|
|
Background Case reports and echocardiographic studies suggest that the ergot-derived dopamine agonists pergolide and cabergoline, used in the treatment of Parkinson's disease and the restless legs syndrome, may increase the risk of cardiac-valve regurgitation.
Methods We used data from the United Kingdom General Practice Research Database to identify a population-based cohort comprising 11,417 subjects 40 to 80 years of age who were prescribed antiparkinsonian drugs between 1988 and 2005. We conducted a nested case–control analysis within this cohort in which each patient with newly diagnosed cardiac-valve regurgitation was matched with up to 25 control subjects from the cohort, according to age, sex, and year of entry into the cohort. Incidence-rate ratios for cardiac-valve regurgitation with the use of different dopamine agonists were estimated by conditional logistic-regression analysis.
Results Of 31 case patients with newly diagnosed cardiac-valve regurgitation, 6 were currently exposed to pergolide, 6 were currently exposed to cabergoline, and 19 had not been exposed to any dopamine agonist within the previous year. The rate of cardiac-valve regurgitation was increased with current use of pergolide (incidence-rate ratio, 7.1; 95% confidence interval [CI], 2.3 to 22.3) and cabergoline (incidence-rate ratio, 4.9; 95% CI, 1.5 to 15.6), but not with current use of other dopamine agonists.
Conclusions In this study, use of the dopamine agonists pergolide and cabergoline was associated with an increased risk of newly diagnosed cardiac-valve regurgitation.
An association has been reported between treatment with the dopamine agonist pergolide (Permax, Eli Lilly) and the development of fibrotic valvular heart disease, particularly when pergolide was administered in high doses over long periods.3,4,5,6,7 On echocardiography, patients had mild-to-severe cardiac-valve regurgitation, often involving more than one valve. Histologic findings have been found to resemble those of valvulopathies related to the carcinoid syndrome and the use of ergot derivatives or fenfluramine.5,6,7 In several cases, the valvulopathy improved when pergolide was discontinued.3,5,8 Two recent case reports have also suggested that cabergoline (Dostinex, Pfizer) is associated with an increased risk of valvular heart disease.5,9 A case of severe tricuspid-valve regurgitation was reported after 5 years of therapy with bromocriptine (Parlodel, Novartis).10
There are mechanistic grounds for believing that not all dopamine agonists are equally likely to be implicated in the development of cardiac-valve regurgitation. Pergolide and cabergoline are potent agonists of the 5-hydroxytryptamine 2B (5-HT2B) receptor expressed on heart valves,11,12 whereas other agents in this class, such as bromocriptine and lisuride (Dopergine, Schering), have antagonistic properties.12,13 Pramipexole (Mirapex, Boehringer Ingelheim) and ropinirole (ReQuip, GlaxoSmithKline) have low affinity to the human 5-HT2B receptor.14 Preferential activation of this receptor has been shown to induce prolonged mitogenic effects in cardiac fibromyoblasts, which could lead to inducing valvular fibroplasia.11,15,16,17
We conducted a cohort study with a nested case–control analysis to investigate the risk of newly diagnosed cardiac-valve regurgitation associated with the use of different dopamine agonists. The hypothesis was that pergolide and cabergoline, but not the other agents, would be associated with an increased risk of cardiac-valve regurgitation.
Methods
Source of Data
Data were obtained from the General Practice Research Database (GPRD), which includes computerized medical records of more than 6.3 million patients from more than 350 general practices in the United Kingdom. Contributing general practitioners initially received 12 months of instruction in the standardized recording of medical information on computers. The information recorded includes the patient's demographic characteristics, symptoms, history, medical diagnoses, and drug prescriptions, as well as details of referrals to specialists and hospitals. More detailed clinical information (e.g., on test results and from hospital discharge summaries and letters from specialists), entered separately by the general practitioner in free text fields, is available on request. The completeness and validity of the recorded information on diagnoses and drug exposures, as checked on an ongoing basis by staff of the GPRD, have been shown in several studies.18,19,20 The database has been used for the study of numerous diseases, including valvular heart disease.21
Acquisition of access to the database for this study was funded by a grant from the Canadian Foundation for Innovation and by the Canadian Institutes of Health Research, and an unrestricted grant from Schering. These sources of the funding were not involved in the design or realization of the study, the analysis of the data, or the decision to publish the results. The scientific and ethics advisory group of the GPRD approved the study. Because anonymized electronic records were used as the source of data, written informed consent was not required.
