The drugs discussed below include nonsteroidal anti-inflammatory drugs (NSAIDs), general and local anesthetics, antiarrhythmics (amiodarone, [3-blockers, calcium channel antagonists, and doxazosin), and antibiotics. Others include steroids, by virtue of their mineralocorticoid properties (but also perivascular fibrosis of the systemic and coronary arterioles, and myocyte necrosis due to potassium depletion, followed by reparative fibrosis) and sym-pathomimetics, by virtue of an increase in afterload (cyclosporine and ketoconazole both act through this mechanism and may induce sharp pressure increases at high doses).
Nonsteroidal anti-inflammatory drugs (NSAIDs)
NSAIDs induce or exacerbate heart failure by promoting fluid retention. Their major mechanism of action is inhibition of prostaglandin synthesis, via inhibition of cyclooxygenase. The same mechanism accounts for NSAID interference in the effects of diuretics and angiotensin-converting enzyme (ACE) inhibitors. In healthy subjects, prostaglandins play a secondary role in renal homeostasis, and do not usually have significant effects on renal hemodynamics. In heart failure, on the other hand, they play a major role in cardiovascular and renal homeostasis: they cause vasodilatation of the afferent arteriole, counter systemic vasoconstriction by angiotensin, and promote salt and water elimination. In such patients, therefore, NSAIDs can unbalance circulatory homeostasis, especially at renal level, blunt the response to diuretics, and destabilize heart failure.
Two cyclooxygenase isoforms, COX-1 and COX-2, have been identified. COX-2 is mainly involved in all stages of the inflammatory response, while COX-1 is responsible for the synthesis of prostaglandins E2 (PGE,) and I, (PGI,) in the kidney and stomach. To prevent adverse renal effects, selective NSAIDs (sulin-dac, nabumetone, and meloxicam) were introduced; their renal and cardiovascular effects have not yet been thoroughly studied.
Low-dose aspirin can be useful in heart failure. In the presence of prostaglandin upregulation, thromboxane can be produced in greater amounts than prostacyclins. Besides favoring platelet activation and aggregation, thromboxane A2 directly induces vasoconstriction. Selective inhibition of thromboxane A, could therefore be useful. As for the problem of aspirin interference with ACE inhibitors, this may be proven clinically: in the Acute Infarction Ramipril Efficacy (AIRE) Study, Cooperative New Scandinavian ENalapril SUrvival Study II (CONSENSUS II), and Studies Of Left Ventricular Dysfunction (SOLVD), enalapril had unequal benefits depending on whether it was given alone or in combination with aspirin. ACE inhibitors antagonize kininase II, thus interfering with the degradation of bradykinins, which are powerful vasodilators with a potentiating effect on prostaglandin production. Aspirin, on the other hand, inhibits prostaglandin synthesis by blocking cyclooxygenase. The current literature on interaction between the two drugs is divided. Explanations include:
Aspirin is an ACE inhibitor antagonist. At high doses it decreases vasodilatation by enalapril in systemic but not in pulmonary vessels. A mechanism independent of prostaglandin synthesis has been hypothesized to explain pulmonary vasodilatation by ACE inhibitors; Aspirin and enalapril have similar mechanisms of action, but their effects do not appear additive; their combined effect is not greater than that of the individual drugs alone. The drugs may interact differently at different levels, with partial agonism and antagonism depending on their relative concentrations and on the balance between vasoconstrictor and vasodilator prostaglandins present at that given moment. Even negative interference between the two remains uncertain, as shown by the similar benefit observed in the heart failure subgroup of the Bezafibrate Infarction Prevention Trial, whether treated with aspirin or not.
Anesthetic agents General anesthetics
Cardiovascular homeostasis during general anesthesia depends on a variety of factors, including patient characteristics, such as age and concomitant disease, intravenous fluid overload, the surgical procedure, and the choice of anesthetic drugs. It is difficult to assign a specific cause to the onset or worsening of heart failure during surgery under general anesthesia. Many drugs used during anesthesia have negative cardiac effects, such as depressing contractility or inducing arrhythmia. Anesthetic agents may sensitize the heart to the arrhythmo-genic effects of endogenous catecholamines or (3-recep-tor agonists. The volatile anesthetics halothane and enflurane have a modest negative inotropic effect, which becomes significant in patients with preexisting myocardial dysfunction. Halothane appears able to block some calcium channels, thus interfering with dihydropyridine binding sites and altering calcium homeostasis. It may also modify the contractile protein response to calcium activation. Isoflurane and desflu-rane appear less cardiotoxic.
