lambert Pharmaceuticals. The drugs listed by brand name in this article are given as examples only. Their listing does not imply any endorsement. 1. Moore PA, Gage TW, Hersh EV, Yagiela JA, Haas DA. Adverse drug interactions in dental practice: professional and educational implications. JADA 1999;130:47-54. 2. Stockley IH. Drug interactions. 4th ed. London: Pharmaceutical Press; 1996. 3. Lepper MH, Dowling HF. Treatment of pneumococcic meningitis with penicillin compared with penicillin plus aureomycin. Arch Intern Med 1951;88(4):489-94. 4. Olson RA, Kirby JC, Romansky MJ. Pneumococcal meningitis in the adult: clinical, therapeutic, and prognostic aspects in forty-three patients. Arch Intern Med 1961;55(4):545-9. 5. Strom J. Penicillin and erythromycin singly and in combination in scarlatina therapy and the interference between them. Antibiot Chemother 1961;11(11):694-7. 6. Manten A. Synergism and antagonism between antibiotic medicines containing erythromycin. Antibiot Chemother 1954;4(12):1228-33. 7. Graham DY, Lew GM, Klein PD, et al. Effect of treatment of Helicobacter pylori infection on the long-term recurrence of gastric and duodenal ulcer. Ann Intern Med 1992;116(9):705-9. 8. Scheiner JA, Altemeier WA. Experimental study of factors inhibiting absorption and effective therapeutic levels of declomycin.

JADA, Vol. 130, February 1999 Copyright 1998-2001 American Dental Association. All rights reserved.