Study Cohort
We identified all patients in the database who were between 40 and 80 years of age and had received at least two prescriptions for antiparkinsonian medications between January 1, 1988, and August 31, 2005. The antiparkinsonian drugs qualifying a patient for entry into the study cohort included the dopamine precursor levodopa, the monoamine oxidase inhibitor selegiline (Eldepryl, Somerset; Zelapar, Valeant Pharm), and the dopamine agonists bromocriptine, cabergoline, pergolide, lisuride, pramipexole, and ropinirole.
The date of the second prescription of an antiparkinsonian medication was defined as the date of entry into the study cohort. Patients with a history of rheumatic heart disease, congenital heart disease, congestive heart failure, dilated cardiomyopathy, endocarditis or myocarditis, the carcinoid syndrome, intravenous drug abuse, or heart-valve abnormalities (including mitral-valve prolapse and cardiac murmurs) before entry into the cohort were excluded. Also excluded were patients who had received fenfluramine, dexfenfluramine, phentermine (Adipex-P, Teva; Ionamine, UCB), ergotamine (Ergomar, Harvest Pharms), dihydroergotamine (DHE-45; Migranal, Valeant), or methysergide.
The date of exit from the cohort (exit date) was defined as the date of the first occurrence of the following events: a diagnosis of valvular heart disease; a condition or an event that was one of the exclusion criteria, with the exception of congestive heart failure (which may precede a diagnosis of valvular heart disease); death; the end of registration with the practice; cessation of the contribution of data to the GPRD by the general practice; or the end of the study period (August 31, 2005).
Identification of Case Patients and Controls
We used diagnostic codes compatible with valvular regurgitation, cardiac-valve interventions, and cardiac murmurs to identify possible new cases of cardiac-valve regurgitation during follow-up in the study cohort of patients taking antiparkinsonian medications. Patients who had had a myocardial infarction within 3 years before receiving a diagnosis of valvular regurgitation were excluded, since this condition may be a cause of valvular insufficiency. For each patient with a recorded diagnosis suggestive of valvular regurgitation, we requested more detailed clinical information available in the database that was related to cardiologic findings (including results of the clinical examination, echocardiography, and heart catheterization), hospital discharge summaries, and letters from specialists. This information was independently reviewed by a senior consulting cardiologist and two trained physicians who were unaware of the patient's history of use of a dopamine agonist. Patients were considered to have newly diagnosed cardiac-valve regurgitation if they had no history of a cardiac-valve abnormality and if they had received a diagnosis of cardiac-valve regurgitation on the basis of echocardiography, heart catheterization, or clinical examination. The index date was defined as the first date of recorded cardiac-valve regurgitation or a related symptom.
For each case patient, we randomly selected up to 25 controls from the study cohort of patients taking antiparkinsonian medications. Controls were matched to case patients according to sex, age (within 2 years), and year of entry into the study cohort. The date that resulted in the same duration of follow-up for the case patient and the control was designated the index date for the control. Patients with a myocardial infarction within 3 years before the index date were not selected as controls. For both case patients and controls, recorded data had to be available for at least 12 months before the index date.
Exposure
For all case patients and controls, we identified exposure to all dopamine agonists during the 12 months before the index date. Case patients and controls were stratified according to the medication prescribed: bromocriptine, cabergoline, pergolide, lisuride, pramipexole, or ropinirole.
We defined two periods of exposure on the basis of the timing of the prescription of a dopamine agonist. Current use was defined as a prescription that was still in effect within 6 months before the index date, and recent use as a prescription that ended between 6 and 12 months before the index date. No current or recent use was defined as no use of any dopamine agonist within the 12 months before the index date. Those who had received more than one dopamine agonist in the year before their index date were grouped according to the dopamine agonist prescribed for the longest period during the 12 months. If two or more dopamine agonists had been taken for the same number of months, the patients were grouped according to the most recent use. Case patients and controls who had taken more than one dopamine agonist during the 6 months before their index date (defined as current multiple use) were excluded from the analysis.
We quantified exposure to a dopamine agonist in two ways. First, we categorized the cumulative duration of current use of a dopamine agonist as less than 6 months or 6 or more months. Second, we calculated the daily dose of a dopamine agonist for patients currently using pergolide or cabergoline by multiplying the tablet strength by the prescribed number of tablets per day and categorized the dose as 3 mg or less daily or more than 3 mg daily (with the defined daily dose [the average maintenance dose per day] used as the cutoff value, as designated by the World Health Organization).