Injectable barbiturates such as thiopental and metho-hexital can depress myocardial contractility. This negative inotropic effect can become clinically significant in frailer patients, such as the elderly and those with left ventricular dysfunction. Propofol is an induction agent that can cause bradycardia, but also hypotension due to potent peripheral vasodilatation, sympathetic inhibition, and reduced myocardial contractility. Heart failure has been observed in children sedated with propofol for long periods; the effect is less frequent in adults.
Lidocaine and mepivacaine rarely have undesirable cardiovascular effects. Cocaine and procainamide, on the other hand, can have major negative effects on the heart. Acute responses to cocaine include decreases in myocardial contractility and coronary diameter, increases in heart rate and blood pressure, and cardiac electrical abnormalities (tachy-bradyarrhythmias, sudden death, ischemia, and myocardial infarction). Chronic cocaine abuse is associated with progressive and accelerated atherosclerosis and severe cardiomyopathy. Cardiotoxicity is due to the pharmacologic effects of cocaine on excitable tissues. These include reversible inhibition of sodium channels and blockade of the transient rise in sodium conductance (local anesthetic effect, membrane stabilizer). Cocaine slows myocyte depolarization, lowers the action potential, slows conduction velocity, and prolongs the absolute refractory period. In addition, it inhibits norepinephrine (and dopamine) reuptake in presynaptic neurons of the central and peripheral nervous systems, thus increasing sympathetic output and catecholamine levels (sympathomimetic effect). It has recently been shown to stimulate central sympathetic activity, causing vasoconstriction and tachycardia. Cocaine stimulates eland P-adrenoceptors, causing an increase in cyclic adenosine monophosphate (cAMP). As a second messenger, cAMP induces an increase in cytosolic calcium, generating a sustained action potential and premature ventricular beats. Conduction is altered and reentry circuits are induced.
In summary, cocaine increases the heart’s requirement for oxygen (by increasing sympathetic activity) while reducing its supply (by coronary vasoconstriction). It may therefore induce myocardial ischemia and/or infarction. Increased platelet aggre-gability has also observed; this compounds the ischemia. Although these mechanisms have been observed in patients with coronary artery disease, it is probable that they also occur in typically young cocaine addicts with normal coronary arteries. Evidence of direct myocardial toxicity is the extensive myofibril depletion and marked sarcoplasmic vacuolization observed in cocaine addicts. These changes are unlikely to have been caused by ischemia because the amount of necrosis is small and focal (even totally absent). Single cells are also found surrounded by inflammatory infiltrates similar to those observed on doxorubicin therapy; coag-ulative myocytolysis is also present. There is evidence of mitochondrial cardiotoxicity, due in part to the acidotic effect. Cocaine directly inhibits the respiratory chain, especially in the nicotinamide adenine dinucleotide (NADH) dehydrogenase region. Abnormal calcium homeostasis is due not only to a cocaine-mediated excess of catecholamines, but also to a direct effect of cocaine on sarcolemmal integrity. These changes can also lead directly to myocyte necrosis, vascular contraction, development of myocyte contraction bands, catecholamine hypersensitivity, and platelet hyperaggregability.
Cocaine addicts also have evidence of hemodynamic overload (elevated atrial natriuretic hormones, ventricular hypertrophy) and fibrosis (accumulation of collagen types I and II). Myocarditis, endocarditis, aortic rupture, and hypothyroidism have also been observed.
Genetic P-adrenoceptor differences have been described in rats and could be responsible for differences in susceptibility to cocaine cardiotoxicity.
