Long-Term Statin Use and the Risk of Parkinson’s Disease | Page 1
Published Online: August 12, 2013
Bitya Friedman, MD; Amnon Lahad, MD, MPH; Yizchak Dresner, MD, MS; and Shlomo Vinker, MD
Parkinson’s disease (PD) is the second-most common neurodegenerative disorder after Alzheimer's disease, and it causes much disability and suffering for patients and their families. It is a progressive disease with no known cure and becomes more common as people get older; thus, it will become an even greater and more frequent burden as the worldwide population ages.1 It is imperative that every effort should be made to identify possible preventable causes and risk factors for this disease.
Recent studies have suggested that cardiovascular factors—hypertension, diabetes, hypercholesterolemia, and smoking—play a role in the etiology of PD. A few researchers examined the association between diabetes,2,3 body mass index,4,5 hypertension,6 and PD, and some found a positive correlation. By contrast, however, there is evidence that smoking is a protective factor for PD.7,8
Some studies have also shown an inverse correlation between lowdensity lipoprotein cholesterol (LDL-C) levels and PD incidence, with lower PD incidence rates in people with higher LDL-C levels.9-11 These findings have prompted further investigations on the influence of cholesterol and cholesterol-lowering medications on PD risk.12-16
CHOLESTEROL, STATINS, AND PARKINSON’S DISEASE
Statins (ie, HMG-CoA reductase inhibitors) have been widely used as cholesterol-lowering drugs for more than 15 years. They have been proved to be both effective and safe for preventive treatment of cardiovascular morbidity and mortality. The most common adverse effects associated with statin use are elevation of liver function tests and myopathy.17,18
Investigating the possible influence of statins on PD risk is of great importance not only for improving our understanding of PD etiology, but also for identifying additional potential effects of this widely used medication.
A number of the hypothesized mechanisms in the pathogenesis of PD may be influenced by blood cholesterol levels or by use of statins. Over the past few years, the biologic actions of statins have been a subject of extensive research. A recent review by Roy and Pahan19 summarizes some of the proposed mechanisms by which statins may affect the pathogenesis and progression of PD. Statins have an anti-inflammatory effect and have been shown to attenuate glial activation, inhibit oxidative stress, and protect dopaminergic neurons in animal models of PD.20 They also suppress the aggregation of α-synuclein protein, an important component in PD, as demonstrated in in vitro models.21
It has also been suggested that statins could affect PD via inhibition of HMG-CoA reductase.22 This pathway is shared by coenzyme Q10, known as ubiquinone, which is an important antioxidant and may play a role in PD. An unintended consequence of statins is lowering of coenzyme Q10 levels, thus potentially increasing the risk and worsening the course of PD.23 Another possible link between statins and cholesterol and PD is the apolipoprotein E e2 allele, which is associated with increased incidence of cases of sporadic PD24 and also related to lower levels of LDL-C in the serum.25
According to these various biologic theories, statins may either have a protective or a deleterious effect regarding PD. Empirical evidence is needed to determine their actual effect. Recently, a number of epidemiologic studies have examined the association between cholesterol levels, statin use, and PD incidence.9-16,26 However, these studies have revealed conflicting results with no clear conclusion.
While some studies found a decrease in PD incidence related to statin use,9,12,13,15 others have shown that differences in PD incidence were related to baseline cholesterol levels (ie, cholesterol metabolism playing a role in PD pathogenesis). 9,10,11,26 Most of the latter found an inverse association between serum cholesterol levels and PD incidence; however, one study showed opposite results.26
The conclusions regarding statins varied, with some studies suggesting they have a protective role in PD,9,10,12,13,15 and others not finding any significant association between statin use and PD.11,14,16 These inconsistent results are perhaps due to the diverse methods used and the different populations studied. Many of the studies mentioned the need for a large epidemiologic study to verify their results.
As statins are known to be very safe, cheap, and widely available, we believe that it is important to determine their effect on PD, because this might have practical implications for prevention of this debilitating disease. We set out to study the association between baseline LDL-C level, long-term statin use, and the incidence of PD using our large patient database.
This population-based historical cohort study was based on the computerized clinical database of Clalit Health Services, the largest health service in Israel. The database includes laboratory results, a central register of medical diagnoses (smoking, diabetes, hypertension, ischemic heart disease, cerebrovascular accidents [CVAs]), information on medication purchase, and sociodemographic details.27
The study population included all eligible people over the age of 45 years living in 1 administrative region in central Israel, between the years of 2000 and 2007. According to Israeli guidelines, LDL-C measurements are taken routinely as a standard screening for all males over the age of 35 years and all females over the age of 45 years.28
The date of entry into the study was defined as the date of the first measurement of LDL-C as documented in the patient’s file, between January 1, 2001, and December 31, 2005. This date was personal for each participant. Exclusion criteria were (1) existing PD as indicated by use of anti-parkinsonian medication prior to the study period (Table 1); (2) statin use before the commencement of the study; (3) neuroleptic drug use, a frequent cause of secondary parkinsonism (Table 2); (4) change of health insurance during the study; and (5) subjects with no recorded LDL-C value during the study period.
We applied these exclusion criteria to the study population during the year prior to study commencement (ie, January 1, 2000, to December 31, 2000) to ensure that participants had not been exposed to statins or diagnosed with PD before the study period.
PDF is available on the last page.