Statistical Analysis
The incidence-rate ratio (the ratio of the incidence rate of newly diagnosed cardiac-valve regurgitation among the patients currently exposed to a dopamine agonist under study to the incidence rate among patients who had not been exposed to a dopamine agonist) was estimated from odds ratios calculated with the use of conditional logistic regression. We constructed separate models characterizing case patients and controls according to the last daily dose of a dopamine agonist and the cumulative duration of current use of pergolide or cabergoline. The model included the use of all dopamine agonists and considered the following covariates: body-mass index; smoking status; the presence or absence of hypertension, diabetes, coronary heart disease, Parkinson's disease, restless legs syndrome, and hyperprolactinemia; and the use of antiparkinsonian drugs other than the dopamine agonists under study. We used a backward-selection procedure, including variables in our final model that led to a change of more than 15% in the risk estimates for valvular regurgitation associated with current use of pergolide or cabergoline, as compared with no current or recent use of any dopamine agonist. The reference category for the analysis comprised case patients or controls who had not been exposed to any dopamine agonist under study during the 12 months before the index date. All reported P values are two-tailed, with a significance level of 0.05, and 95% confidence intervals (CIs) were calculated for all incidence-rate ratios. The excess risk of cardiac-valve regurgitation incurred by current use of pergolide or cabergoline was calculated by multiplying the incidence-rate ratio minus 1 by the incidence rate among patients who had not been exposed to a dopamine agonist.
Results
A total of 11,417 patients were included in the final cohort, all of whom had received at least two prescriptions for antiparkinsonian drugs and met all the inclusion and exclusion criteria (Figure 1). The mean age at entry into the study cohort was 69 years, and the mean duration of follow-up (from entry to exit from the study cohort) was 4.2 years. The numbers of patients taking dopamine agonists at any time during follow-up and the respective duration (person-years) of exposure during follow-up were as follows: 828 patients taking bromocriptine (1683 person-years), 61 taking lisuride (88 person-years), 931 taking pergolide (2031 person-years), 1228 taking cabergoline (1812 person-years), 492 taking pramipexole (648 person-years), and 993 taking ropinirole (1565 person-years). Patients may have been included in more than one of these categories of exposure because of changes in their use of dopamine agonists during follow-up. The total number of person-years of no exposure to dopamine agonists was 34,548. There were 7702 patients who were never exposed to any dopamine agonist during follow-up, accounting for 28,892 person-years.
|
|
|
The rate of cardiac-valve regurgitation was elevated among patients who were currently exposed to either pergolide (adjusted incidence-rate ratio, 7.1; 95% CI, 2.3 to 22.3) or cabergoline (adjusted incidence-rate ratio, 4.9; 95% CI, 1.5 to 15.6) (Table 3), but not among those who were currently exposed to other dopamine agonists. For amantadine, the only concurrent medication found to have a significant association with cardiac-valve regurgitation, the adjusted incidence-rate ratio was 3.5 (95%, CI, 1.1 to 11.3). The adjusted incidence-rate ratios were particularly elevated for daily doses greater than 3 mg of pergolide (37.1; 95% CI, 5.1 to 270.6) and 3 mg of cabergoline (50.3; 95% CI, 6.6 to 381.4), as well as for a duration of use of 6 months or more (Table 4).
|
|
We did not have systematic clinical follow-up data on the case patients with cardiac-valve regurgitation identified in this analysis. One patient had echocardiographic evidence of regression of aortic regurgitation after discontinuation of cabergoline. Valve replacement was considered in another patient, who had been exposed to pergolide, but the procedure was not performed.
Discussion
Our study showed that the use of pergolide or cabergoline was associated with a significantly increased risk of newly diagnosed cardiac-valve regurgitation. This risk was particularly high among patients who had taken daily doses of pergolide or cabergoline that exceeded 3 mg; the risk was increased only among those who had taken either drug for 6 or more months. The risk was not increased among patients treated with other ergot-derived dopamine agonists or with dopamine agonists that are not derived from ergot.
