IMPH Research in Pharmacoepidemiology

Medications and the Risk of Sudden Cardiac Death 
The MIND-USA Study 
Opiod Analgesics and the Risk of Serious Infections in Seniors 
Opiod Selection and the Risk of Serious Infections in Older Adults 
Prescribed Opiod Safety in Children 
Preventing Gastroduodenal Bleeding in Oral Anticoagulant Users

Medications and the Risk of Sudden Cardiac Death

Prescribed medications can increase the risk of sudden cardiac death, one of the single most common causes of death in industrialized countries. Thus, an important strategy for prevention is identifying medications that increase the risk of sudden cardiac death and using this information to guide clinical practice. The prevailing approach to identification of high-risk medications has been study of electrophysiologic markers of arrhythmia risk and case reports of torsade de pointes. However, while these methods can identify extreme risks, they often are inconclusive. We have utilized the Tennessee Medicaid automated database to identify important, yet unexpected, increased risks of sudden cardiac death for several commonly prescribed medications, including oral erythromycin, cyclic antidepressants, and antipsychotics. We propose to utilize this methodology to study other medications taken by more than 11 million patients in the U.S. with a strong signal suggesting increased risk for sudden cardiac death, but inconclusive evidence: contemporary antidepressants, methadone, and concurrent use of antipsychotics with drugs likely to inhibit their metabolism. Electrophysiologic studies and case reports of torsade de pointes suggest four specific widely used antidepressants--fluoxetine, citalopram, escitalopram, and trazodone--may increase risk of sudden cardiac death. The m-opioid agonist methadone, now most commonly used for chronic pain, prolongs cardiac repolarization and can cause torsade de pointes, suggesting it may increase the risk of sudden cardiac death. Several medications commonly prescribed with antipsychotics, such as fluoxetine and ciprofloxacin, markedly inhibit antipsychotic metabolism, which may increase antipsychotic effective dose and thus the risk of sudden cardiac death. To better define the cardiac risks of these widely used drugs, we propose a series of controlled epidemiologic studies in Tennessee Medicaid with three specific aims: 1. Test the hypothesis that sudden cardiac death risk varies for individual antidepressants and that four drugs--fluoxetine, citalopram, escitalopram, and trazodone--increase risk. 2. Test the hypothesis that in patients treated for chronic pain, the risk of sudden cardiac death in methadone users is greater than that for comparable opioid analgesics. 3. Test the hypothesis that concurrent antipsychotic use with strong inhibitors of their metabolism increases the risk of sudden cardiac death relative to such use without metabolic inhibitors. The novel quantitative data these studies will provide are critical for optimal clinical practice, as they will enable safer drug choices, particularly for patients with high baseline cardiovascular risk. Read more.

Funding Source: NIH/NHLBI 
PI: Wayne Ray

The MIND-USA Study

The long-term objective of the proposed MIND-USA (Modifying the Impact of ICU-Induced Neurological Dysfunction-USA) Study is to define the role of antipsychotics in the management of delirium in vulnerable critically ill patients. We and others have shown that delirium is an independent predictor of more death, longer stay, higher cost, and long-term cognitive impairment often commensurate with moderate dementia. The rapidly expanding aging ICU population is especially vulnerable to develop delirium, with 7 of 10 medical and surgical ICU patients developing this organ dysfunction. Antipsychotics are the first-line pharmacological agents recommended to treat delirium, and over the past 30 years they gained widespread use in hospitalized patients globally prior to adequate testing of efficacy and safety for this indication. Haloperidol, the most commonly chosen antipsychotic, is used by over 80% of ICU doctors for delirium, while atypical antipsychotics are prescribed by 40%. Antipsychotics safety concerns include lethal cardiac arrhythmias, extrapyramidal symptoms, and the highly publicized increased mortality associated with their use in non-ICU geriatric populations. The overarching hypothesis is that administration of typical and atypical antipsychotics-haloperidol and ziprasidone, in this case-to critically ill patients with delirium will improve short- and long-term clinical outcomes. Aim 1 will determine whether haloperidol or ziprasidone will increase days alive without acute brain dysfunction (referred to as delirium/coma-free days or DCFDs) over a 14-day period compared with placebo and compared to one another. Aim 2 will determine whether haloperidol or ziprasidone will improve 30-day, 90-day, and 1-year survival compared with placebo and compared to one another. Aim 3 will determine whether haloperidol or ziprasidone will reduce ICU length of stay compared with placebo and compared to one another. Aim 4 will determine whether haloperidol or ziprasidone will reduce the incidence, severity, and/or duration of long-term neuropsychological dysfunction and improve quality of life at 90-day and 1-year follow-up compared with placebo and compared to one another. To address these Aims, we will conduct this multi-center, double blind, randomized, placebo-controlled investigation in 876 critically ill, delirious medical/surgical ICU patients who are (a) on mechanical ventilation or non-invasive positive pressure ventilation or (b) in shock on vasopressors. In each group (haloperidol, ziprasidone, and placebo), 292 patients will be enrolled and treated until delirium has resolved for 48 hours or to 14 days (whichever occurs first) and followed for 1 year. We will monitor many safety parameters such as cardiac dysrhythmias and extrapyramidal symptoms. This study will have adequate power to detect the effect of antipsychotics in 4 important subgroups including age >65 years, severity of illness (APACHE II > 25), severe sepsis at enrollment, and medical vs. surgical ICU patients, and a hypothesis generating analysis of patients with pre- existing cognitive impairment. Read more.

