Amphetamines
Mortality Associated With Anorectics
By James C. Harris, M. D.
This is about the heart problems caused by using anoretics to treat obesity. It doesn't mean that using anoretics recreationally will destroy your heart, but if you have, or there is a family history of heart problems, you might want to read this before using anoretics, either recreationally, or medically. (ed.)
During 1997, several reports provided evidence of substantial extra cardiovascular mortality incidence to the use of certain anorectic drugs for the treatment of obesity. In response to this information and because of a significant number of applications for insurance from individuals with a recent history of such drug use, Dr. Harris reviewed the limited literature on this subject and, in November 1997, produced a white paper for AUL's internal use in dealing with the mortality risk posed by anorectic therapy.
The voluntary withdrawal of the implicated drugs from the market and a substantial reduction in the number of applications from individuals with exposure to the drugs have provided reason to believe that the anorectic risk will soon become a non-issue.
Moreover, as recently reported at a meeting of the American College of Cardiology, one study has failed to detect cardiovascular complications with the use of one of the offending anorectics -- although the period of treatment was considerably less than that observed with prior published reports.
In spite of what appears to be a diminishing concern over the anorectic risk, AUL Reinsurance still receives inquiries from client companies on the risk classification of proposed insureds whose drug histories suggest adverse mortality. Because of this continuing interest, Dr. Harris' review of the anorectic risk is reproduced, and updated, as the singular scientific subject of this issue of MR.
Background
Pharmacologic appetite control in the treatment of obesity has been attempted since before I entered private practice in 1964. One approach in the past was the use of thyroid hormones (irrespective of thyroid status) to boost the metabolic rate. This approach achieved very little except to either
(1) produce a placebo psychologic effect while prescribing doses too low to cause any physiologic effect or
(2) create a new disease (iatrogenic hyperthyroidism) to replace the original disease (obesity). An equally ill-conceived drug intervention of the past was the periodic administration of oral or parental diuretics -- which, of course, resulted in abrupt, transient weight loss in the form of dehydration.
In the 1960s and early '70s, I observed a proliferation of medical entrepreneurs operating fat clinics. Typically, these medicine mongers (armed with a nurse, a scales, a sphygmomanometer and peanut case full of pills) would combine the thyroid medication and diuretics with one of the popular appetite suppressant drugs (commonly referred to as anorectics or anorexigenics), most which are amphetamine congeners (amphetamine look-alike drugs).
These amphetamine derivatives all have sympathomemetic (andrenergic) activity -- meaning that they produce effects mimicking sympathetic nerve stimulation (most notably: central nervous system and cardiovascular excitation). The innovative obesitologists who prescribed these drugs commonly dispensed them also (often at an inflated price) -- thus forming the basis for labeling their operations as pill mills.
Iatrogenic Disease Resulting From Anorectics
As a consequence of overzealous weight control efforts and resulting adverse drug reactions, there emerged a new population of victims suffering from corpulence (or delusions of corpulence) -- all typically thyrotoxic, dehydrated and complaining of symptoms of adrenergic overstimulation: nervousness, anxiety, sweating, insomnia, palpitation and a wide variety of gastrointestinal symptoms.
I had the pleasure of treating a number of these unfortunate individuals 20-30 years ago -- including two physicians' wives who were self-medicating with amphetamine samples affectionately provided by their husbands.
One of these dear ladies was brought to my office following an automobile accident caused by her uncontrolled tremor while driving home from the grocery. The other lady was hypotensive and pre-syncopal as the result of a prolonged paroxysm of atrial tachycardia (documented at 220 beats per minute).
The doctors' wives were easily prevailed upon to abandon the use of amphetamines, and both underwent complete recovery. (One was actually sightly below average weight for her height before taking the drug.)
However, in those days it was very difficult to dissuade most patients from participating in potentially harmful treatments, especially after they had been cleverly brainwashed with unrealistic claims regarding the efficacy of appetite suppressants. An often-heard argument went something like: Well I know it works; it helped me lose weight on six different occassions! -- testimony to the transient nature of anorectic success. (I cringed at such remarks but often lost the battle.)