Our finding of an increased risk of valvular regurgitation with pergolide is supported by a number of case reports3,4,5,6,7,22 and echocardiographic studies.3,8 One echocardiographic study showed an odds ratio of 3.7 (95% CI, 0.7 to 19.2) and 4.0 (95% CI, 1.3 to 12.2) for clinically significant aortic or mitral regurgitation, respectively, and an odds ratio of 18.4 (95% CI, 1.2 to 283) for clinically significant tricuspid regurgitation among 46 patients treated with pergolide, as compared with an age-matched historical control group.3 Evidence of an increased risk of cardiac-valve regurgitation with cabergoline consists primarily of case reports.5,9 However, a recent echocardiographic study showed a similar prevalence of cardiac-valve regurgitation among patients exposed to cabergoline and pergolide23: clinically significant regurgitation was present in 6 of 13 patients (46%) taking cabergoline and in 9 of 29 patients (31%) taking pergolide, but in only 6 of 49 (12%) age-matched controls who did not have Parkinson's disease. In contrast, another retrospective evaluation of cardiopulmonary fibrotic side effects in 234 patients treated with cabergoline showed a low risk of fibrotic side effects, particularly cardiac valvulopathy (in 1 patient).24
In our study, no patients with newly diagnosed valve regurgitation were identified among those treated with bromocriptine or lisuride (ergot-derived dopamine agonist) or with ropinirole or pramipexole (dopamine agonists that are not derived from ergot) within the 12 months before the patients' index dates. Among the controls, the prevalence of exposure to pramipexole, ropinirole, and bromocriptine was similar to that of exposure to pergolide and cabergoline (Table 3). Nevertheless, the upper bounds of the 95% CIs derived from the size of the population studied and the absence of observed events suggest an upper bound to the rates that is compatible with values up to 17.8 events per 10,000 persons per year for bromocriptine, 19.1 events per 10,000 per year for ropinirole, 46.1 events per 10,000 per year for pramipexole, and 334.7 events per 10,000 per year for lisuride. Differences in the affinity of the various dopamine agonists for valvular 5-HT2B receptors11,12,13,14,15 could explain the differences in risk observed in our study. The unexpected finding of an increased risk of cardiac-valve regurgitation associated with amantadine use requires further investigation, since this drug is not known to activate 5-HT2B receptors.
Some limitations of our study need to be considered. All information was recorded prospectively, ruling out recall bias. Selection bias in the choice of control subjects is unlikely, because we used a nested case–control design in a defined cohort of both case patients and controls. Detection bias could be of concern, since pergolide was discussed as a possible cause of cardiac valvulopathy before the end of the study period. The British Committee on Safety of Medicines published an alert on pergolide-associated valvulopathy in September 2003,25 and this alert may have led to an increased use of diagnostic measures in patients receiving pergolide in order to detect valvulopathies. To investigate this potential source of bias, we conducted a subgroup analysis involving only patients in whom cardiac-valve regurgitation had been diagnosed before September 2003. This analysis did not materially change our results.
Our study may have underestimated the incidence of asymptomatic cases, since it was not based on echocardiographic monitoring of all patients in the cohort. If asymptomatic cases associated with the use of pergolide or cabergoline were underdiagnosed in the study, our reported estimates would, in fact, be underestimates of the true risks. Since we did not have data on echocardiography or heart catheterization for all the case patients, we conducted a subgroup analysis involving 16 case patients for whom confirmation of the diagnosis by echocardiography or heart catheterization was recorded. This subgroup analysis also showed an increase in the risk of cardiac-valve regurgitation associated with pergolide and cabergoline, although the results were not significant because of the reduced power of this analysis.
In conclusion, our study showed that treatment with either pergolide or cabergoline, particularly at daily doses greater than 3 mg and for periods of 6 months or longer, was associated with a substantially increased risk of newly diagnosed cardiac-valve regurgitation. There was no evidence of such an increase in risk with the use of other dopamine agonists.
Dr. Schade reports receiving a grant from Schering; Dr. Suissa, receiving consulting fees from GlaxoSmithKline and Schering, lecture fees from Boehringer Ingelheim and Pfizer, grant support from the Canadian Institute of Health Research, AstraZeneca, and Organon, and fees for serving as an expert witness for Schering; Dr. Haverkamp, receiving consulting fees from Schering, GlaxoSmithKline, and Pfizer and lecture fees from Novartis; and Dr. Garbe, receiving consulting fees and an unrestricted grant from Schering for acquisition of access to the GPRD. No other potential conflict of interest relevant to this article was reported.
We thank the general practitioners who contribute data to the GPRD for their continued participation and care of patients.
Source Information
From the Department of Clinical Pharmacology, Charité Campus Mitte (R.S., F.A., E.G.), and the Department of Cardiology, Charité Campus Virchow (W.H.), Charité-Universitätsmedizin Berlin, Berlin; and the McGill Pharmacoepidemiology Research Unit, Division of Clinical Epidemiology, Royal Victoria Hospital, McGill University Health Centre, Montreal (S.S.).
References
| |||||||||||||||||||||||||||||||||||||||||||||||||||||
Related Letters:
Dopamine Agonists and Valvular Heart Disease
Stephens J. W., Price D. E., Ionescu A., Linkova H., Ruzicka E., Penicka M., Kast R. E., Altschuler E. L., Ziegler J. L., Bukhman G. Y., Sartiani L., Cerbai E., Mugelli A., Schade R., Andersohn F., Garbe E., Zanettini R., Pezzoli G., Roth B. L.
Extract |
Full Text |
PDF
N Engl J Med 2007;
356:1676-1680, Apr 19, 2007.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |
Previous
