Valvular heart disease induced by drugs

INTRODUCTION — Drug-induced valvular heart disease refers to morphological and functional abnormalities of valve leaflets secondary to exposure to various medications. Ergot derivatives (ergotamine and methysergide) were the first drugs linked to valvulopathy. Subsequently, other drugs, such as fenfluramine, pergolide, and cabergoline, have been implicated in causing valvular abnormalities [1]. The use of 3,4-methylenedioxymethamphetamine (MDMA, also known as ecstasy) may also induce valve disease. The common pathogenetic link is increased serotonin activity, since many of the drugs and their metabolites are potent 5-hydroxytryptamine_2B (5-HT_2B) receptor agonists [2].

PATHOGENESIS — The echocardiographic and pathologic features of the valve disease seen in patients with drug-induced valvular disease are similar to those occurring in carcinoid heart valve disease [3]. (See "Carcinoid heart disease".)

The link between these two conditions appears to be serotonin, a neurotransmitter that is elevated in serum and plasma of most patients with carcinoid heart valve disease [4]. As noted above, fenfluramine and dexfenfluramine augment serotonergic activity, while phentermine may contribute to the fenfluramine valvulopathy by interfering with the pulmonary clearance of serotonin.

Serotonin is also thought to play a role in the valve disease associated with ergot derivatives ergotamine [5], methysergide [6], and the dopamine agonists pergolide and cabergoline [7-9]. These drugs are structurally similar to serotonin and are thought to produce valve disease by stimulation of serotonin (5-HT 2B) receptors [7,10,11]. (See 'Parkinson disease' below.)

Although serotonin appears to play a central role in drug-induced valve disease, selective serotonin reuptake inhibitors (fluoxetine, paroxetine, and sertraline), which are commonly used for the treatment of depression and other disorders, do not appear to be associated with valve disease [12] (see "Serotonin-norepinephrine reuptake inhibitors: Pharmacology, administration, and side effects"). This may be because activation of serotonin-2B receptors, which are located on both mitral and aortic valves, is required to induce valve disease. Support for this hypothesis comes from a study that found that drugs associated with valvular heart disease or metabolites of such drugs, including ergotamine and metabolites of fenfluramine, dexfenfluramine, and methysergide, had high affinity for and were partial to full agonists of the serotonin-2B receptor [10]. By contrast, drugs not associated with valvular disease had low affinity for this receptor (phentermine and fluoxetine) or were a potent antagonist (trazodone).

Serotonin stimulates fibroblast growth and fibrogenesis, which appears to underlie the valve disease seen in patients with carcinoid and anorectic drug and ergot derivative related heart valve disease. More direct evidence for a central role of serotonin comes from studies in animals:

●In a rat model in which daily serotonin injections were given for three months, valvular abnormalities (aortic and/or pulmonary regurgitation) developed in six of ten rats, with histopathologic examination showed carcinoid-like plaques similar to those seen in humans with carcinoid heart disease [13]. There were no valvular changes in control rats.

●Another mechanism of raising serotonin levels is to interfere with its metabolism. This process is mediated in part by the 5-hydroxytryptamine (5-HT) transporter, which is responsible for serotonin inactivation as it passes through the lungs. Both valvular hyperplasia and fibrosis were observed in a mouse model in which the 5-HT transporter gene was knocked out [14].

PREVENTION AND MANAGEMENT — Initial recommendations from the Centers for Disease Control and Prevention, US Food and Drug Administration, and National Institutes of Health were issued in 1997 for patients who were exposed to fenfluramine or dexfenfluramine for any length of time [15]. More definitive recommendations were included in the 2003 American College of Cardiology/American Heart Association/American Society of Echocardiography guidelines on clinical application of echocardiography [16]:

●Drugs linked to valvular heart disease such as fenfluramine and dexfenfluramine should be avoided. Caution is advised regarding the use of nonprescription anorectic drugs, herbal anorectics, and even diet supplements. The use of some ergot derivatives has also been shown to be associated with a valvulopathy.

●A cardiac history and physical examination, emphasizing cardiac auscultation, is indicated in all exposed patients. In the absence of abnormal findings, a repeat history and physical examination can be performed in six to eight months. The validity of this approach was illustrated in a study of 223 patients receiving dexfenfluramine for seven months; the absence of heart murmurs predicted the absence of echocardiographic valvular regurgitation in 93 percent of patients, while the presence of a murmur had a positive predictive value of only 19 percent [17].

