Statins in Neuro-ophthalmology

ABSTRACT

Statins are effective and well-tolerated hypolipidaemic agents which have been increasingly studied for their pleiotropic immunomodulatory and anti-inflammatory effects. Statins have potential therapeutic benefit in a range of neuro-ophthalmological conditions but may also induce or exacerbate certain neurological disorders. This literature review examines evidence from clinical and in vitro studies assessing the effects of statins in myasthenia gravis, myopathy, multiple sclerosis, neuromyelitis optica, idiopathic intracranial hypertension (pseudotumour cerebri), migraine, giant cell arteritis, Bell’s palsy, ocular ischaemia, stroke, Alzheimer’s disease and Parkinson’s disease.

Introduction

Statins are a class of lipid-lowering medication that act by inhibiting the rate-limiting enzyme in cholesterol biosynthesis, 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase. Typically, statins are used as front-line therapeutic agents for both the primary and secondary prevention of coronary heart disease.¹ Statins have been proposed to exert pleiotropic effects by modulating immuno-inflammatory processes in a range of pathologies independent of low-density lipoprotein cholesterol levels. The immunomodulatory properties underlying the pleiotropic activity of statins have been investigated in a number of in vitro studies. Suppression of major histocompatibility complex class II expression, inhibition of leukocyte function antigen-1/intercellular adhesion molecule-1 interaction, and inhibition of CD40 (cluster of differentiation molecule) and its ligand signalling are among some of the non-lipid related functions that have been reviewed.²

Statins also inhibit the biosynthesis of isoprenoid intermediates in the mevalonate pathway. Isoprenoids serve as lipid attachments for intracellular proteins centrally involved in cell growth and signal transduction pathways.³ Inhibition of these processes alters the expression of endothelial nitric oxide synthase, the stability of atherosclerotic plaques, the production of pro-inflammatory cytokines and reactive oxygen species, the reactivity of platelets, and the development of cardiac hypertrophy and fibrosis.¹ These pathogenic mechanisms are associated with various cardiovascular as well as neurological disorders, and therefore numerous clinical and experimental studies have investigated the neuro-protective efficacy of statins.

However, statins have also been implicated in the induction or exacerbation of some diseases due to their effects upon the immune system, such as increasing the T-helper 2 (Th2) cell response. This excessive response leads to decreased B-cell suppression, which increases B-cell activity, causes loss of immune tolerance and produces pathogenic autoantibodies.³

Statins are generally well-tolerated with the most common adverse event being myalgia without biochemical evidence of muscle damage, which is reversible with discontinuation of statin use.⁴ Some individuals may develop more serious reactions, such as severe necrotising myopathy or rare but potentially fatal acute rhabdomyolysis, renal failure and myoglobinuria.⁴ A side effect of statins is that they may induce or exacerbate neuromuscular disorders. The risk factors for ocular and systemic side effects include increasing age, male gender, diabetes mellitus, renal impairment, cardiovascular disease, certain interacting drugs and mutations of the SLCO1B1 (Solute Carrier Organic Anion Transporter Family Member 1B1) gene, which encodes a transporter protein in the liver.⁴

Recently, the effects of statins in ophthalmology have been reviewed.^5,6 It was found that statins may have a role in reducing the burden of blepharitis-associated dry eye by reducing the excess cholesterol contamination of the tear film associated with meibomian gland and other accessory glands dysfunction. The severity of open angle glaucoma may also be reduced by statins, as blood flow through the optic nerve may be optimised with the lowering of serum cholesterol levels. In addition, statins may benefit in reducing the burden of diabetic retinopathy and age-related macular degeneration (ARMD) through not only cholesterol lowering but also their anti-inflammatory properties.

This review reports evidence for the role of statins in clinical practice, with respect to their cholesterol-lowering and pleiotropic effects, in a range of neuro-ophthalmological and associated conditions as a narrative synthesis of the literature.

