Plasma cholesterol and Parkinson's disease: Is the puzzle only apparent?
G. Scigliano,Gabriele Ronchetti,F. Girotti
DOI: https://doi.org/10.1002/mds.22626
IF: 9.698
2010-04-15
Movement Disorders
Abstract:Recent articles (two in Movement Disorders) have provided conflicting results regarding the association between plasma cholesterol levels and risk of Parkinson’s disease (PD). In a case-control study on 124 consecutive idiopathic PD patients, blood cholesterol was significantly lower in PD patients than controls. The authors hypothesized an etiological role of low cholesterol in the development of PD. Paradoxically, they also found that 17% of cases vs. 34% of controls used statins, and further suggested that statin use may have a neuroprotective effect against PD. It appears to us that the word ‘‘association’’ is being interpreted too strongly and that causes and effects are being confounded. Lower cholesterol levels and less frequent use of statins may be ‘‘associated’’ with PD, but this does not imply a causal relationship. In a similar case-control study, we found that PD patients had lower blood glucose and lipid levels, and lower frequency of diabetes and hypertension than controls: applying the logic of the authors cited above we would conclude that high glucose and lipids, presence of diabetes and hypertension, or even use of glucose-lowering and blood pressure-lowering drugs, were protective against PD. However, we thought it better to explain our results as an effect of the chronic illness which damages the autonomic system, resulting in reduced sympathetic activity with decrease of circulating levels of catecholamines and cortisol. Because physiologically these substances have diabetogenic and hypertensive effects and also increase plasma lipids, their reduction is likely to be an important cause of the reduced frequencies of diabetes and hypertension, and lowered glucose, cholesterol and triglycerides seen in PD patients. Nevertheless, because cholesterol is an essential constituent of neuronal cell membranes, it is possible that cholesterol homeostasis plays an etiological role in PD. Investigation of cholesterol levels before disease onset might elucidate such a relationship, because any cholesterol alterations at that stage are unlikely to be caused by autonomic damage. Large prospective population-based studies have examined the association between plasma cholesterol and incident PD. Some found increased risk of PD in persons with high total cholesterol at baseline, others found increased risk of PD in those with low cholesterol. In the study of Hu et al., PD cases were identified by use of anti-PD drugs reported in the National Insurance Register, which entitles patients to medications free of charge. The diagnosis of PD is often difficult, even for neurologists trained in extrapyramidal disorders, and typically requires follow-up or neuroradiological investigations for confirmation. Furthermore, free-of-charge drugs would not be denied patients with multiple system atrophy or vascular Parkinsonism, so the study may well have included such patients. Inadequate diagnostic criteria for PD may also have been used in the study of Huang et al., where PD cases were identified from hospital records, death certificates, and self-reported histories of parkinsonism (only the latter confirmed by a neurologist). The study, confined to men, found 41 incident cases of PD, with average age at onset 79.3 years (range 73–89). The mean age at onset of PD is approximately 60 years. Since parkinsonisms and other neurodegenerative disorders are increasingly frequent over 65 years, it is likely that the authors studied the association of cholesterol with parkinsonisms, and not idiopathic PD. In the well-conducted study of de Lau et al. the age of PD onset was not stated, but mean age at baseline was 69 years, and 87 new PD cases were diagnosed over the 9.4 years of follow-up. It is evident therefore that these patients were rather old, and late-onset PD could well differ from the common form of PD. The authors reported that high plasma cholesterol was associated with decreased risk of PD in women, but not men. Interestingly, they pointed out that: ‘‘as the intact blood-brain barrier is impermeable to cholesteroltransporting lipoproteins and most brain cholesterol is synthesized in situ, it is [. . .] unclear whether serum cholesterol levels reflect changes in cholesterol metabolism in the central nervous system.’’ We conclude that the available prospective populationbased studies fail to provide compelling evidence that altered cholesterol metabolism is involved in PD genesis, and that the hypothesis should either be abandoned or be investigated by approaches that include only carefully diagnosed cases of idiopathic PD.