Newer Treats
While the original adverse reactions to amphetamines were troublesome and sometimes serious, there was solace in the fact that these undesirable effects were functional rather than structural. As such, the resulting conditions were usually completely reversible upon withdrawal from the drug.
Unfortunately, however, more recent attempts to develop the ultimate appetite suppressant have resulted in newer amphetamines which still exhibit the same old sympathomimetic threats but which also appear to be causing very serious irreversible structural abnormalities associated with substantial extra mortality.
Actually, such complications came to light in the late 1960s and early 1970s, when in Europe the use of an amphetamine anorectic (aminorex) was associated with an epidemic of primary (idiopathic) pulmonary hypertension (PPH) in which the mortality rate was 50 percent among affected individuals and in which the average survival was 3-1/2 years among those who died.1
The lesson to be learned from this experience failed to thwart the introduction in the USA of three newer amphetamine anorectics which have received FDA approval in recent years. A review of these drugs follows.
Fenfluramine (PONDIMIN; IONAMIN)
Class: Sympathominetic amine; amphetamine congener
Action: Serotonergic effect -- release of serotonin from terminal nerve endings of cerebral neurons and blockage of reuptake of serotonic by nerve endings; build-up of serotonin in the brain and circulation.
Adrenergic effect -- See Amphetamine and Serotonin Pharmacology
1973 -- FDA approval for short-term (a few weeks) treatment of obesity
1986, 1993 -- Reported to be associated with irreversible pulmonary hypertension 2,3
1996 -- European prospective case-control study of 95 patients with primary pulmonary hypertension and 355 controls demonstrated an association between the use of fenfluramine or dexfenfluramine and PPH. The odds ratio of PPH among users, compared to untreated controls was 1.8 when the drug use was less than three months, sharply rising to 23.1 when drug use was less than three months. The odds ratio dropped sharply when drug use was concluded 12 or more months prior to the index date (2.4, compared to 10.1 encountered among those using fenfluramine or dexfenfluramine during the past 12 months).
1997 -- Two cases of valvular heart disease (VHD) following the treatment with fenfluramine alone reported by FDA.
1997 (September) -- FDA urged withdrawal from market; Wyeth-Ayerst complied.
Dexfenfluramine (REDUX)
Class: Sympathomimetic amine; amphetamine congener
Action: Serotonergic effect -- release of serotonin from terminal nerve endings of cerebral neurons and blockage of reuptake of serotonin by nerve endings; build-up of serotonin in the brain and circulation.
Adrenergic effect -- See Amphetamine Pharmacology
1996 (April) -- FDA approval for the long-term treatment of obesity
1996 (August) -- European prospective case-control study of 95 patients with primary pulmonary hypertension and 335 controls demonstrated an association between the use of dexfenfluramine and PPH. The odds ratio of PPH among users, compared to untreated controls was approximately 23 when drug use was less than three months.4
1997 (August) -- Seven cases of valvular heart disease following treatment with dexfenfluramine (two of which were in combination with phentermine) reported by FDA.5
1997 (September) -- FDA urged withdrawal from market; Wyeth-Ayerst complied.6
1998 (March) -- Georgetown University Medical Center study found no statistically significant increase in valvular heart disease among patients taking dexfenfluramine for a median duration of 77 days, compared to a control group taking placebo.7
Phentermine (FASTIN)
Class: Sympathominetic amine; amphetamine congener.
Action: Serotonergic effect -- interference with pulmonary clearance of serotonin; resulting in serotonin; resulting in serotonin build-up in the circulation.
Adrenergic effect -- See Amphetamine and Serotonin Pharmacology
1991: FDA approval for short -term (a few weeks) treatment of obesity.
No reports of adverse reactions (for phentermine alone).
FDA has not issued a recall.
Fen-Phen (fenfluramine+phentermine, in combination)
Individual drugs approved by FDA but combination not studied.
1992 -- Beginning use of fen-phen combo for treating obesity.
1996 -- more than 18 million prescriptions written in USA
1997 (August) -- Case report from Mass Gen Hosp & Harvard Medical School published in NEJM: 29-year-old female took Fen-Phen for 23 days and developed right heart failure five months later.8 Cardiac catheterization revealed severe pulmonary hypertension. Other studies ruled out other causes of pulmonary hypertension. Death from cor pulmonale occurred about three months after diagnosis. An autopsy revealed features typical of plexogenic pulmonary hypertension. The authors call attention to the fact that the histologic age of the pulmonary lesions correlated with the time interval since fen-phen ingestion.
