Realistic Assessment of Drug-induced Adverse Events: A Double-edged Sword

Abstract

Listen

Translate article

Angiotensin-converting enzyme (ACE) inhibitors represent a class of first-line medications for hypertensive patients with coronary artery disease, heart failure, atrial fibrillation, and post-stroke. There is an established outcome benefit of ACE inhibitors in reduction of all-cause mortality (risk ratio [RR], 0.87; 95% confidence interval [CI], 0.81‐0.94); nonfatal myocardial infarction (RR, 0.83; confidence interval, 0.73‐0.94); and, with less robust evidence, stroke (RR, 0.93; CI, 0.84‐ 1.03). 1-3 Despite such impressive improvement in clinical outcomes, use of ACE inhibitors is associated with the plethora of side effects including renal impairment, hyperkalemia, angioedema, headache, and fatigue somewhat jeopardizing their chronic use and increasing associated withdrawal risks. 4 In addition, dry, persistent, nonproductive cough is the most common well-described class effect of ACE inhibitors. 5,6 The mechanism of such adverse association remains unresolved but likely involves the excessive accumulation of bradykinin 7 and substance P, 8 agents that are degraded by ACE. These mediators accumulate in the upper respiratory tract or lungs when the enzyme is inhibited and modulate local airway release of prostaglandins, the production of which may be stimulated by bradykinin. Despite obvious concerns, it is still unclear how common this complication is and what are the “real life” associated threats, and withdrawal risks. A timely, elegant, well-designed, and convincing study in almost 200,000 pooled patients from 125 studies published in this issue of The American Journal of Medicine is an important contribution to the field. 9 The article concludes that all ACE inhibitors cause double-digit rates of cough, suggesting a class effect, that is many times higher than the overoptimistic 1.3% risk outlined in the Physicians’ Desk Reference/drug label. Moreover, the documented withdrawal rate was 2.57%, that was 31-fold higher compared to the reported rate in the Physicians’ Desk Reference/drug label (0.1%). The alarming data yielded from the index study have few important practical implications. First, patients in randomized trials are better controlled than in routine clinical practice clearly suggesting even higher risks of cough and associated withdrawal rates in real life than reported in the index article. Second, it is unclear to what degree the hidden, unreported to physicians, lack of compliance impacts the ACE inhibitors use on top of coughassociated complications. Third, we should look well beyond antihypertensives, especially focusing on the drug classes with no early benefits, but annoying immediate even minor side effects. Antiplatelet agents, anticoagulants, and antithrombins should be of most concern. Without appropriate patient education, withdrawal from these medicines is probably even higher due to frequent bleeding episodes, and gastrointestinal discomfort way exceeding officially recognized risks. Fourth, there is and will be no easy solutions, since pharmaceutical companies are somewhat resisting adequate reporting of adverse reactions protecting the safety profile image of their products. Finally, The Food & Drug Administration should require collecting registries independent from the drug company’s funding, especially when phase 3 trials reveal some unexpected complication.