Funding Source: NIH/NIA 
PI: E. Wesley Ely

Opiod Analgesics and the Risk of Serious Infections in Seniors

Opioid analgesics are one of the most commonly prescribed medication classes in the United States. Although these medications have been available for medical use for decades, their safety and effectiveness remain unclear. Available data from clinical trials are limited to highly selected patients and short term efficacy outcomes. Of great concern, information on the safety profile of these medications is even more scant. Several lines of in vivo experimental evidence indicate that opioid analgesics have significant immunosuppressive properties and could render opioid users susceptible to serious, potentially life-threatening infections. The immunosuppressive effects of opioid analgesics could be particularly troublesome for the elderly, who are commonly affected by persistent pain and are already at increased risk for infections. Given concerns about the safety of alternate analgesics (e.g. NSAIDs and COX-2 inhibitors), and the increasing widespread use of opioids, the determination and quantification of this potential risk is of great public health interest. Furthermore, there are a number of opioid analgesics currently available, but not all are expected to have the same immunosuppressive properties. Identifying those opioids with the lowest propensity to facilitate serious infections will be crucial to inform the selection of analgesics for the elderly. We propose to conduct a retrospective study of elderly persons enrolled in TennCare (Tennessee Medicaid) with the following specific aims: 1) To test the hypothesis that use of opioid analgesics increases the risk of serious infections; and, 2) To test the hypothesis that there is variation of the risk of serious infections for individual opioids. Th proposed studies will use a self-controlled case-series design, in which cases serve as their own controls, cancelling out the effects of fixed measured and unmeasured confounders, allowing control of relevant time-varying covariates; and, thus, allowing an unbiased estimation of the relative risk. Read more.

Funding Source: NIH/NIA 
PI: Carlos G. Grijalva

Opiod Selection and the Risk of Serious Infections in Older Adults

There is substantial evidence that opioid analgesic use impairs immune system responses and there has been a long-standing concern that these effects increase the risk of serious, potentially life-threatening infections. In vivo studies in animal models and human subjects have demonstrated that opioids affect surrogate markers of immune functions that are crucial for prevention of serious infections, such as white cell migration and phagocytosis. Although studies have demonstrated significant dose-dependent suppression of immunological functions in animal models and humans as well as a dose-dependent increase in the risk of serious infections in animal models, the clinical relevance of these findings in humans remains unclear. These concerns are particularly relevant for older adults, who are commonly affected by pain and are at increased risk for infections. A number of opioid analgesics are currently available, but not all have the same immunosuppressive properties. Studies in animal models suggest that, taking the chemical structure of morphine as reference, opioids with hydroxyl groups at both C3 and C6 (e.g. morphine), have the strongest immunosuppressive effects, whereas modification at C3 alone (e.g. codeine) reduces immunosuppression, and substitution of a carbonyl group at C6 (e.g. hydromorphone) eliminates the immunosuppressive effects. Identifying those opioids that are the least likely to increase the risk of serious infections will be crucial to inform the selection of analgesics for vulnerable older adults. Furthermore, the immunosuppressive effects of opioids are dose-dependent and for several opioids, in vivo data suggest that concurrent use of opioids and other commonly used medications that inhibit opioid metabolism could markedly increase the serum concentration of opioids. We propose to conduct a series of studies of older adults with the following specific aims: 1) Test the hypothesis that the risk of serious infections in new users of codeine (which is metabolized to morphine) is greater than in new users of other opioids with comparable analgesic properties; 2) Test the hypothesis that the risk of serious infections in new users of morphine is greater than in new users of other opioids with comparable analgesic properties; and, 3) Test the hypothesis that concurrent use of oxycodone or methadone and strong inhibitors of their metabolism increases the risk of serious infections relative to such use without metabolic inhibitors. The proposed studies will use a retrospective cohort study design and data from Tennessee Medicaid, to compare the incidence of serious infections associated with the use of selected opioids while controlling for the effect of relevant baseline and time-varying covariates. Our research team has the combination of experience and expertise needed to successfully complete the proposed projects. The proposed studies are designed to have a high impact on the field of pain therapeutics, advance our understanding of the effects of opioid analgesics on the risk of serious infections and inform the clinical care of older adults. Read more.