Methods

Published studies investigating the effect of statins upon neuro-ophthalmological diseases and associated co-morbidities were searched in MedLine (Ovid), EMBASE (Ovid), Cochrane Library and Pubmed in September 2019. The search terms used were “statin” (Medical Subject Heading [MeSH]) or “Hydroxymethylglutaryl-CoA Reductase Inhibitors” (MeSH), “multiple sclerosis” (MeSH), “myasthenia gravis” (MeSH), “giant cell arteritis” (MeSH), “Alzheimer’s disease” (MeSH), “Parkinson’s disease” (MeSH), “migraine” (MesH), “Bells palsy” (MeSH), “extraocular muscle” (MeSH), “oculomotor muscles” (MeSH), “optic nerve” (MeSH), “depression” (MeSH), “amaurosis fugax” (MeSH), “stroke” (MeSH), “ischaemic stroke” (MeSH), “haemorrhagic stroke” (MeSH), “idiopathic intracranial hypertension” (MeSH), “central retinal artery occlusion” (MeSH), “neuromyelitis optica” (MeSH), “migraine” (MeSH), “retinal migraine” (MeSH) and “non-arteritic anterior ischaemic optic neuropathy” (MeSH).

Records were identified through database searching and screened for relevancy to neurological or ophthalmological conditions and statins. Articles from animal models and clinical studies that described the use of statins in neurological or ophthalmological conditions and associated co-morbidities were included. Additionally, in vitro studies reporting the effect of statins upon biochemical markers related to neurological or ophthalmological conditions were included. Articles published in languages other than English or were not available as a full text were excluded. Additional articles identified through references lists of searched articles were also screened for relevancy and included in this review.

Results

Myasthenia gravis

Myasthenia gravis (MG) is a chronic autoimmune disorder of neuromuscular transmission with antibodies directed towards the nicotinic acetylcholine receptor (AChR) of the neuromuscular junction.⁷ It is characterised by fatiguable muscle weakness that improves with rest. It has an annual incidence of up to five per 100,000 people with various subtypes.⁸ The age of disease onset in MG has a bimodal pattern, with one peak at 20–30 years (particularly for white females) and the other at 60–70 years.⁹ Ocular MG, an early disease manifestation in approximately 50% of patients with MG, is clinically isolated to the extra-ocular, levator palpebrae and orbicularis oculi muscles, resulting in variable ptosis and diplopia.⁸

In a recent disproportionality analysis of 184,284 reported cases of adverse drug reactions related to HMG CoA reductase inhibitors from a World Health Organisation associated pharmacovigilance global database, there were 169 suspected cases of statin-induced MG from a total of 3967 reports mentioning MG. A weak drug safety signal was found linking the occurrence of MG and the use of statins in general (reporting odds ratio [OR] 2.66, 95% confidence intervals [CI] 2.28–3.10).¹⁰ One case-control study (n = 84), however, found no evidence to suggest statin usage induces MG (OR 1.70, 95% CI 0.60–4.87, p = .32). It was noted, however, that insufficient sample size may have been a limitation and larger sample size studies were suggested.¹¹

There are case reports documenting statin-induced ocular MG including de novo seropositive cases and disease relapse from long-term remission.^12–16 The supporting evidence is based on myasthenic symptoms with ocular involvement that were temporally associated with the initiation, re-commencement, dose adjustment or discontinuation of statin therapy.^{12,14,16–19} Across case reports, the time to clinical manifestation of ocular signs ranged from one week to three months.^15,18 The time to partial or complete resolution following cessation of statin therapy varied substantially, from between a few days to three months and may have been due to differences in immunosuppressive therapy.¹⁵ There were two cases of complete resolution following discontinuation of statin therapy in combination with prednisone therapy and two cases of recurrence following retrial of a statin.^15,18 The limitation of evidence of this nature is that it is difficult to distinguish between the effects that statin cessation and immunosuppressive therapy have upon clinical improvement of myasthenic symptoms.