Antiarrhythmics exert differing degrees of negative inotropic effect on ventricular function. In particular, they can destabilize patients with preexisting left ventricular dysfunction. The mechanism involves changes in the intracellular calcium concentration. Antiarrhythmics also have other adverse cardiovascular effects, such as peripheral vasodilatation and arrhythmia (prolonged action potential with an increase in the time available for calcium entry). Randomized, double-blind, placebo-controlled trials have shown a higher risk of developing heart failure among patients (mostly with atrial fibrillation or previous myocardial infarction) receiving class I antiarrhythmics; quinidine, procainamide, disopyramide, mexiletine, tocainamide, flecainide, propafenone, and moricizine. Despite their individual differences, almost all antiarrhythmics must be viewed as potentially dangerous in patients with chronic heart failure. They should therefore be avoided, unless there are clear indications to the contrary, in which case their use must be carefully monitored. This is particularly important in that drug clearance is decreased in heart failure, with an attendant increase in the risk of accumulation and toxicity.
Amiodarone, a noncompetitive P-blocker and weak calcium channel antagonist, is generally well-tolerated in heart failure, as shown in several randomized trials. It is less cardiotoxic than other antiarrhythmics. Outside a few case reports, there are no negative hemodynamic effects. The only risk of destabilization may lie in excessive bradycardia (2.4% of cases in one meta-analysis).
P-Blockers have become a recommended heart failure treatment. Nevertheless, their negative inotro-pism dictates gradual and cautious low-dose initiation followed by closely monitored increments. The bradycardia and hypotension induced by P-blockers can also be destabilizing. Guidelines define a heart rate ^50 beats per minute and a systolic pressure =S90 mm Hg as indicating discontinuation of dose escalation. Trials have reported destabilization rates of 1% to 5% in advanced heart failure, sotalol prolongs the action potential and may thus be less negatively inotropic than other P-blockers. It induces torsades de pointes in 2% to 4% of heart failure patients. The D-isomer (o-sotalol) has no P-block-ing action, but in the Survival With ORal D-sotalol (SWORD) trial it increased mortality in infarcted patients with weak left ventricular function.
Calcium channel antagonists
The mechanisms involved in the adverse effect of calcium channel antagonists in patients with ventricular dysfunction include negative inotropism, caused by blockade of cellular transmembrane transport, and activation of the sympathetic nervous system and renin-angiotensin-aldosterone neuroendocrine system. Marked hemodynamic and clinical deterioration can result in patients with chronic heart failure. In the Multicenter Diltiazem Postinfarction Trial (MDPIT), mortality and emergent heart failure were significantly higher in patients with a decreased ejection fraction treated with diltiazem. Two large trials were recently
Some antibiotics can be cardiotoxic if given at excessive doses and/or to patients with preexisting cardiac abnormalities. The main mechanisms and clinical manifestations are listed.
Aminoglycosides (streptomycin, gentamicin, kanamycin, amikacin, etc) are associated with direct cardiodepres-sion. Gentamicin inhibits the slow calcium channels in cardiac muscle. It has therefore been suggested that aminoglycosides inhibit the uptake or binding of calcium to specific sites on the sarcolemma, thus decreasing the amount bound to the membrane and available for migration during sarcolemmal depolarization.
Other antibiotics can cause cardiac dysfunction via as yet unelucidated mechanisms. Erythromycin may induce arrhythmias (torsades de pointes). Tetracyclines and chloramphenicol appear to depress contractility, interacting directly with the myocyte or indirectly decreasing plasma or extracellular calcium levels. Some antifungals (amphotericin B) depress myocardial contractility by blocking the slow calcium channels and inhibiting sodium flow conducted with dihydropyridines in heart failure: in the Vasodilator Heart Failure Trial III (V-HeFT III), the effect of nifedipine was neutral; in the Prospective Randomized Amlodlpine Survival Evaluation (PRAISE) trial, amlodipine was neutral in terms of clinical events and overall mortality, but potentially beneficial in the subgroup with nonischemic heart failure.
The Antihypertensive and Lipid-Lowering treatment to prevent Heart Attack Trial (ALLHAT) showed that, compared with chlorthalidone controls, hypertensive patients treated with the 01-blocker doxazosin had twice the risk of serious cardiac events; in particular, their risk of developing heart failure exceeded 100% (RR 2.04; 1.79-2.32). The reason is not yet clear. The difference in hypotensive effect, which was greater on doxazosin, was too small to explain the different incidence of heart failure. Until, and unless, evidence to the contrary becomes available, doxazosin should therefore not be used in heart failure, despite its symptomatic benefit in prostatic hypertrophy.
drug; destabilization factor; myocardial dysfunction; NSAID; anesthetic agent; antiarrhythmic drug; calcium channel blocker; antibiotic