249

PHARMACOLOGY
Conner DP, Cantilena LR. Changes in the pharmacokinetics and electrocardiographic pharmacodynamics of terfenadine with concomitant administration of erythromycin. Clin Pharmacol Ther 1992;52(3):231-8. 55. Honig P, Wortham D, Zamani K, Conner D, Cantilena L. Effect of erythromycin, clarithromycin and azithromycin on pharmacokinetics of terfenadine. Clin Pharmacol Ther 1993;53(2):161. 56. Biglin KE, Faraon MS, Constance TD, Lieh-Lai M. Drug-induced torsades de pointes: a possible interaction of terfenadine and erythromycin. Ann Pharmacother 1994;28(2):282. 57. Goss JE, Ramo BW, Blake K. Torsades de pointes associated with astemizole (Hismanal) therapy. Arch Intern Med 1993;153(23):2705. 58. Krvisto KT, Neuvonen PJ, Klotz U. Inhibition of terfenadine metabolism: pharmacokinetic and pharmacodynamic consequences. Clin Pharmacokinet 1994;27(1):1-5. 59. Honig PK, Wortham DC, Zamani K, Conner DP, Mullin JC, Cantilena LR. Terfenadine-ketoconazole interaction: pharmacokinetic and electrocardiographic consequences. JAMA 1993;269(12):1513-8. 60. Wysowski DK, Bacsanyi J. Cisapride and fatal arrhythmia. N Engl J Med 1996;335(4):290-1. 61. Sekkarie MA. Torsades de pointes in two chronic renal failure patients treated with cisapride and clarithromycin. Am J Kid Dis 1997;30(3):437-9. 62. Propulsid (cisapride tablets/suspension). In: Physicians desk reference. 52nd ed. Montvale, N.J.: Medical Economics Company; 1998:52:1308-9. 63. Bartkowski RR, McDonnell TE. Prolonged alfentanil effect following erythromycin administration. Anesthesiology 1990;73(3):566-8. 64. Bartkowski RR, Goldberg ME, Larijani GE, Boerner T. Inhibition of alfentanil metabolism by erythromycin. Clin Pharmacol Ther 1989;46(1):99-102. 65. Nelson MV, Berchou RC, Kareti D, LeWitt PA. Pharmacokinetic evaluation of erythromycin and caffeine administered with bromocriptine. Clin Pharmacol Ther 1990;47(6):694-7. 66. Wong YY, Ludden TM, Bell RD. Effect of erythromycin on carbamazepine kinetics. Clin Pharmacol Ther 1983;33(4):460-4. 67. Wroblewski BA, Singer WD, Whyte J. Carbamazepine-erythromycin interaction: case studies and clinical significance. JAMA 1986;255(9):1165-7. 68. Macnab AJ, Robinson JL, Adderly RJ, DOrsogna L. Heart block secondary to erythromycin-induced carbamazepine toxicity. Pediatrics 1987;80(6):951-3. 69. OConnor NK, Fris J. Clarithromycincarbamazepine interaction in a clinical setting. J Am Board Fam Pract 1994;7(6):489-92. 70. Albani F, Riva R, Baruzzi A. Clarithromycin-carbamazepine interaction: a case report. Epilepsia 1993;34(1):161-2. 71. First MR, Schroeder TJ, Alexander JW, et al. Cyclosporine dose reduction by ketoconazole administration in renal transplant recipients. Transplantation 1991;51(2):36570. 72. Ferguson RM, Sutherland DE, Simmonds RL, Najaran JS. Ketoconazole, cyclosporin metabolism and renal transplantation. Lancet 1982;ii(Oct 16):882-3. 73. Gomez DY, Wacher VJ, Tomlanovich SJ, Herbert MF, Benet LZ. The effects of ketoconazole on the intestinal metabolism and bioavailability of cyclosporine. Clin Pharmacol Ther 1995;58(1):15-9. 74. Ferrari SL, Goffin E, Mourad M, Wallemacq P, Squifflet JP, Pirson Y. The interaction between clarithromycin and cyclosporine in kidney transplant recipients. Transplantation 1994;58(6):725-7. 75. Jensen CW, Flechner SM, Van Buren CT, et al. Exacerbation of cyclosporine toxicity by concomitant administration of erythromycin. Transplantation 1987;43(2):26370. 76. Freeman DJ, Martell R, Carruthers SG, Heinrichs D, Keown PA, Stiller CR. Cyclosporin-erythromycin interaction in normal subjects. Br J Clin Pharmacol 1987;23(6):776-8. 77. Neuvonen PJ, Suhonen R. Itraconazole interacts with felodipine. J Am Acad Dermatol 1995;33(1):134-5. 78. Bailey DG, Arnold JMO, Tran LT, Ahktar J. Marked effects of both erythromycin and grapefruit juice on felodipine pharmacokinetics. Clin Pharmacol Ther 1994;55(2):165. 79. Glynn AM, Slaughter RL, Brass C, DAmbrosio R, Jusko WJ. Effects of ketoconazole on methylprednisolone pharmacokinetics and cortisol secretion. Clin Pharmacol Ther 1986;39(6):654-9. 80. Zurcher RM, Frey BM, Frey FJ. Impact of ketoconazole on the metabolism of prednisolone. Clin Pharmacol Ther 1989;45(4):366-72. 81. LaForce CF, Szefler SJ, Miller MF, Ebling W, Brenner M. Inhibition of methylprednisolone elimination in the presence of erythromycin therapy. J Allergy Clin Immunol 1983;72(1):34-9. 