●Echocardiography to assess valve morphology and regurgitation is indicated in patients with a history of use of any drug associated with valvular heart disease who have a heart murmur, symptoms of valvular heart disease, or a technically inadequate auscultatory examination.

●The optimal timing of follow-up echocardiography to determine the progression, regression, or stabilization of valvular lesions is not known. Routine repetition of echocardiography was not recommended in past users of anorectic drugs with normal studies or known trivial valvular abnormalities

ANORECTIC DRUGS — The consequences of obesity include a variety of adverse health outcomes as well as social stigma [18] (see "Overweight and obesity in adults: Health consequences"). As a result, aggressive strategies have been developed for weight reduction in individuals with obesity. One such strategy is the prescription of appetite suppressing or anorectic drugs, usually in combination with diet, exercise, and behavior modification. Some of these drugs have been implicated in the pathogenesis of a characteristic form of valvular heart disease [19,20].

Patients who developed valve disease have generally been treated with an off-label fenfluramine in combination with phentermine (called "fen-phen"). (See "Obesity in adults: Drug therapy".)

●Fenfluramine – Fenfluramine (Pondimin) is a sympathomimetic amine that promotes the release of serotonin and blocks its neuronal uptake. The ensuing increase in serotonergic activity has an anorectic effect. Fenfluramine was originally approved by the US Food and Drug Administration (FDA) in 1973 for single-drug, short term use as an appetite suppressant.

●Dexfenfluramine – Dexfenfluramine (Redux) is the dextro isomer of fenfluramine. Its activity is relatively selective to serotonergic pathways in the brain, but the appetite suppression is similar to that of fenfluramine. In 1996, the FDA approved dexfenfluramine as a single-drug appetite suppressant for longer term use in individuals with marked obesity.

●Benflourex is an amphetamine derivative related to fenfluramine and dexfenfluramine. This agent was withdrawn from the European markets in 2010 following several reports linking this agent to three-fold increase in hospitalization for valve regurgitation [21-23].

●Phentermine – Phentermine (Adipex, Fastin, Ionamin) is a noradrenergic central nervous system stimulant that was approved by the FDA as an appetite suppressant in 1959. When given alone, this agent has not been implicated in the pathogenesis of valvular heart disease.

Combinations of fenfluramine and phentermine (called fen/phen) have been used with the hope that weight reduction might be achieved with lower doses and fewer side effects (see "Obesity in adults: Drug therapy"). Most of the reported cases of valvular disease in patients taking anorectic drugs were in patients using this combination [19,20].

Approximately 18 million prescriptions were written for either fenfluramine or phentermine in 1996 [24]. In 1997 the FDA asked the manufacturers to withdraw fenfluramine and dexfenfluramine (but not phentermine) from the market. Assuming that the average prescription was written for one month and that the treatment course lasted three to twelve months, it is likely that several million Americans were exposed to these drugs.

Relation to valvular disease — In 1997, cardiac ultrasonographers at the MeritCare Medical Center in Fargo, North Dakota and clinicians at the Mayo Clinic in Rochester, Minnesota described valvular heart disease in 24 women (average age 44) who had been treated with the combination of fenfluramine and phentermine [19]. All 24 had cardiac symptoms and/or a heart murmur. Aortic or mitral regurgitation (or both) was present in all 24; one-half also had tricuspid involvement. The average dose of fenfluramine was 60 mg/day and the dose of phentermine was 30 mg/day. The mean duration of treatment with fenfluramine-phentermine was 11 months (range 1 to 28 months). Five patients required valve replacement, while less severe valvular regurgitation was present in the remaining patients. (See "Auscultation of cardiac murmurs in adults".)

An additional 28 cases were reported simultaneously by physicians from the FDA [20]. These patients, all women, exhibited a similar history of valvular heart disease in association with the use of a combination of fenfluramine and phentermine. As a result of these observations, the manufacturer withdrew fenfluramine and dexfenfluramine from the market (September 1997).

Subsequently, more than 100 additional provider-initiated case reports were received by the FDA. Valvulopathy was associated with the use of fenfluramine or dexfenfluramine, alone or in combination with phentermine. There were no cases with the use of phentermine alone.