Funding Source: NIH/NIA 
PI: Carlos G. Grijalva

Prescribed Opiod Safety in Children

The epidemic of toxicity related to prescription opioid use among U.S. adults is thought to be related to drug abuse. Thus, the public health response has focused on programs to reduce non-medical uses. However, little attention has been given to the possibility of a parallel epidemic among children, particularly vulnerable to opioid toxicit. Preliminary Tennessee Medicaid data suggest this epidemic now possibly affects children: the proportion of children 2-17 years old prescribed opioid analgesics increased from 5% in 1996 to 11% in 2007, with a comparable trend in medical care possibly consistent with opioid toxicity. What is the appropriate public health response to this potential threat to the safety of children? Because children's prescriptions must be filled by a parent or guardian, non-medical use is less plausible, particularly for young children. For this reason, pediatric opioid toxicity is most likey to be an unintended consequence of therapeutic use, which would need to be considered as part of prescription risk-benefit evaluation. Thus, there is an urgent need to quantify the incidence of opioid toxicity in children to inform pediatric practitioners' decision-making. Furthermore, opioid prescribing practices associated with elevated risk need to be identified so these can be avoided when possible. Potentially hazardous practices include higher doses, high-potency opioids such as oxycodone, and concurrent central nervous system depressants. To address this unmet public health need, we will conduct a large retrospective cohort study in an estimated 500,000 Tennessee Medicaid children 2-17 years of age with 1,000,000 filled opioid prescriptions and 1400 confirmed cases of opioid toxicity. There are two specific aims: Aim 1. Quantify the incidence of toxicity related to prescription opioid use according to a) the child's age, b) toxicity severity, and c) whether or not the toxicity was related to therapeutic use. Aim 2 Test the hypothesis that opioid toxicity risk increases with: a) increased opioid dose, b) high-potency opioids, and c) concurrent use of other CNS-depressant drugs. These data on the risks of opioid medications increasingly used by children will provide a sound basis for altering pediatric practice to address a novel threat to the safety of this vulnerable population. Read more.

Funding Source: NIH/NICHD
Pi: Wayne Ray

Preventing Gastroduodenal Bleeding in Oral Anticoagulant Users

Warfarin is a mainstay of cardiovascular pharmacotherapy and one of the most frequently prescribed medications in the U.S. However, major bleeding is a frequent and potentially devastating complication of its use, affecting approximately 3% of patients annually. Serious warfarin-related bleeding most commonly occurs in the gastrointestinal (GI) tract, accounting for about 1/3 of major bleeds. This suggests proton-pump inhibitors (PPIs) might improve warfarin safety. This hypothesis is supported by our understanding of the pathophysiology of warfarin-related bleeding as well as by studies of other pro-hemorrhagic drugs. A benefit for PPI cotherapy would be greatest for the many warfarin users with other risk factors for GI bleeding. However, the hypothesis that PPIs prevent serious warfarin-related GI bleeding, although plausible, has not been tested. The available evidence is largely indirect and cannot, by itself, support recommending PPI cotherapy to warfarin patients. Indeed, contemporary anticoagulant guidelines do not mention PPIs. Histamine-2 receptor antagonists (H2RAs) also might be considered to prevent serious warfarin-related GI bleeding, although they may be less effective than PPIs. Novel oral anticoagulants, more convenient to use than warfarin, will account for a growing proportion of oral anticoagulant use. However, these also cause serious GI bleeding; some have risk greater than that of warfarin. Thus, PPIs may have the potential to markedly improve the safety of novel anticoagulants. We will conduct a large historical cohort study in Tennessee Medicaid and the national 5% Medicare sample assessing whether concurrent PPI/H2RA in oral anticoagulant users reduces risk of serious GI bleeding, critical data to improve the safety of these widely used medications. We will test two hypotheses: Aim 1: In Tennessee Medicaid warfarin users, concurrent PPI or H2RA use decreases risk of hospitalizations for gastroduodenal bleeding. Aim 2: In novel oral anticoagulant users in the Medicare sample, concurrent PPI or H2RA use decreases risk of hospitalizations for gastroduodenal bleeding. For each of these aims, we also will test whether the protective effect is greater for PPIs than for H2RAs as well as whether the absolute benefit varies according to number of GI bleeding risk factors. Read more.

Funding Source: NIH/NHLBI
PI: Wayne Ray