Whilst various theories exist explaining how statins may induce MG or a myasthenic syndrome, the exact pathogenesis remains unclear.¹³ One mechanism is thought to be related to the immunomodulatory properties of statins. Experimental animal studies have demonstrated that HMG-CoA reductase inhibitors upregulate Th2 cell secretion of cytokines, such as interleukin (IL)-4 and IL-10, which stimulate the anti-AChR response in MG.^20,21 This has been observed in several case reports of increased AChR antibody titre in patients experiencing myasthenia-like symptoms that were temporally-associated with statin usage.^3,19,22 The immunomodulatory theory is also supported by cases of other statin-induced autoimmune disorders such as dermatomyositis, polymyalgia rheumatica and a lupus-like syndrome.¹⁷ Alternatively, it has been proposed that statin myotoxicity may exacerbate or unmask subclinical weakness in patients with underlying MG, as observed in other myopathies and motor neuropathies such as McArdle’s disease and sporadic rippling muscle disease.^23,24 In patients with pre-existing MG, exacerbation of symptoms or development of a myalgic syndrome has been associated with statin therapy commencement in 11–25% of patients.^11,19

In summary, although there is one small case-control study that did not show a significant association between statin usage and MG onset, pharmacovigilance data suggest that statins may induce or exacerbate MG. The exact pathogenic mechanism remains largely debated. Patients with MG requiring statins should be monitored for possible worsening of their symptoms.

Myopathy

Statin-induced or -associated myopathy (SAM) is defined as the occurrence of muscle symptoms coinciding with a rise in serum creatine kinase greater than 10 times the upper limit of normal.²⁵ Potential involvement of ocular muscles in SAM mimicking MG was first reported in India in a middle-aged woman on simvastatin for six months who presented with pupil-sparing unilateral ptosis, painless external ophthalmoplegia and had negative results in her investigations for MG. Simvastatin cessation resulted in normalisation of symptoms within two months.²⁶ This was differentiated from statin-induced orbital myositis, which the authors have previously reviewed, where pain is a feature.⁵

Several mechanisms have been postulated including cholesterol depletion altering membrane fluidity causing membrane hyperexcitability.²⁷ Statins may impair calcium homoeostasis increasing intracytosolic calcium causing mitochondrial depolarisation.²⁷ Elevated intracytosolic calcium can also trigger apoptosis of skeletal muscles through activation of the mitochondrial-mediated apoptotic signalling cascade.²⁸ Another theory is that statins may cause mitochondrial dysfunction by depleting endogenous coenzyme Q₁₀ (CoQ₁₀) due to inhibition of the mevalonate pathway.²⁹ This effect maybe most likely a function of lipoprotein reduction as these proteins serve as carriers for CoQ₁₀.³⁰ CoQ₁₀ is a key component in mitochondrial bioenergy transfer, facilitating electron transfer in the generation of ATP. Deficiency of CoQ₁₀ has been reported in muscle biopsies of patients with isolated myopathy, hence suggesting a possible mechanism for statin-induced myopathies.³¹ In a recent randomised double-blind trial (n = 105), the intensity of statin adverse effects were significantly reduced in the group with the addition of CoQ₁₀, including a significant reduction in hepatic enzymes activity and inflammatory markers as compared with a control group of stain usage without CoQ₁₀ for the treatment of dyslipidaemia and elevated triacylglyceride levels.³² Further studies investigating the use of CoQ₁₀ supplements concurrently with statins in various conditions are needed as they confer potential benefit with low risk of toxicity.³²

Bell’s palsy

Bell’s palsy is an acute unilateral facial nerve palsy with an estimated global incidence of 15 to 30 per 100,000 persons annually and can affect patients of all ages.³³ Statins may affect the incidence of Bell’s palsy by inducing physical neuronal changes. Several studies have reported the negative impact of statins upon neurite outgrowth and maintenance, oligodendrocyte and myelin formation, and vulnerability to demyelination induced by viral infections as seen in Bell’s palsy.^34–37 Statins may also stimulate pro-inflammatory responses and trigger autoimmunity, as previously reported in cases of statin-induced autoimmune disorders.¹⁷