1997 (August) -- Study from Mayo Clinic reported in NEJM on 24 females presenting with recent onset valvular heart disease and no antecedent history of cardiac disease.9 Eight of these patients also suffered from pulmonary hypertension. All had received fenfluramine-phentermine therapy for appetite suppression for periods of one to 28 months up to the time of entering the study. Most lesions were classified as moderate or severe, and five required surgery. Pathologically, the valve deformities were unlike congenital, rheumatic or degenerative valvular disease. Rather, the valve leaflets and chorda tendinae were encased in a white, glistening plaque-like substance identical to that found with carcinoid syndrome and complicating ergotamine and methysergide treatment of migraine.
1997 (August) -- 28 additional cases of moderate to severe valvular heart disease associated with fen-phen therapy reported by the FDA, and 10 of these patients also suffered from pulmonary hypertension.5
Amphetamine and Serotonin Pharmacology Adrenergic effects. Amphetamines and their derivatives are felt to suppress appetite through inhibition of the appetite (feeding) center in the lateral hypothalamus. Amphetamine congeners operate through the induction of secretion of norepinephrine (a neurotransmitter) from the terminal nerve endings of the sympathetic division of the autonomic nervous system.
Norepinephrine binds to receptors of effector cells in various body tissues, mainly producing excitatory effects such as an accelerated heart rate; increased cardiac contractility and excitability; vasoconstriction of some vessels; increased blood pressure; pupillary dilation; glucose release from the liver; increased skeletal muscle strength; and increased mental activity. Norepineprine also, to a lesser extent, mediated inhibitory effects, including decreased peristalsis and vasodilatation in certain tissues.
Serotonergic Effects
Fenfluramine and dexfenfluramine have been shown to stimulate the release of serotonin (another neurotransmitter) from cerebral nerve endings and to block reuptake of serotonin by these nerve endings -- resulting in an accumulation of serotonin in brain tissue and increased levels of circulating serotonin.
Taking this analysis one step further, carcinoid tumors are known to secrete serotonin, and there exists a remarkable structural similarity among serotonin, ergotamine and methysergide (the latter two drugs having been long used in the treatment of migraine and other vascular headaches). Finally, carcinoid tumors and ergotomine and methysergide therapy are all known to be complicated by predominantly right-sided cardiac valvular lesions that are microscopically identical with the Mayo clinic findings (cited above) in patients receiving fenfluramine-phentermine therapy.
The Mayo Clinic report on fen-phen and valvular heart disease also identified echocardiographic, surgical and pathologic features in their patients that were indistinguishable from carcinoid and ergotamine-induced heart valve disease.9 Additionally, serotonin may play a role in the development of pulmonary hypertension. It has been postulated that the pathogenesis of this form of PPH may be related to serotonin-induced pulmonary vasoconstriction and/or serotonin inhibition of a potassium current in pulmonary vascular smooth muscle cells.10
Clincopathologic Features
Primary Pulmonary Hypertension. The histopathologic features of PPH are collectively referred to as plexogenic pulmonary arteriopathy and are distinctively unlike that seen in the more common secondary forms of pulmonary hypertension, such as that resulting from chronic obstructive lung disease, interstitial pulmonary fibrosis, pulmonary thromboemolic disease and connective tissue disease. PPH histology includes pulmonary arterial medial hyperplasia, intimal proliferation and fibrosis, in situ thrombosis, diffuse pulmonary arteritis, pulmonary vascular necrosis and pulmonary-bronchial arterial shunting.
These pathologic changes result in a decreased cross-sectional area of the pulmonary vascular bed -- leading to chronic right ventricular pressure and volume overload, right ventricular hypertrophy and triscuspid regurgitation. The end result is right ventricular failure (cor pulmonale), signaling and expected survival of only about six months.
