Use of astemizole in a large group practice

TIMOTHY J. CRAIG, DO MARK GREENWALD, MD VANESSA KAUFFMANN JILL CRAIG, MS

Astemizole was released in 1988. Shortly after the release of astemizole and terfe­ In late 1992, a new warning label was added nadine, it became evident that both were associat­ in response to reports of syncope and death ed with prolonged QT intervals and arrhythmias, from arrhythmia. Records of patients given new mainly torsades de pointes. Because of this associ­ prescriptions for astemizole were reviewed ation, the Food and Drug Administration (FDA) and to assess compliance with the warnings in a the manufacturers provided a warning letter to large multispecialty practice. The indication physicians in 1992.1 For astemizole, the warning was appropriate in 89% of cases. Excessive stated: (1) That arrhythmias have usually occurred doses were used in 4% of cases. Two percent when the dose of 10 mg/d (the recommended dose) of prescriptions were given to patients with was exceeded. Exceeding this dose and the use of contraindications. Only two complications loading doses should be avoided. (2) Serum levels were documented. Despite carrying a drug of astemizole may be elevated by ketoconazole, ery­ warning, astemizole continues to be used thromycin;, and itraconazole. These drugs should inappropriately and is a medicolegal concern. not be used together. (3) Presyncope and syncope Education and drug evaluations can be used may precede fatal arrhythmias and calls for dis­ to enhance compliance and decrease the risk continuing astemizole. (4) The drug should be avoid­ associated with the use of astemizole. ed in patients with significant liver disease. When (Key words: Astemizole, antihistamine, astemizole is used as directed, with the preceding pre­ adverse effects, toxicity, quality assurance) cautions kept in mind, it is considered a safe and effective nonsedating antihistamine. Astemizole was first introduced into the US mar­ The intent of our study was to evaluate the com­ ket in 1988. It was the second of a new generation pliance of the "new warnings" in a large multispe­ of antihistamines, the nonsedating antihistamines. cialty group practice. Frequently, a particular med­ The first of this class was terfenadine. More recent­ ication is selected for a formulary based on price ly, loratadine has become available. The safety pro­ without consideration of the drug's adverse profile. file of loratadine appears to be improved over the In our group practice, the only nonsedating anti­ other nonsedating antihistamines, but is more expen­ histamine available was astemizole. If the drug was sive. The nonsedating antihistamines are preferred frequently misused, it would suggest that the cost­ over the sedating antihistamines in that their adverse savings may be outweighed by the risks. In this effect profile is more acceptable to patients. case, the use of alternate agents may avoid unnec­ essary medicolegal concerns. From the University of Iowa, Iowa City (Dr Craig), and Charleston Naval Hospital (Dr Greenwald, Ms Kauffmann, Methods Ms Craig). Currently, Dr Craig is an associate professor, Divi­ sion of Pulmonary and Critical Care, Section of Allergy, Depart­ Outpatient records were reviewed monthly on all patients ment of Medicine, Pennsylvania State University, Hershey with new prescriptions for astemizole without regard Medical Center, Hershey, Pa. to the patient's age, sex, or previous health. The infor­ The views and opinions expressed in this article are those of mation was collected during the year 1993. The review­ the authors. They are not the official position of the Charleston ers included four physicians from the Pharmacy and Naval Hospital or the United States government. Therapeutics Committee. The group practice that was Correspondence to Timothy J. Craig, DO, Division of Pul­ monary and Critical Care, Section of Allergy, Department ofMed­ reviewed contained approximately 70 physicians of the icine, Pennsylvania State University, Hershey Medical Cen­ surgical, medical, family practice, and pediatric spe­ ter, 500 University Dr, Suite 5860, PO Box 850, Hershey, PA cialties. The physicians reviewed the records by fol­ 17033. lowing a checklist, noting the following for each record:

352 • JAOA • Vol 96 • No 6 • ,Tune 1996 Clinical practice • Craig et a1 Table Data on Use of Astemizole in a Large Group Practice*