There are, however, limited studies investigating the association of statin therapy with the incidence of Bell’s palsy. A single Taiwanese case-control study (n = 7908) compared subjects Bell’s palsy patients (n = 1977) with sex- and age-matched control subjects (n = 5931).³⁸ After adjusting for diabetes mellitus, hypertension and hyperlipidaemia, previous regular statin use (defined as use ≥ 60 days within six months prior to the index date) was found to be significantly associated with patients with Bell’s palsy (OR 1.46, 95% CI 1.28–1.67). There was no increased adjusted OR of irregular statin use for subjects with Bell’s palsy compared with controls (OR 1.09, 95% CI 0.82–1.46). However, as the data were sourced from a population-based national health insurance database, there are various limitations which impact interpretation of these findings. Uncertain reliability and incomplete data such as statin dose, demographic information and medical history limits the ability to adjust for covariates such as genetic predispositions and other co-morbidities that may predispose to Bell’s palsy. Furthermore, it was not possible to assess a dose-dependent relationship.³⁸ Further studies are needed to establish whether statin neurotoxicity, which has been documented to cause peripheral neuropathy, adversely impacts the central nervous system (CNS).

In summary, there is little evidence for the impact of statin use upon the incidence of Bell’s palsy and further studies are required to investigate this potentially detrimental effect.

Multiple sclerosis (including optic neuritis)

Multiple sclerosis (MS) is an autoimmune demyelinating disorder of the CNS that often manifests with afferent and efferent visual pathway involvement with optic neuritis being the most common ocular manifestation. Additionally, MS may be associated with ocular inflammatory disease in the form of pars planitis and retinal periphlebitis.³⁹

Clinical trials in MS have indicated that statins exert pleiotropic immunomodulatory effects by inhibiting the biosynthesis of isoprenoids, resulting in a decrease of autoimmune responses.^40,41 However, limited evidence supports statin therapy in reducing visual symptoms associated with MS. A double-blind, randomised, controlled trial (RCT) (n = 64) found treatment of acute optic neuritis with simvastatin 80 mg per day for six months had a significant effect on decreasing visual evoked potential (VEP) latency (p = .0132) and increasing amplitude (p = .0132). The study also found statins slightly improved contrast sensitivity as assessed by Arden gratings (p = .0572), colour perception as assessed by Velhagen pseudoisochromatic plates (p = .0531) and significantly improved self-evaluated visual function as assessed by visual analogue scales (p = .0391).⁴² Post-hoc analysis of a double-blind RCT (n = 140) of patients with secondary progressive MS found high dose simvastatin (80 mg) had a positive effect, independent of serum cholesterol levels, upon disability as measured by the Expanded Disability Status Scale (EDSS) and brain atrophy measured on brain magnetic resonance imaging (MRI).^43,44 A weaker simvastatin effect on visuospatial memory measured using a block design test was mediated by slowing brain atrophy rates.⁴⁴

Various combinations of statins with immunomodulatory therapies for MS have demonstrated additive anti-inflammatory effects in a range of studies measuring inflammatory and immunological biomarkers. Low doses of simvastatin, lovastatin and mevastatin have been demonstrated in vitro to inhibit proliferation of peripheral blood mononuclear cells of patients with relapsing-remitting MS (n = 60) and secondary progressive MS (n = 14).⁴⁰ The effects were observed in a dose-dependent manner with highly lipophilic simvastatin and lovastatin demonstrating the greatest potency, as well as exerting synergistic effects when combined with interferon (IFN) β-1b.

The potency of lipophilic statins is related to their ability to cross the blood brain barrier.⁴⁵ Adjunctive use of atorvastatin has been shown in a RCT (n = 28) to significantly (p < .05) lower serum levels of high-sensitivity C-reactive protein, which are elevated in relapsing-remitting MS patients treated with IFN β-1b.⁴⁶ In another RCT (n = 38), the combination of atorvastatin and methylprednisolone significantly increased anti-inflammatory cytokines IL-13, IL-10, IL-35 (p < .05) whilst lowering inflammatory IFN-γ (p = .001).⁴⁷ A recent double-blind RCT (n = 95) also found levels of IL-10, transforming growth factor (TGF)-β and IL-4 increased significantly after 18 months of atorvastatin 40 mg treatment combined with IFN β-1a (p = .0001).⁴⁸ Furthermore, levels of tumour necrosis factor (TNF)-α and IL-17 decreased significantly compared to the baseline level (p = .0001) but levels of IFN-γ were unaffected.