Unfortunately, these initial descriptions of an association between valvulopathy and anorectic drugs do not represent epidemiologic or even case-control studies. Subsequent randomized and case-controlled studies provided additional information [25-27], but several issues remain unclear:

●The prevalence of regurgitant valvular lesions in patients with obesity not exposed to these drugs is not well-established. The patients in these reports did not routinely have echocardiography prior to drug therapy and it is therefore unclear how many may have had preexistent valvular abnormalities.

●The relative risks of fenfluramine and dexfenfluramine, alone or in combination with phentermine, as well as the effects of dose and duration of exposure are not well defined.

●The risk factors for valvulopathy have not been defined. Baseline body weight, concomitant medications, and other demographic factors must be assessed.

A nonrandomized study compared the echocardiographic findings of 257 patients using any anorectic drugs with 239 control patients [26]. Mild or greater aortic regurgitation or moderate or greater mitral regurgitation were present in 1.3 percent of controls, 12.8 percent of patients receiving dexfenfluramine (odds ratio 12.7), 22.8 percent of those taking the combination of dexfenfluramine and phentermine (odds ratio 24.5), and 25.2 percent of patients taking both fenfluramine and phentermine (odds ratio 26.3) (figure 1).

A much lower prevalence of valvulopathy has been reported by others. One nonrandomized study examined 223 patients exposed to dexfenfluramine for a mean of 6.9 months and 189 unexposed controls; at least mild aortic regurgitation or moderate mitral regurgitation was significantly more frequent in patients treated with dexfenfluramine (7.6 versus 2.1 percent for controls) [28]. The difference was primarily due to more frequent mild aortic regurgitation (6.3 versus 1.6 percent) (figure 2). Factors independently associated with any grade of aortic regurgitation were older age, diastolic blood pressure greater than 95 mmHg (figure 3), and a shorter time from drug discontinuation to echocardiogram, which suggests that dexfenfluramine-related valve regurgitation may regress after drug discontinuation.

Another nonrandomized, but reader-blinded, controlled echocardiographic study of 934 patients treated for at least 30 days with dexfenfluramine or phentermine/fenfluramine and 539 untreated matched controls found an increase in the prevalence of aortic regurgitation with these drugs (8.9 and 13.7 versus 4.1 percent for placebo) but no difference in the prevalence of mitral regurgitation [29]. There was no difference in the rate of serious cardiac events (myocardial infarction, heart failure, or ventricular arrhythmia) in treated and untreated subjects.

In a prospective randomized trial of dexfenfluramine or sustained-release dexfenfluramine versus placebo administered for an average of 71 days, echocardiograms were obtained in 1072 patients within one month of treatment discontinuation [25]. There was a nonsignificant trend towards more valvular regurgitation with dexfenfluramine compared with placebo (6.5 and 7.3 versus 4.5 percent, respectively). The low incidence of valvulopathy in this report may have been due to the brief duration of therapy; the course of the valvular disease in these patients is discussed below. (See 'Course' below.)

Support for a relationship between the duration of therapy and the incidence of valvular abnormalities comes from the Boston Collaborative Drug Surveillance Program which conducted a population-based and a nested case-control analysis of 6532 subjects treated with dexfenfluramine, 2371 who received fenfluramine, 862 treated with phentermine, and 9281 individuals with obesity who did not take these drugs [27]. The five year cumulative incidence of an idiopathic valvular disorder was:

●0 per 10,000 among subjects who had not taken appetite suppressants or those who had taken phentermine alone

●7.1 per 10,000 among those treated with fenfluramine or dexfenfluramine for less than four months

●35 per 10,000 among patients treated with fenfluramine or dexfenfluramine for four or more months

Another report of 226 individuals with obesity followed prospectively for up to 30 months found significant, but not severe, aortic regurgitation in 6.6 percent and mitral regurgitation in 1.3 percent; there was no association with the dose administered or duration of therapy [30]. The prevalence of valvular regurgitation in this study was comparable to what has been found in the general population in the Framingham Heart Study [31].

Pathology — The gross pathologic features included thickening of the mitral leaflets and chordae, with a glistening white appearance. The histopathologic picture is one of intact leaflet architecture with a plaque-like encasement of the leaflets; proliferative myofibroblasts are surrounded by an abundant extracellular matrix [32].