Valvular Heart Disease. The pathologic features of anorectic-induced valvular heart disease are (as noted above) the same as those associated with carcinoid and treatment with ergotamine and methysergide. Multiple valves may be involved.. The aortic and mitral valves are most commonly affected. Tricuspid lesions are less common, and pulmonic lesions are infrequent.
Grossly, the affected valves leaflets are thickened and encased by a white glistening plaque, resulting in retraction, diminished mobility and shortening of the chordae tendinae -- leading to loss of coaptation of the leaflets and hemodynamically significant regurgitant valvular lesions. Stenotic lesions have not been reported. The clinical course and surgical indications are much the same as those encountered with other acquired heart valve lesions.
Cause And Effect
Because of insufficient control group studies, it is not possible to definitely conclude that pulmonary hypertension and valvular heart disease occur as the result of treatment with fenfluramine, dexfenfluramine or the combination of fenfluramine and phetermine. On the other hand, there exists very compelling evidence of such an association.
A recapitulation of this evidence follows:
Three decades ago the anorectic aminorex was undeniably implicated in a highly lethal epidemic of primary pulmonary hypertension.
Fenfluramine and dexfenfluramine are strikingly similar to molecular structure to aminorex.
Fenfluramine and dexfenfluramine are known to increase circulating levels of serotonin.
Carcinoid tumors which secrete serotonin and structurally similar ergotamine and methysergide are well-known to result in an unusual form of valvular heart disease.
The Valvular lesions associated with carcinoid, ergotamine and methysergide are identical to those reported following aminorex and fenfluramine-phentermine therapy.
A 1996 prospective case-control study of 95 patients with priary pulmonary hypertension and 335 controls revealed the risk of PPH to be 23 times greater among patients receiving treatment with fenfluramine or dexfenfluramine as compared to untreated patients. In 1997, an additional case of rapidly fatal PPH occurred in a 29-year-old female following only 23 days of fenfluramine-phentermine treatment. The histologic age of the lesions correlated with the time period of fen-phen therapy.
In August 1997, two reports identified 52 female patients with recent onset of valvular heart disease following treatment with combined fenfluramine and phentermine. The lesions were unique and undistinguishable from previously described valvular lesions associated with carcinoid and ergotamine and methysergide treatment. Similar reports of valvular lesions followed treatment with fenfluramine alone, dexfenfluramine alone and combined dexfenfluramine and phentermine.
Phentermine alone has not been implicated as a cause of primary pulmonary hypertension. However, this drug has been reported to inhibit the usual pulmonary clearance (degradation) of serotonin, an effect which would augment the level of circulating serotonin and its pathologic consequences.
Interestingly, this phenomenon provides and explanation for the apparent dissimilarity between the predominantly right-sided valvular lesions of carcinoid/ergotamine/methysergide and the predilection of phen-phen to involve the aortic and mitral valves. Simply stated, the pulmonary degradation of serotonin may be viewed as protective to the left side of the heart, and the inhibition of this degradation by phentermine should be expected to counter that protection.
More recently, as reported at the 47th Annual Scientific Session of American College of Cardiology (March 1998), the first prospective, randomized, placebo-controlled study compared outcomes among patients who had received dexfenfluramine for a median duration of 77 days against a control group.
Based on subsequent echocardiography of 1,072 patients, the incidence of valvular heart disease was 6.5 percent in the dexfenfluramine group, 7.3 percent in a group taking sustained-release dexfenfluramine and 4.5 percent in the placebo group -- statistically insignificant.
However, caution must be used in interpreting this data. Consider the following:
The study involved only dexfenfluramine, not fenfluramine or phentermine.
The duration of treatment was largely limited to two to three months. Longer exposure may have produced results more reflective of earlier reports.
The study provides no information on long-term, late-onset complications following short-term dexfenfluramine use.
The study did not address the evidence implicating dexfenfluramine in the development of primary pulmonary hypertension. Based on all available evidence, it would seem that there is good reason to suspect that fenfluramine and dexfenfluramine may be responsible for a substantial number of cases of both primary pulmonary hypertension and valvular heart disease and that these adverse effects are medicated through the induction of serotonin secretion by sympathetic nerve endings and blockage of the reuptake of serotonin.
To delay underwriting recognition of this association until the emergence of absolute scientific proof of a cause-effect relationship would seem imprudent.