The impact of adjunctive statin therapy upon MRI outcomes such as cerebral atrophy and quantity of new, enlarging or contrast-enhancing lesions has also been reported extensively with varying results. In patients with secondary progressive MS, a multi-centre, double-blind RCT (n = 140) showed that high dose (80 mg) simvastatin significantly reduced the annualised rate of whole-brain atrophy when compared to placebo by 43%.⁴³ In patients with relapsing-remitting MS, a double-blind RCT (n = 154) demonstrated atorvastatin 40 mg as add-on therapy to IFN β-1b did not significantly change brain atrophy rates when compared to placebo. However, an open-label clinical trial (n = 30) demonstrated that treatment with high dose simvastatin significantly reduced the number and volume of gadolinium-enhancing (Gd) lesions by 44%, (p < .0001) and 41% (p = .0018) respectively.⁴¹ Further trials, however, were unable to replicate such results. In one double-blind RCT (n = 85) of patients with relapsing-remitting MS on IFN β-1a therapy, adjunctive treatment with a lower 40 mg dose of simvastatin only demonstrated a decreasing trend in lesions on MRI (p = .62).⁴⁹ Post-hoc analysis of a large, multi-centre RCT (n = 582) in which relapsing-remitting MS patients (n = 40) on IFN β-1a therapy were also treated with either atorvastatin (n = 26) or simvastatin (n = 13) showed no significant difference in the number of Gd-enhancing lesions (p = .604), or number of new or enlarging T2-hyperintense lesions (p = .802) on MRI at two years.⁵⁰ This result was also observed in a double-blind RCT (n = 95), which found no significant difference in the number of Gd-enhancing lesions in patients receiving combined atorvastatin 40 mg and IFN β-1a treatment.⁴⁸ Similarly, in another RCT of relapsing-remitting MS patients (n = 307) treated with IFN β-1a, co-medication with high dose simvastatin did not yield any difference in the mean number of new or enlarging T2-weighted lesions from baseline to 12 months.⁵¹

In contrast to other studies, one double-blind RCT of relapsing-remitting MS patients (n = 26) on IFN β-1a therapy found the majority of patients on either atorvastatin 40 mg or 80 mg for six months developed either new T2 or Gd-enhancing lesions on brain MRI (hazard ratio [HR] 8.25, 95% CI 1.06–64.2, p = .044).⁵² A definite dose response difference in terms of changes in disease activity between patients on either low or high dosages of atorvastatin could not, however, be found.

In summary, although there is a relative paucity of studies reporting the effect of statins on the visual symptoms of MS such as in optic neuritis, the literature suggests that there may be an overall benefit to MS patients’ biochemical inflammatory marker profile and radiological outcome. Statins also appear to be beneficial as monotherapy or in combination with standard immunotherapies for MS. Further clinical trials of statins with high lipophilicity (such as simvastatin and lovastatin) and possibly with higher doses may, therefore, help delineate their efficacy before recommendations can be provided.

Neuromyelitis optica

Neuromyelitis optica (NMO) is an inflammatory disorder predominantly resulting in attacks of optic neuritis and myelitis, secondary to pathogenic autoantibodies to aquaporin-4 (NMO-IgG or AQP4-Ab). AQP4-Ab are anti-astrocytic serum autoantibodies, which target the most abundant water channel protein in the CNS.⁵³ This immune response results in astrocyte loss and the deposition of immunoglobulins and complement proteins followed by infiltration of inflammatory cells, cytokine release, oligodendrocyte death and demyelination.⁵⁴ The mainstay of NMO treatment involves the usage of steroids and steroid-sparing immunotherapies. Currently, no clinical trials have focused on the usage of statins in NMO. The potential statin pleiotropic benefits in NMO have, however, recently been investigated in vitro in regard to the vasculocentric deposition of activated complement in NMO-affected human tissues. Atorvastatin, simvastatin, lovastatin and fluvastatin have been shown to increase the major complement regulator protein CD55 expression in primary cultures. Furthermore, oral atorvastatin at 10–20 mg/kg/day for three days has been shown to strongly upregulate CD55 immunofluorescence and reduce NMO pathology in the mouse brain and spinal cord post-intracerebral AQP4-Ab injection.⁵⁵