When cardiac valves excised from patients treated with anorexic agents were compared with normal and rheumatic valves, distinctive histopathologic features were identified [33]. For example, valves exposed to anorexic drugs showed a larger number of onlays (plaques) per valve, while rheumatic valves showed the greatest average onlay size and thickness. The valve onlays from anorexic agent-exposed valves contained a higher percentage of glycosaminoglycans compared with normal and rheumatic valves.

Echocardiography — The 2003 guidelines from the American College of Cardiology, the American Heart Association, and the American Society of Echocardiography (ACC/AHA/ASE) recommended echocardiography to assess valve morphology and regurgitation in patients with a history of use of anorectic drugs or any other drug known to be associated with valvular heart disease, particularly those who are symptomatic, have cardiac murmurs, or have a technically inadequate auscultatory examination [16]. Routine screening of all patients who have taken anorectic drugs was not recommended.

The echocardiographic features of the valvulopathy are similar to those seen in some patients with rheumatic heart disease, but valve obstruction is conspicuously absent. The most common mitral valve findings include the following:

●Thickening and diastolic doming of the anterior mitral leaflets with relative immobility of the posterior leaflets

●Prominent subvalvular disease with thickening and shortening of the chordae tendineae

●Leaflet tethering and malcoaptation contribute to the mitral regurgitation

The aortic valves appear thickened with leaflet retraction, although aortic regurgitation can occur in the presence of echocardiographically normal leaflets.

FDA case definition — Minor degrees of regurgitation (trace or mild mitral regurgitation or trace aortic regurgitation) are relatively common in the general population and may occur without valvular disease [34]. Therefore, these lesser degrees of regurgitation were excluded from the 1997 FDA case definition of fenfluramine or dexfenfluramine-associated valvulopathy. The FDA case definition requires documented mild or greater aortic regurgitation OR moderate or greater mitral regurgitation [15].

Although not included in the FDA case definition, rates of significant (greater than mild) tricuspid regurgitation may approach those for significant mitral regurgitation [35,36]. Tricuspid leaflets may be thickened with restricted mobility.

Course — Echocardiography has been used to evaluate the course of valvular heart disease associated with anorectic drugs. The valvular lesions tend to be stable or improve in the great majority of patients after the cessation of these drugs [37-40].

●One of the trials described above randomly assigned over 1000 patients to dexfenfluramine, sustained-release dexfenfluramine, or placebo for an average of only 71 days [25]. Echocardiograms were obtained one month and one year after discontinuation in 914 patients [37]. The major finding was greater regression of aortic regurgitation compared with the placebo group but no change in mitral regurgitation.

●Similar findings were noted in a second report of 50 patients with previous exposure to fenfluramines who had at least mild mitral (76 percent) and/or aortic regurgitation (86 percent) [39]. An initial echocardiogram was obtained a mean of 190 days after drug cessation and repeat echocardiography performed approximately one year later. Among the 38 patients with mitral regurgitation, 17 improved, 19 experienced no change, and 2 worsened by at least one grade. Among the 43 patients with aortic regurgitation, 19 improved, 22 were stable, and two worsened by at least one grade.

ERGOT DERIVATIVES — Valve disease has resulted from use of ergot derivatives in the treatment of migraine and, less commonly, in the treatment of hyperprolactinemia. Ergot derivatives are no longer used to treat Parkinson disease given the risk of valve disease.

Migraine — The ergot derivatives ergotamine and methysergide have limited utility for treatment and prophylaxis of migraines largely due to their adverse effects which include induction of valvular disease [5,6,41-45]. Although most reported cases of valve disease have been asymptomatic, some patients have developed symptoms and required valve replacement. (See "Acute treatment of migraine in adults" and "Preventive treatment of episodic migraine in adults".)

Hyperprolactinemia — Cabergoline associated valvular disease has only rarely been reported in patients treated for hyperprolactinemia. In the United Kingdom study cited above, for example, only one of the 31 patients with valve disease were being treated for hyperprolactinemia compared with 29 for Parkinson disease [8]. This difference is at least in part dose-dependent. As noted above, the risk of valve disease is particularly increased at doses higher than 3 mg/day [8]; in comparison, the recommended cabergoline dose for hyperprolactinemia is only 0.5 to 2 mg per week. (See "Management of hyperprolactinemia", section on 'Overview of dopamine agonists'.)