No. of Proper Proper Warnings Compli- Steroids Study period charts indication, % dose, % present, % cations, % used, %

First 6 months 194 89 96 2 1 37

.~ Second 6 months 58 89 100 6 0 48 - Total 252 89 98 3 0 39 - *Percentages rounded to whole numbers

o indication, of the 252 patients had adverse effects, which includ­ o dosage, ed an arrhythmia (type undocumented) and an o concurrent medications, and episode of syncope. Nasal steroids were used in 39% o medical illnesses, including hepatic or cardiac dis- ofthe patients. The effectiveness of astemizole was ease, presyncope, and syncope. not assessed. They recorded adverse effects to astemizole. All patients who received astemizole were older than 12 years. Fur­ Discussion ther demographics were not assessed. Despite a drug warning by the FDA and the man­ An electrocardiogram (ECG) was not required to be ufacturer, astemizole continues to be used in inap­ available or obtained before patients started taking propriate doses and in patients with relative con­ astemizole. If an ECG was in the record, it was evalu­ traindications. The concern is that physician education ated for a prolonged QT interval, arrhythmias, or con­ duction abnormalities, all of which may complicate ther­ may not be up to date and that quality assurance apy with astemizole. programs may be necessary to improve compliance Feedback was given monthly to physicians prescribing to the "standard of care." In large group practices, the drug inappropriately in order to enhance compli­ proper use of medications is most easily accom­ ance. This interaction was not punitive, but simply a plished by drug utilization reviews. Drugs with brief reeducation of the proper use of astemizole. The com­ high usage, high price, or potentially fatal adverse bined use of nasal steroids and astemizole was assessed. effects are the most appropriate to study. If selec­ This assessment was deemed important because of the tion of a medication is based on price and alternate increasing use of nasal steroids and the belief by some agents are available, drug reviews should be done that nasal steroids are the preferred mode of therapy for to ensure that savings are not outweighed by risk. allergic rhinitis. Additionally, the data collected should become avail­ The data were descriptive only; thus, statistics were able to individual physicians to enhance effective not indicated. drug usage. During this drug utilization review, feedback was provided to physicians prescribing Results the drug inappropriately. During the first 6 month's A total of 252 records were reviewed. The data are of the study, 4% of patients received excessive doses, compiled in the Table. The indication was appro­ whereas during the last 6 months, all doses were with­ priate in 89% of the cases. Inappropriate use includ­ in guidelines. This compliance was attributed to ed nonallergic rhinitis and single-dose therapy for the ongoing monthly feedback given to physicians acute allergic diseases, such as urticaria. The dose during the study. was appropriate in 98% of cases. During the first 6 Arrhythmias complicating therapy with astem­ months ofthe study, five (4%) of the patients received izole have been documented. The frequency of inappropriate doses. Four patients (1.6%) received arrhythmia, however, is extremely rare. Between a loading dose, while one patient (0.4%) received 1988 and 1992, there were 44 serious cardiovas­ twice the recommended dosage. All doses were cular events associated with astemizole. These appropriate during the second 6 months of the events included 23 cases of torsades de pointes, 10 study. cases of ventricular tachycardia, 9 cardiac arrests, Eight (3%) of the patients receiving prescriptions and 5 cardiovascular deaths.2 Almost all events for astemizole had relative contraindications: 3 were associated with overdose. Our data are con­ patients (1.2%) had prolonged QT intervals, 2 sistent with this experience, in that adverse effects patients (0.8%) were receiving erythromycin, 1 were infrequent and could not be directly attrib­ patient (0.4%) had a significant conduction abnor­ uted to astemizole. mality, and 2 patients (0.8%) had documented Both astemizole and terfenadine are metabolized arrhythmias. Complications were rare. Only 2 (0.8%) by cytochrome p450. Erythromycin 3 and ketocona-

