Non-selective nonsteroidal anti-inflammatory drugs and cardiovascular events: is aldosterone the silent partner in crime?

Abstract

Non-selective nonsteroidal anti-inflammatory drugs (NSAIDs) have long been known to cause gastrointestinal and renal toxicity. More recently, adverse cardiovascular effects have been associated with the selective COX-2 inhibitors. However, current studies that show an increased cardiovascular risk with non-selective NSAIDs raise the question of the exclusive contribution of COX-2 to this type of toxicity. Aldosterone, a key cardiovascular hormone, can induce deleterious effects, such as myocardial fibrosis and vascular stiffening. Non-selective NSAIDs inhibit the metabolism of aldosterone in vitro by human renal tissue, predicating an increased plasma aldosterone concentration in vivo. The question remains whether inhibition of aldosterone metabolism by non-selective NSAIDs is a casual or causal factor in NSAID-induced cardiovascular toxicity.

Nonsteroidal anti-inflammatory drugs (NSAIDs) are among the most commonly used drugs world-wide for the treatment of a variety of clinical conditions characterized by pain, inflammation and fever. Recent experience with the selective COX-2 inhibitors, culminating in the withdrawal of rofecoxib, has expanded the spectrum of adverse effects beyond the gastrointestinal and renal systems to encompass the cardiovascular system; but is the increased incidence of cardiovascular events with COX-2 inhibitors relevant to the non-selective NSAIDs?

Murmurings that the adverse cardiovascular effects were restricted to COX-2 inhibitors soon abated with the publication of two studies linking traditional non-selective NSAIDs to an increased risk of cardiovascular events [1, 2]. Publication of these papers unleashed a flurry of criticism and created intense debate. Recently, Gislason et al.[3] reported a higher risk of death in patients taking either COX-2 inhibitors or non-selective NSAIDs after myocardial infarction, using data from the Danish National Patients Registry. The hazard ratio for all-cause death in those taking diclofenac (over 100 mg day⁻¹) was reported as 3.76. After adjustment for age, year of myocardial infarction, sex, social status, other medical conditions and concomitant medications, the authors concluded that those who were taking higher doses of COX-2 inhibitors or other non-selective NSAIDs were at a ‘strikingly higher’ risk of death than those who were not taking the drugs [3]. Using the UK General Practice Research Database, Rodriguez and Gonzalez-Perez [4] reported a small excess risk of nonfatal myocardial infarction with chronic exposure to non-selective NSAIDs. Again the relative risk was greatest with diclofenac, 1.38 (95% confidence interval 1.00, 1.90), while there was a small reduction with naproxen and an undetectable risk with ibuprofen. All three drugs are less than fivefold COX-2 selective {log [IC₈₀ ratio (COX-2/COX-1)]}, in contrast to rofecoxib (which is over 50-fold COX-2 selective), in the human whole blood assay system, with a rank order of COX-2 selectivity of diclofenac > ibuprofen > naproxen [5]. However, if the prothrombotic potential of COX-2 inhibitors is simply an imbalance in favour of thromboxane A₂, what is the explanation for the cardiovascular risk associated with non-selective (COX-1/COX-2) inhibitors?

NSAIDs promote sodium and water retention, and this has generally been explained by a reduction in prostaglandin-induced inhibition of both renal chloride reabsorption and the action of antidiuretic hormone. The same mechanism has been used to explain the reduction in effectiveness of antihypertensive drugs. Although the use of antihypertensive drugs has increased, blood pressure control remains suboptimal and the incidence of renal disease, stroke and heart failure has continued to rise in most countries. This suggests the existence of factors that limit the effectiveness of antihypertensive drugs and at the same time favour the onset and progression of target organ damage.

The renin–angiotensin–aldosterone system (RAAS) plays an integral role in blood pressure regulation and has long been a target of pharmacological intervention to control blood pressure (for example, with ACE inhibitors and angiotensin II receptor antagonists). However, the focus has centred on blocking angiotensin II, with the underlying assumption of a consequential reduction in the synthesis of aldosterone. Little, if any, attention has been placed on the second ‘A’ in RAAS, perhaps a reflection of constant references to the renin–angiotensin system. In the meantime, aldosterone has silently stalked the cardiovascular system. Although the classic actions of aldosterone relate to regulation of water and electrolytes via actions on renal epithelial cells, it is well established that aldosterone contributes to the progression of chronic heart failure and end-organ damage and promotes perivascular and interstitial myocardial fibrosis, endothelial dysfunction and vascular stiffening [6, 7].

Limited attention has been directed towards drugs that inhibit the actions of aldosterone, although an increasing body of evidence supports the view that aldosterone causes cardiovascular and renal injury through mineralocorticoid receptor-dependent mechanisms. Support for this view is evident from the RALES study [8], in which patients randomized to receive spironolactone in addition to standard therapy (an ACE inhibitor and a diuretic) had a 30% reduced risk of cardiac mortality and a 35% lower rate of hospitalization than those who received standard therapy plus placebo. Similarly, the EPHESUS trial [9] evaluated the selective aldosterone blocker eplerenone plus standard therapy on morbidity and survival in patients who developed heart failure soon after myocardial infarction; combined therapy that included aldosterone blockade resulted in a 21% reduction in the rate of sudden cardiac death and a 15% reduction in rate of hospitalization from heart failure. Undoubtedly, aldosterone is a key cardiovascular hormone. Could it be the silent partner that links non-selective NSAIDs with adverse cardiovascular events?

It has recently been reported that commonly prescribed NSAIDs (e.g. diclofenac, ketoprofen, mefenamic acid, naproxen and tiaprofenic acid), at clinically relevant concentrations, inhibit the glucuronidation of aldosterone by human kidney (and liver) microsomes [10]. Calculations based on these in vitro data suggest that selected NSAIDs may increase plasma and tissue (renal) aldosterone concentrations from 30 to 320% (unpublished data).

If cardiovascular risk is the outcome and non-selective NSAIDs are at the centre of the web, is the link NSAID-induced sodium retention, which fuels the profibrotic effects of aldosterone, exacerbating hypertension and heart failure in some individuals? Clearly, it is time to look beyond the gastrointestinal and renal effects of non-selective NSAIDs to ascertain if chronic NSAID administration increases the cardiovascular risk and to consider whether aldosterone is the silent partner in crime.