CERTs.About.Annual Reports.Year 4.CERTs Progress.Advancing.Knowledge

Medical therapies are designed with one goal in mind: to improve health. The underlying challenge is that all therapies invariably involve some degree of risk. The CERTs are providing information to help health care providers, policymakers, and patients understand the risks and benefits of medical therapies in order to maximize those benefits and minimize the risks in their use.

UNDERSTANDING THE RISKS OF ADVERSE REACTIONS TO FREQUENTLY PRESCRIBED MEDICATIONS

Prescription drugs play an important and beneficial role in American medicine. But in some cases, they produce unwanted side effects, ranging from relatively mild allergic reactions to sudden death. The CERTs have undertaken a number of studies to uncover factors that increase risks associated with some frequently prescribed medications.

Erythromycin, a commonly prescribed antibiotic, has been reported to cause episodes of serious arrhythmia or sudden cardiac death in both oral and intravenous forms. Previous studies of the association of erythromycin and arrhythmia have focused on the intravenous use of erythromycin, and there has been a perception that oral use of erythromycin is not associated with arrhythmias.

For the study, CERTs investigators examined computerized death certificate and hospital discharge records of patients enrolled in the Tennessee Medicaid program, TennCare, to identify potential cases of sudden cardiac death. They identified 4,404 patients for whom these records suggested occurrence of sudden cardiac death. They found that the association between the use of erythromycin and the risk of sudden death from cardiac arrest applies to the oral use of erythromycin as well as intravenous use.

Even more alarming was the elevated risk associated with using erythromycin in combination with other common medications that inhibit a specific type of drug enzyme known as cytochrome P450 3A (CYP3A). The risk of death for people using both erythromycin and CYP3A inhibitors was five times as high as the risk for those who were not using any of the antibiotic drugs from the study or any CYP3A inhibitors.

Among CYP3A inhibitors are the antibiotic clarithromycin (Biaxin®) and certain anti-fungal drugs used to treat toenail fungus and yeast infections, such as fluconazole (Difulcan®), ketoconazole (Nizoral®), and itraconazole (Sporanox®). CYP3A inhibitors slow down erythromycin's breakdown, thereby increasing its concentration in the bloodstream. The higher levels of erythromycin can trigger an abnormal, potentially fatal heart rhythm.

With safer antibiotics, such as amoxicillin, readily available, researchers conclude that practitioners should avoid prescribing erythromycin in conjunction with CYP3A inhibitors.

Penicillin is a commonly used antibiotic that provokes an allergic response in some people. Approximately 10 to 15 percent of adults report having had an allergic reaction to penicillin, but little was known about how often such patients are re-exposed to penicillin or the risks caused by such re-exposure.

Most information about allergic-like reactions to penicillin has come from hospitalized patients, a population that represents only a small portion of the people who use penicillin. Because it might sometimes be necessary to re-administer penicillin to someone who has previously had an allergic-like reaction, it is important to gather more information about how often such re-exposure to penicillin occurs and the risks of such exposure.

CERTs investigators examined a large group of patients who were not hospitalized but had received prescriptions for penicillin more than 60 days apart. They identified patients who had symptoms similar to those of an allergic reaction on the day of or within 30 days after a prescription. Researchers found that an alarming number of patients were being prescribed penicillin a second time, despite a prior allergic-like event. The data indicated that almost half of those who had experienced such a response to penicillin were later given a second prescription, and in their subsequent use of penicillin, these patients faced a marked increase in the risk of an allergic-like reaction. The absolute risk of an allergic reaction was 0.18 percent following the first prescription, but it increased to 1.89 percent following the second prescription. These data emphasize the necessity for careful prescribing, and specifically the importance of avoiding prescribing penicillin to patients who suffered a previous reaction.

Antibiotics are a crucial weapon in the arsenal against infectious diseases, and yet widespread use undermines their effectiveness. Now in our fifth year, the CERTs have conducted numerous studies and interventions on the subject of bacterial resistance in an attempt to lower antibiotic use in situations where antibiotics are needlessly prescribed. Bacterial resistance to these medicines is a persistent problem in our country, and the CERTs are focusing on several different angles of this dilemma.