82. Reisz G, Pingleton SK, Melethil S, Ryan P. The effect of erythromycin on theophylline pharmacokinetics in chronic bronchitis. Am Rev Respir Dis 1983;127(5):581-4. 83. Cummins LH, Kozak PP, Gilman SA. Erythromycins effects on theophylline blood levels. Pediatr 1977;59(1):144-5. 84. May DC, Jarboe CH, Ellenberg DT, Roe EJ, Karibo J. The effects of erythromycin on theophylline elimination in normal males. J Clin Pharmacol 1982;22(2/3):125-30. 85. Zarowitz BJM, Szefler SJ, Lasezkay GM. Effect of erythromycin base on theophylline kinetics. Clin Pharmacol Ther 1981;29(5):601-5. 86. Spach DH, Bauwens JE, Clark CD, Burke WG. Rhabdomyolysis associated with lovastatin and erythromycin use. West J Med 1991;154(2):213-5. 87. Lees RS, Lees AM. Rhabdomyolysis from the coadministration of lovastatin and the antifungal agent itraconazole. N Engl J Med 1995;333(10):664-5. 88. Neuvonen PJ, Jalava K. Itraconazole drastically increases plasma concentrations of lovastatin and lovastatin acid. Clin Pharmacol Ther 1996;60(1):54-61. 89. Hiller A, Olkkola KT, Isohanni P, Saarnivaara L. Unconsciousness associated with midazolam and erythromycin. Br J Anaesth 1990;65(6)826-8. 90. Olkkola KT, Aranko K, Luurila H, et al. A potentially hazardous interaction between erythromycin and midazolam. Clin Pharmacol Ther 1993;53(3):298-305. 91. Mattila MJ, Idanpaan-Heikila JJ, Tornwall M, Vanakoski J. Oral single doses of erythromycin and roxithromycin may increase the effects of midazolam on human performance. Pharmacol Toxicol 1993;73(3):180-5. 92. Phillips JP, Antal EJ, Smith RB. A pharmacokinetic interaction between erythromycin and triazolam. J Clin Psychopharmacol 1986;6(5):297-9. 93. Olkkola KT, Backman JT, Neuvonen PJ. Midazolam should be avoided in patients receiving systemic antimycotics ketoconazole or itraconzaole. Clin Pharmacol Ther 1994;55(5):481-5. 94. Varhe A, Olkkola KT, Neuvonen PJ. Oral triazolam is potentially hazardous to patients receiving systemic antimycotics ketoconazole or itraconazole. Clin Pharmacol Ther 1994; 56(6):601-7. 95. Abramowicz M, ed. Fexofenadine. Med Lett Drugs Ther 1996;38:95-6. 96. Bran S, Murray W, Hirsch IB, Palmer JP. Long QT syndrome during high-dose cisapride. Arch Intern Med 1995;155(7):765-8. 97. Mattila MJ, Vanakoski J, IdanpaanHeikkila JJ. Azithromycin does not alter the effects of oral midazolam on human performance. Eur J Clin Pharmacol 1994;47(1):4952. 98. Loeffler PM. Oral benzodiazepines and conscious sedation: a review. J Oral Maxillofac Surg 1992;50(9):989-97. 99. Goldfien A. The gonadal hormones and inhibitors. In: Katzung BG, ed. Basic and clinical pharmacology. 7th ed. Stamford, Conn.: Appleton & Lange; 1998:653-83. 100. Barnett ML. Inhibition of oral contraceptive effectiveness by concurrent antibiotic administration. J Periodontol 1985;56(1):1820. 101. Helms SE, Bredle DL, Zajic J, Jarjoura D, Brodell RT, Krishnarao I. Oral contraceptive failure rates and oral antibiotics. J Am Acad Dermatol 1997;36(5):705-10. 102. Sondheimer SJ. Update on oral contraceptive pills and postcoital contraception. Curr Opin Obstet Gynecol 1992;4(4):502-5. 103. Reimers D, Jezek A. The simultaneous use of rifampicin and other antitubercular agents with oral contraceptives. Prax Pneumol 1971;25(5):255-62. 104. Noche-Finck l, Breuer H, Reimers D. Effects of rifampicin on the menstrual cycle and on estrogen excretion in patients taking oral contraceptives. JAMA 1973;226(3):378. 105. Shenfield GM. Oral contraceptives. Are drug interactions of clinical significance? Drug Safety 1993;9(1):21-37. 106. Szoka PR, Edgren RA. Drug interactions with oral contraceptives: compilation and analysis of an adverse experience report data base. Fertil Steril 1988;49(5 suppl 2):31S-8S. 107. Back DJ, Breckenridge AM, Crawford FE, et al. The effect of rifampin on the pharmacokinetics of ethinylestradiol in women. Contraception 1980;21(2):135-43. 108. Back DJ, Breckenridge AM, Crawford F, et al. The effect of rifampin on norethisterone pharmacokinetics. Eur J Pharmacol 1979;15(3):193-7. 109. Dossetor J. Drug interaction with oral contraceptives. Br Med J 1975;4(5994):46768. 110. Bacon JF, Shenfield GM. Pregnancy attributable to interaction between tetracycline and oral contraceptives. Br Med J 1980;280(6210):293. 111. DeSano EA, Hurley SC. Possible interactions of antihistamines and antibiotics with oral contraceptive effectiveness. Fertil Steril 1982;37(6):853-4. 112. Bainton R. Interaction between antibiotic therapy and contraceptive medication.