Parkinson disease — Ergot-derived dopamine agonists pergolide and cabergoline were previously used to treat Parkinson disease and were associated with a clinically significant risk of valvular heart disease, with lesions similar to those associated with carcinoid, fenfluramine-induced, and other ergot-induced valvular disease [7-9,46-50]. The risk of valvular disease appears to increase relative to the cumulative dose of pergolide or cabergoline. As with anorectic drugs, the mechanism is probably related to pergolide and cabergoline activation of serotonin (5-HT 2B type) receptors expressed on heart valves, which in turn leads to valvular overgrowth [11]. (See 'Pathogenesis' above.)

●In a nested case-control study from the United Kingdom involving a cohort of 11,417 patients taking antiparkinsonian drugs, 31 patients with newly diagnosed cardiac valve regurgitation were compared with 664 controls; six were taking pergolide and six were taking cabergoline [8]. The risk of cardiac valve regurgitation was significantly increased with current use of pergolide (incidence-rate ratio [IRR] 7.1, 95% CI 2.3-22.3) and cabergoline (IRR 4.9, 95% CI 1.5-15.6) but not with other dopamine agonists. The risk of cardiac valve regurgitation was increased only for patients taking pergolide or cabergoline for six months or longer, and was particularly increased for both medications at doses higher than 3 mg daily.

●In a case-control study from Italy involving 155 patients taking dopamine agonists for Parkinson disease and 90 control subjects, the frequency of clinically important valve regurgitation determined by echocardiography was significantly higher in patients taking pergolide or cabergoline compared with controls (23.4 and 28.6 versus 5.6 percent, respectively) but not in those taking nonergot dopamine agonists (0 percent) [9].

Thus, pergolide and cabergoline should NOT be used as treatment for Parkinson disease. In the United States, pergolide was voluntarily withdrawn from the market in March 2007 due to the potential risk of heart valve damage [51].

MDMA (ECSTASY) — Limited data suggest that use of 3,4-Methylenedioxymethamphetamine (MDMA; known as ecstasy) may cause valve disease. An in vitro study found that MDMA and its N-demethylated metabolite 3,4-methylenedioxyamphetamine activated the 5-hydroxytryptamine 2B (5-HT2B) serotonin receptor and elicited a prolonged mitogenic response in human valvular interstitial cells similar to that induced by fenfluramine [52].

A preliminary report of 29 current or former users of MDMA and 29 sex- and age-matched controls found that 8 of 29 MDMA users had abnormal echocardiographic results (using the FDA case definition for valvulopathy due to appetite suppressant drugs [15] (see 'FDA case definition' above)), compared with none in the control group [53]. Among MDMA users, 13 had greater than mild tricuspid regurgitation and six had valvular strands; neither of these findings was present in any of the control subjects.

Other cardiovascular effects of MDMA are discussed separately. (See "MDMA (ecstasy) intoxication".)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Cardiac valve disease".)

SUMMARY AND RECOMMENDATIONS

●Drugs linked to valvular heart disease such as fenfluramine and dexfenfluramine should be avoided. Caution is advised regarding the use of nonprescription anorectic drugs, herbal anorectics, and even diet supplements. The use of some ergot derivatives has also been shown to be associated with a valvulopathy. (See 'Prevention and management' above.)

●A cardiac history and physical examination, emphasizing cardiac auscultation, is indicated in all patients exposed to drugs associated with valvular heart disease. (See 'Prevention and management' above.)

●Echocardiography to assess valve morphology and regurgitation is indicated in patients with a history of use of any drug associated with valvular heart disease who have a heart murmur, symptoms of valvular heart disease, or a technically inadequate auscultatory examination. (See 'Prevention and management' above.)

●The optimal timing of follow-up echocardiography to determine the progression, regression, or stabilization of valvular lesions is not known. Routine repetition of echocardiography was not recommended in past users of anorectic drugs with normal studies or known trivial valvular abnormalities. (See 'Prevention and management' above.)

ACKNOWLEDGMENTS

The UpToDate editorial staff acknowledges William H Gaasch, MD (deceased), who contributed to an earlier version of this topic review.

The UpToDate editorial staff acknowledges Gerard P Aurigemma, MD, who contributed to an earlier version of this topic review.