The Anorectic Benefit Myth
It has been argued that the potential health benefits of pharmacologic appetite suppression outweigh the risks associated with today's annorectics. However, these purported benefits must be viewed with some degree of skepticism. Consider the following:
Weight loss is beneficial only when it is maintained for a long time period.
No scientific study has shown that appetite suppressant therapy can maintain weight reduction indefinitely.
It is generally accepted that the phenomenon of adaptation to appetite suppressants will limit therapeutic weight reduction to no more than 5-10 percent of the patient's original weight.
The safety of anorectic therapy must be questioned, particularly in view of the increasing incidence of adverse reactions with longer duration of treatment.
Intermittent appetite suppression leads to weight cycling -- which is worse than stable obesity (and associated with an increased risk of coronary heart disease and death).
No studies have shown that appetite suppression prevents obesity-related diseases or improves survival. These realities have led Curfman,11 in a recent New England Journal of Medicine editorial, to conclude that:
The only justifiable medical use of anorectic drugs is in seriously obese patients who have obesity-related illnesses such as coronary heart disease, diabetes, hypertension and hyperlipidemia.
...we should demand a moratorium on the use of anorectic drugs for the purpose of cosmetic weight loss.... For generally healthy people who want to lose a few pounds, there are safer alternatives.
Recent reports of pulmonary hypertension and valvular heart disease following anorectic therapy are chilling reminders that succumbing to the allure of diet pills as a quick fix for excess weight may be courting disaster.
Risk Assessment
While there is good reason to accept a cause-effect relationship among anorectic therapy and both VHD and PPH, there is very little basis for quantitating the resulting extra mortality. The lack of long-term follow up and the paucity of controlled studies makes this determination difficult. However, a simplified method of approximating mortality is suggested from the available data.
On October 1, 1997, data were released to the lay press12 calling attention to clinical investigations from five USA centers on follow up of patients who had received fenfluramine or dexfenfluramine for an average of one year during the recent past.
These investigations were solicited by and reported to the FDA. Briefly, the follow-up of all 291 patients who received these drugs identified 92 patients with valvular heart disease (VHD) -- 32 percent of the total. Subsequent to this preliminary report, these five centers have conducted studies on a few hundred more patients -- with very similar results. These findings naturally prompt the question of whether or not some of these patients might have had pre-existing valvular disease. However, the prevalence of acquired VHD in the general population is only about 3 percent -- providing a relative risk of 11:1.
An August 1997 report from the FDA reports details of the first 28 cases among the investigations in which the combination of fenfluramine and phentermine had been taken for two to 36 months (average: 10 months) followed by the detection of VHD. Fourteen percent of these patients were asymptomatic (incidental murmur only).
A Mayo Clinic report in August 1997 reported five (21 percent) of a series of 24 patients with VHD following fenfluramine-phentermine therapy to be asymptomatic at the time of diagnosis. For shorter intervals between drug exposure and diagnosis, the percentage of affected individuals remaining asymptomatic would naturally be significantly greater. The valvular lesions were described as moderate to severe in 18 (78 percent) of the patients. Interestingly, 10 (36 percent) of these patients also suffered from pulmonary hypertension.
This information allows an analysis of the expectation of extra mortality among proposed insureds with a history of conclusion of fenfluramine or dexfenfluramine therapy 12-36 months prior to application, and in whom cardiovascular symptoms are absent and no objective evaluation for possible VHD (auscultation or echocardiography) has been carried out.
Studies from 30 years ago pointed to a 50 percent mortality among patients developing PPH after treatment with a drug closely resembling fenfluramine (structurally and pharmacologicaly). Furthermore, the survival of the patients who died was an average of 3-1/2 years. This information points to the need for individuals who have been treated with fenfluramine or dexfenfluramine during the past 12 months.
As noted above, 10 (36 percent) of the FDA-reported 28 cases of VHD following fenfluramine/phentermine therapy also had PPH.
Referring to the cohort of 291 patients receiving fenfluramine or dexfenfluramine, even if we assume that the 199 patients without VHD were also free of PPH, the incidence of PPH in the entire cohort might be expected to be 12 percent (36 percent of the 32 percent with VHD).