250

JADA, Vol. 130, February 1999 Copyright 1998-2001 American Dental Association. All rights reserved.

PHARMACOLOGY
Oral Surg Oral Med Oral Pathol 1986;61(5)453-5. 113. Back DJ, Grimmer SF, Orme ML, Proudlove C, Mann RD, Breckenridge AM. Evaluation of Committee on Safety of Medicines yellow card reports on oral contraceptive-drug interactions with anticonvulsants and antibiotics. Br J Clin Pharmacol 1988;25(5):527-32. 114. Orme ML, Back DJ. Interactions between oral contraceptive steroids and broadspectrum antibiotics. Clin Exp Dermatol 1986;11(4):327-31. 115. Donley TG, Smith RF, Roy B. Reduced oral contraceptive effectiveness with concurrent antibiotic use: a protocol for prescribing antibiotics to women of childbearing age. Compend Contin Educ Dent 1990;9(6):392-6. 116. Miller DM, Helms SE, Brodell R. A practical approach to antibiotic treatment in women taking oral contraceptives. J Am Acad Dermatol 1994;30(6):1008-11. 117. Wynn RL. The top 50 drugs dispensed by pharmacies in 1996. Gen Dent 1997;45(5):416-20. 118. Fazio A. Oral contraceptive drug interactions: important considerations. South Med J 1991;84(8):997-1002. 119. Pyle MA, Faddoul FF, Terezhalmy GT. Clinical implication of drugs taken by our patients. Dent Clin North Am 1993;37(1):73-90. 120. ADA Health Foundation Research Institute, Department of Toxicology. Antibiotic interference with oral contraceptives. JADA 1991;122(12):79. 121. Ortho-Novum 1/50 tablets (norethindrone/mestranol), Ortho-Novum tablets (norethindrone/ethinyl estradiol). In: Physicians desk reference. 52nd ed. Montvale, N.J.: Medical Economics Company; 1998;52:2027-40. 122. Phiras D. Wrongful birth and birth injury claims: new risks for oral surgeons. J Mass Dent Soc 1986;41(2):82-3. 123. Becker DE. Drug interactions in dental practice: a summary of facts and controversies. Compend Contin Educ Dent 1994;15(10):1228-44. 124. Shenfield GM. Drug interactions with oral contraceptive preparations. Med J Aust 1986;144(4):205-11. 125. Orme ML, Back DJ. Interactions between oral contraceptive steroids and broadspectrum antibiotics. Clin Exp Dermatol 1986;11(4):327-31. 126. Orme M, Back DJ. Oral contraceptive steroidspharmacological issues of interest to the prescribing physician. Adv Contracept 1991;7(4):325-31. 127. Back DJ, Breckenridge AM, Challiner M, et al. The effect of antibiotics on the enterohepatic recirculation of ethinylestradiol and norhisterone in the rat. J Ster Biochem 1978;9(6):527-31. 128. Back DJ, Breckenridge AM, Cross KJ, Orme ML, Thomas E. An antibiotic interaction with ethinyloestradiol in the rat and rabbit. J Steroid Biochem 1982;16(3):407-13. 129. Willman K, Pulkkinen MO. Reduced maternal plasma and urinary estriol during ampicillin treatment. Am J Obstet Gynecol 1971;109(6):893-6. 130. Joshi JV, Joshi UM, Sankholi GM, et al. A study of interaction of low-dose combination oral contraceptive with ampicillin and metronidazole. Contraception 1980;22(6):64352. 131. Back DJ, Breckenridge AM, MacIver M, et al. The effects of ampicillin on oral contraceptive steroids in women. Br J Clin Pharmacol 1982;14(1):43-8. 132. Murphy AA, Zacur HA, Charache P, Burkman RT. The effect of tetracycline on levels of oral contraceptives. Am J Obstet Gynecol 1991;164(1):28-33. 133. Neely JL, Abate M, Swinker M, DAngio R. The effect of doxycycline on serum levels of ethinyl estradiol, norethindrone, and endogenous progesterone. Obstet Gynecol 1991;77(3):416-20. 134. Back DJ, Tjia J, Martin C, Millar E, Salmon P, Orme M. The interaction between clarithromycin and oral contraceptive steroids. J Pharm Med 1991;2(1):81-7. 135. Back DJ, Tjia J, Martin C, et al. The lack of interaction between temafloxacin and combined oral contraceptive steroids. Contraception 1991;43(4):317-23. 136. Lazar JD, Wilner KD. Drug interactions with fluconazole. Rev Infect Dis 1990;12(suppl 3):S327-33. 137. Grimmer SF, Allen WL, Back DJ, Breckenridge AM, Orme M, Tija J. The effect of cotrimoxazole on oral contraceptive steroids in women. Contraception 1983;28(1):53-9. 138. Meyer B, Muller F, Wessels P, Maree J. A model to detect interactions between roxithromycin and oral contraceptives. Clin Pharmacol Ther 1990;47(6):671-4. 139. London BM, Lookingbill DP. Frequency of pregnancy in acne patients taking oral antibiotics and oral contraceptives. Arch Dermatol 1994;130(3):392-3. 140. Shenfield GM, Griffin JM. Clinical pharmacokinetics of contraceptive steroids: an update. Clin Pharmacokinet 1991;20(1):1537. 141. California court denies wrongful birth claim. J Law Med Ethics 1996;24(3):273-4.

JADA, Vol. 130, February 1999 Copyright 1998-2001 American Dental Association. All rights reserved.

251