Given the anti-inflammatory effects of statins, they may be well-placed to trial alone or potentially in combination with other NMO therapies.⁵⁵ However, caution should be taken, given their effects on cholesterol biosynthesis and hence potential for inhibition of remyelination.³⁶ Furthermore, simvastatin has been demonstrated to downregulate the expression of AQP4 in ischaemic stroke rats which may reduce the brain oedema associated with stroke. This may be beneficial in cerebral ischaemia but may be detrimental in NMO.⁵⁶

Idiopathic intracranial hypertension

Idiopathic intracranial hypertension (IIH) is a disorder of unknown aetiology affecting primarily obese women of childbearing age.⁵⁷ Obesity plays a central role given that weight loss alleviates its course.⁵⁸ Nearly all systemic complications of obesity such as hypercholesterolaemia have been linked to visceral (central, abdominal or truncal) adiposity, on the assumption of the unfavourable fat cell metabolic features comprising the visceral depot.⁵⁹ In female patients with IIH, fat distribution is shifted to the lower body in a gynaecoid pattern compared with obese women of a similar age.⁶⁰ In accordance with this difference, the overall cardiac burden, including dyslipidaemia was also less. Total cholesterol and triglycerides levels were significantly increased (p < .001) in the reference cohort of obese women compared with the IIH cohort who were mostly within the normal range.⁶⁰

A recent epidemiological study using 28 years of data, however, has revealed that in a population-based match controlled cohort study matched by body mass index and age, women with IIH appear to have a 55% increase in risk of hypertension, a 30% increase in risk of diabetes mellitus and an increased risk of composite cardiovascular events such as heart failure, ischaemic heart disease and stroke. In this study it was concluded that given IIH in women was associated with a two-fold increase in cardiovascular disease risk, modifying risk factors may reduce long-term morbidity in this cohort.⁶¹

As dyslipidaemia does not appear to be a significant characteristic in the make-up of IIH patients, statin usage would not appear to play a role in the routine management of IIH. Moreover, a comprehensive review of papilloedema denotes no significant inflammatory component amenable to the pleiotropic effects of statins.⁶² Further prospective longitudinal studies may be warranted, especially in those IIH patients with borderline dyslipidaemia where statins may be beneficial.⁶³ This is especially the case given that IIH patients are typically identified at a young age and there may be an opportunity to address modifiable risk factors to improve long-term cardiometabolic outcomes.⁶¹

Migraine

Migraine is a chronic neurological disorder characterised by paroxysmal attacks of head pain with reversible neurological and systemic symptoms. It has an estimated global prevalence of 12%.^64,65 Visual aura occurs in over 90% of patients with aura as spark photopsias, scotoma or teichopsia in a hemifield, with other less diagnostic visual symptoms such as shimmering, undulations or so-called heatwaves also described by patients.⁶⁴ The phenomenon arises from cortical spreading depression that may activate perivascular nerve afferents, leading to vasodilatation and neurogenic inflammation of the meningeal blood vessels. Epidemiological evidence suggests that migraineurs have an unfavourable cardiovascular risk profile with elevated levels of total cholesterol and an increased ratio of total/high-density lipoprotein C.⁶⁶ Among patients with migraine, young women and those with typical aura with migraine are at an increased risk of early-onset (<45 years of age) ischaemic stroke.^67,68 Apart from their lipid-lowering ability, the anti-inflammatory effects of statins may help regulate endothelial dysfunction, which has been implicated in the pathogenesis of migraine, by reducing oxidative stress and increasing nitric oxide bioavailability, thereby aiding vasodilatation.^69–74 This may particularly benefit patients with typical aura with migraine as there is an impairment in the adaptive cerebral haemodynamic mechanisms, which may respond to statins.⁶⁴