Given the 50 percent mortality and the extremely short time previously reported between diagnosis of PPH and death (3-1/2 years), the projected mortality from PPH would be 17 deaths/1,000/year [(120 x 0.5)/3.5] -- representing an extra mortality among females of approximately 4,800 percent at age 20, ranging to 1,500 percent at age 50 (that group most commonly receiving anorectics) -- unless such individuals are determined not to have pulmonary hypertension (at least through Doppler echocardiography).
Because of the substantial drop in the probability of PPH among subjects who have remained asymptomatic 12-36 months after discontinuation of the drugs,4 offers of insurance would seem to be justified for those proposed insureds in this group. A reasonable approach would be to assume a modest increase in mortality, additive to the mortality associated with the co-existing risk of VHD. Beyond 36 months after discontinuation of the drugs, the risk of extra mortality from either VHD or PPH would likely be negligible among asymptomatic subjects.
Conclusion
During 1996 and 1997, several reports pointed to a causal relationship between the ingestion of certain anorectics and the subsequent development of primary pulmonary hypertension and a distinctive form of regurgitant valvular heart disease. Significant extra mortality has been observed to follow the administration of these drugs (fenfluramine or dexfenfluramine, either singly or in combination with phetermine).
While a single control group study casts some doubt on a connection between dexfenfluramine treatment and valvular heart disease, other very compelling evidence exists to suggest an association between these drugs and both valvular abnormalities and pulmonary hypertension. Accordingly, any recent history of treatment with the implicated drugs should prompt underwriting consideration of possible resulting extra mortality.
References
1. Gutner, HP. Amenorex and pulmonary hypertension. Cor Vasa 1985; 27:160-171.
2. McMurray, J; Bloomfield, P; Miller, HC. Irreversible pulmonary hypertension after treatment with fenfluramine. BMJ 1986; 293:51-52.
3. Brenot, F; Herve, P; Repitpretz, P; et al. Primary pulmonary hypertension and fenfluramine use. BR Heart J 1993; 70:537-541.
4. Abenhaim, L; MOrid, Y; Brenot, F; et al. Appetite-suppressant drugs and the risk of pulmonary hypertension. N Engl J Med 1996; 35:609-616
5. Graham, DJ; Green, L. Further cases of valvular heart disease associated with fenfluramine-phentermine. N Engl J Med 1997; 337:635.
6. Shelton, DL. Doctors feel the squeeze in FDA's anti-obesity drug ban. AMA News 1997; 40 (30):1.
7. Weissman, NJ; Tighe, JF; Gottdiener, JT; Gwynne, JT. Prevalence of valvular abnormality in patients exposed to dexfenfluramine; results of a randomized placebo controlled trial. Presented at the American College of Cardiology 47th Annual Scientific Session. March 290April 1. Atlanta.
8. Mark, EJ; Patalas, ED; Chang, HT; et al. Fatal pulmonary hypertension associated with short-term use of fenfluamine and phentermine. N Engl J Med 1997; 337:602-606.
9. Connolly, HM; Crary, JL, McGoon, MD; et al. Valvular heart disease associated with fenfluramine-phentermine. N Engl J Med 1997; 337:581-588.
10. Weir, EK; Reeve, HL; Huang, JM; et al. Anorexic drugs aminorex, fenfluramine and dexfenfluramine inhibit potassium current in rat pulmonary vascular smooth muscle and cause pulmonary vasoconstriction. Circulation 1996; 94:2216-2220.
11. Curfman GD. Editorial: Diet pills redux. N Engl J Med; 337:629-630.
12. Johannes L. New diet-drug data spark more controversy. In: The Wall Street Journal. Oct. 1, 1997; B1.
Books The Speed Culture:
Amphetamine Use and Abuse in America
The only non biased, factual book about the history of speed I have found in print. Written in 1975, it doesn't cover recent history, but does give a good history up to the time of publication.
It was written by the author of Marihuana Reconsidered and describes how amphetamines have been used both medically and recreationally in the United States.
Speed Culture
More Amphetamine Books Amphetamine books from Amazon
Amphetamine Related More Amphetamine Articles Various Amphetamine Links