In one study, researchers looked at the rates of resistance to penicillin in Streptococcus pneumoniae (S. pneumoniae), which is the most common bacterial pathogen and among the most virulent. Instances in some communities of resistance to penicillin or other drugs recommended by national treatment guidelines raise the possibility that these guidelines may no longer be applicable nationwide. When choosing between prescribing older antibiotics or newer medicines, providers may need to rely less on national guidelines than on up-to-date information about local rates of drug resistance.

A second CERTs study examined data from acute-care hospitals in a five-county area of eastern Pennsylvania to determine whether a specific hospital's rates of antibiotic resistance could be a useful indicator of rates of drug resistance in the surrounding community. The results suggest that the information about drug resistance within one hospital poorly reflects the underlying rate of drug resistance in the community. The study demonstrates the danger that comes with the inability to predict or anticipate local drug resistance rates and the importance of additional research to identify reliable indicators of these resistance rates.

In another study, the CERTs examined long-term trends in resistance to fluoroquinolone (fQ) antibiotics among Enterobacteriaceae, such as E. coli and Salmonella. These are organisms that cause many inpatient and outpatient infections, including urinary tract infections and bloodstream infections. The data indicated increasing resistance in samples of these bacteria obtained from both inpatients and outpatients. However, the patterns of resistance differed across various Enterobacteriaceae and among clinical settings. The implications are that FQ antibiotics are used frequently to treat these common infections, but that if resistance continues to increase, the usefulness of these antibiotics will become severely limited.

Oftentimes, adults and children respond differently to medications and treatment regimens. Children have different metabolic rates, their bodies change rapidly, and their ability to understand and express information varies widely. The CERTs study how children respond to common medical therapies to enable parents and providers to make more informed decisions about treatment options.

The use of antipsychotic medications in children ages 18 and younger for indications other than psychosis or Tourette Syndrome, such as bedwetting, is controversial. But with the development of newer medications that avoid some of the dangerous side effects of traditional antipsychotic drugs, health care providers may be prescribing these medications more frequently, despite their potential side effects, such as weight gain, diabetes, galactorrhea, and adverse cardiovascular effects, such as life-threatening arrhythmias.

CERTs researchers examined data from children enrolled in TennCare and found that the proportion of these children using antipsychotic medications nearly doubled from 1996 to 2001, with a substantial increase in the use of antipsychotics for attention deficit hyperactivity disorder (ADHD) and affective disorders, such as depression. Given the risks of serious side effects, the CERTs study demonstrated a clear need for clinical studies to determine whether the benefits of these medications outweigh the risks in children.

The "recommended prescribed dose" of a drug may work very well for many patients but not for others because of underlying differences from patient to patient, including differences in the way drugs are absorbed and metabolized. Less is known about these individual differences in children than in adults. Children may also metabolize drugs differently as they grow and develop. These differences may mean that the "recommended" dose may not have the treatment effect a sick child needs. HIV provides an important example. More than 20,000 children and teenagers in the United States may be infected with human immunodeficiency virus (HIV). How they respond to given doses of anti-HIV drugs such as protease inhibitors may differ from how adults respond.

Therapeutic drug monitoring (TDM) is a way of measuring individual patients' response to drugs and tailoring drug doses to individual patients' needs. The CERTs have assessed the concentrations of protease inhibitors in the blood of children infected with HIV. Concentrations that are too high can be toxic, and concentrations that are too low can cause the virus to become resistant to the drug.

The group used a separation technique (high-pressure liquid chromatography) to develop a test for measuring four of the most commonly used protease inhibitors in blood samples. In a pilot study, they found that in a group of 15 pediatric patients taking at least one of the medications, four children, or 27 percent, had no detectable protease inhibitors in their blood plasma. One child had a suspiciously high concentration, suggesting that the patient had taken an incorrect amount of medication, or that the patient may have had abnormal metabolism of the drug.

The CERTs are now engaged in a larger study to further explore the use of TDM to tailor the treatment of HIV-infected children in order to get the best effect for each child.

In a collaborative effort with an AHRQ Evidence-based Practice Center, the Ambulatory Pediatric Association (APA) published guidelines on three clinical topics: acute otitis media; otitis media with effusion; and management of the infant with hyperbilirubinemia. The new guidelines are based on evidence developed by AHRQ.

After cataloging the resources identified by the experts who drafted the guidelines, a CERTs-APA team will work with practitioners to devise a set of tools and good practices to promote and disseminate the guidelines as widely as possible.

About CERTs

Annual Report Year 5

CERTs Progress

Advancing Knowledge