Preliminary studies supporting statin use suggested that three months of therapy can be potentially efficacious in migraine prophylaxis or result in complete resolution and long term remission, as reported in one isolated case.^75,76 In an open-label parallel group study of female patients, propranolol 60 mg daily (n = 25) yielded a comparable responder rate of 88% to simvastatin 20 mg daily (n = 29) (83%, p = .7112).⁷⁵ Similarly, a more recent RCT found atorvastatin 40 mg daily (n = 46) demonstrated comparable efficacy and superior safety to sodium valproate 500 mg daily (n = 36) (responder rate 65% vs. 72% respectively, p = .499) following three months of treatment.⁶⁵ Animal studies have also suggested statins may have a role in migraine prophylaxis by attenuating activation of nuclear factor-κB in the trigeminal nucleus caudalis, which is believed to be partly involved in the pathogenesis of migraine.⁷⁷

The potential benefit of high serum vitamin D levels has also been studied in association with statin therapy in migraine. Protection against vitamin D deficiency in the context of statin use may provide a synergistic anti-inflammatory effect and improve endothelial dysfunction whilst also reducing the risk of statin-associated musculoskeletal pain. A cross-sectional population based study (n = 5938) and subsequent double-blind RCT (n = 57) found that both statin users (OR 0.67, 95% CI 0.46–0.98, p = .04) and statin users with higher serum vitamin D levels (25(OH)D > 57 nmol/l) (OR 0.48, 95% CI 0.32–0.71, p = .001) were significantly associated with lower odds of having severe headache or migraine.^78,79 In patients treated with simvastatin 20 mg twice daily and vitamin D3 1000 international units, 25% of the cohort experienced a 50% reduction (responder rate) in the number of migraine days from baseline at 12 weeks, which increased to a responder rate of 29% at 24 weeks. This was significantly different from the placebo arm where the responder rate was 3% at both 12 and 24 weeks (p = .03).

Conversely, statin therapy itself has also been implicated in the pathogenesis of migraine by induction of mitochondrial dysfunction, given their ability to reduce circulating CoQ₁₀ concentrations, resulting in impaired oxygen metabolism.^80,81 The theory is supported by reported benefits in clinical trials investigating CoQ₁₀, an established adjunctive agent in the treatment of mitochondrial disorders, as a preventative therapy for migraine. Statistically significant reductions in migraine attack frequency of up to 55.3% from baseline have been reported in both an open-label trial (n = 31) and a double-blind RCT (n = 43) following three months therapy with CoQ₁₀ at daily doses of 300 mg (p < .001) and 150 mg (p = .01) respectively.^82,83 The average number of days with migraine also significantly decreased with CoQ₁₀ use in both studies (p < .001 and p = .04 respectively).^82,83 The responder rate ranged from 47.6% to 61.3% for CoQ₁₀ treated groups and was 14.4% for patients receiving placebo.^82,83

In summary, statin therapy alone has not been found to significantly reduce migraine symptoms but there is emerging evidence of a benefit with respect to prophylaxis. Further studies are needed to investigate whether statins alone or supplemented with vitamin D or CoQ₁₀ may have a role in this prophylaxis.

Giant cell arteritis

Giant cell arteritis (GCA), or temporal arteritis, is a chronic granulomatous vasculitis of medium- and large-sized vessels that primarily affects patients aged over 50 years of age.⁸⁴ Prompt initiation of high-dose corticosteroids is the mainstay of treatment with subsequent low-dose corticosteroid maintenance therapy for often over one year to control disease activity.^84–86 Long-term corticosteroid therapy has been associated with the development of dyslipidaemia by inducing insulin resistance, increasing the hepatic synthesis of very low-density lipoproteins and triglycerides, enhancing the activity of HMG-CoA reductase, and inhibiting lipoprotein lipase.⁸⁷ Hence, patients receiving corticosteroid therapy are often considered for lipid-lowering agents such as statins. Furthermore, cardiovascular risk in systemic vasculitis is suggested to be higher than in the normal population. A matched-controlled study (n = 315) found males with GCA were associated with an increased risk of a cardiovascular event (OR 2.04, CI 95% 1.06–3.95, p = .03), however, treatment with statins was significantly protective (OR 0.38, CI 95% 0.18–0.80, p = .01).⁸⁸