Clinical practice • Craig et al JAOA· Vol 96 • No 6· June 1996 • 353 zole4 are metabolized by the same enzyme and inhib­ mias are rare with the use of astemizole.2 it the metabolism of astemizole, increasing serum The most common indication for astemizole was levels and the potential for toxicity. Because of the allergic rhinitis. Nasal congestion is often the most similarity of fluconazole, metronidazole, itraconazole, troublesome symptom for the patient with allergic and miconazole to ketoconazole, these drugs should rhinitis. Antihistamines provide little or no benefit also be avoided until data are available suggesting in decreasing nasal congestion. The regular use of nasal their combined safety. The macrolides trolean­ steroids decreases congestion and is effective in domycin and clarithromycin are metabolized sim­ greater than 90% of patients with allergic rhinitis. 5 ilarly to erythromycin and should be avoided when It is surprising that nasal steroids were used in only using astemizole. 39% of the study population. Prolongation of the QT interval has been docu­ Inappropriate indications for astemizole includ­ mented with pentamidine isethionate, cisapride, ed: (1) nonallergic rhinitis (for which antihistamines amiodarone, probucol, tricyclic antidepressants, have minimal effect; (2) as a single dose on an as­ phenothiazines, terfenadine, bepridil hydrochlo­ needed basis; and (3) for acute transient rashes. ride, and the antiarrhythmics quinidine, disopyra­ The last two indications are ineffective uses of mide phosphate, and procainamide hydrochloride. astemizole because of the drug's prolonged half-life Combining these medications with astemizole may and delayed effect.4 The long half-life also accounts increase the potential for prolongation of the QT for astemizole's prolonged effect in suppression of interval and torsades de pointes. Diuretics can cause skin tests. This inhibition may persist for weeks, hypomagnesemia and hypokalemia, both of which making astemizole a poor choice to start before skin predispose to prolonged QT interval and arrhythmias. testing.6 This may increase the risk for arrhythmias and adverse effects from astemizole. Our data suggest Comment that erythromycin still poses the greatest risk for drug Our study reconfirms that drug utilization reviews interactions. This drug interaction is assumed to are an effective means of tracking physician's pre­ be secondary to the frequent use of erythromycin and scribing practices and education of physicians. that both drugs are used in a similar patient pop­ Although risk exists with use of astemizole, if the ulation. drug is appropriately used, it appears to re safe Our study did not require that an ECG be avail­ and have little potential for adverse outcomes. 7 able before prescribing astemizole. If a record con­ tained an ECG, the ECG was reviewed for preexisting abnormalities that could predispose to adverse References effects from astemizole. Electrocardiographic abnor­ 1. Nightinggale SL: Warnings issued on nonsedating antihist­ amines terfenadine and astemizole. JAM.A 1992;268:705. malities that contraindicate the use of astemizole were 2. Kemp J : Antihistamines- is there anything safe to prescribe? infrequently found. The most common abnormali­ Ann Allerg 1993;69:276-279. ty was prolonged QT interval. An ECG does not 3. Goss JE, Ramo BW, Blake T: Torsades de pointes associated appear to be indicated before starting astemizole with astemizole therapy. A rch Intern Med 1993; 153: 2705. therapy, except inpatients who have a history of, 4. Botstein P: Is QT interval prolongation harmful? A regulatory or are high risk for, arrhythmias. perspective. Am J Cardiol 1993;72:50B-54B. Only two adverse effects occurred. One patient had 5. Naclerio R: Dr ug therapy, allergic rhinitis. N Engl J Med syncope, and another had an arrhythmia (unspec­ 1991;3 25:860-869. 6. Lantin JP, Hugnuenot C, Pecoud AR: Effect of the HI antag­ ified). Neither adverse effect could be directly attrib­ onist , astemizole, on the skin reaction induced by histamine, uted to astemizole but, because ofthe possibility, the codeine and allergens. Curr Ther Res 1990;47 :683-692. medication was discontinued. This experience coin­ 7. Simons FE: New HI receptor antagonists: Worth the price? cides with past studies demonstrating that arrhyth- Ann Allerg 1992; 163-165.

354 · JAOA • Vol 96 • No 6· June 1996 Clinical practice • Craig et al