Statins may provide benefit by modulating inflammatory and immune responses, which are now recognised to have a pathogenic role in atherosclerotic cardiovascular disease.^89,90 Clinical studies demonstrate that statins reduce circulating C-reactive protein, pro-inflammatory cytokines (IL-1 and IL-6), TNF-α and cell adhesion molecule levels in patients with cardiovascular risk factors.^91–93 In patients with GCA, statins have also been shown to lower erythrocyte sedimentation rate and downregulate key inflammatory mediators (IL-17 and IL-16).^94,95

Statin therapy, however, does not appear to impact vision-related outcomes in patients with GCA. Various retrospective chart reviews and case-controlled studies comparing statin-users with non-users among a cohort of GCA patients did not find evidence for statistically significant reductions in incidence of acute visual ischaemic complications (see Table 1).^87,96–99 One retrospective study (n = 120) of patients with biopsy-proven GCA found a statistically significant association of cranial ischaemic complications in statin-users (p = .004). Statin usage here, however, was described as perhaps a surrogate marker for associated increased cardiovascular risk for the development of cranial ischaemic complications in this cohort of patients.¹⁰⁰ There is one reported case of GCA with polymyalgia rheumatica, which developed in a 78-year-old female with hypercholesterolaemia and hypertension following the commencement of statin therapy. Her symptoms, which resolved with discontinuation of statin-therapy and recurred with the resumption of statin-therapy, suggested a temporal association between disease activity and statin exposure may exist and that the GCA was statin-induced.¹⁰¹ The authors did acknowledge, however, that there was no clear explanatory mechanism. Statin-induced myalgias may have worsened polymyalgia rheumatica symptoms on re-challenge and the concurrent GCA development may have been coincidental.

Table 1.

The association of statin therapy upon vision related outcomes in patients with giant cell arteritis

Study	Location	Study design	Cohort	Statin(s)	Period	Findings
Garcia-Martinez 2004	Spain	Retrospective case-control study	54	Simvastatin (n = 7) Lovastatin (n = 3) Atorvastatin (n = 4) Pravastatin (n = 3)	2.8 years (mean)	Statin users (n = 17) vs. non-users (n = 37) Permanent visual loss: 5.8% vs. 16.2% Diplopia: 5.8% vs. 10.8%
Ma 2016	Canada	Non-randomised, retrospective chart review	137	Multiple (n = 52)	13 years	Anterior ischaemic optic neuropathy: RR 1.16, p > .05 Visual loss: RR = 0.95, p > .05
Narvaez 2007	Spain	Non-randomised, retrospective, pilot, case-control study	121	Simvastatin (n = 10) Atorvastatin (n = 12) Lovastatin (n = 5) Pravastatin (n = 3)	3.6 years (mean)	Statin users (n = 30) vs. non-users (n = 91) Visual manifestations: n = 8 (27%) vs. n = 14 (15%), p = .16 Diplopia: n = 1 (3%) vs. n = 1 (1%), p = .43 Transient visual loss: n = 6 (20%) vs. n = 10 (11%), p = .27 Permanent blindness: n = 1 (3%) vs. n = 3 (3%), p = .99
Olivas 2015	Spain	Retrospective chart review	120	Multiple (n = 21)	13 years	(visual loss or ischaemic stroke) Visual loss (n = 18) Stroke (n = 11) Both (n = 2) Statin users with CIC (n = 10) vs. statin users without CIC (n = 11), p = .004
Schmidt 2013	United States of America	Retrospective case-control chart review	594	Common: Simvastatin, atorvastatin Uncommon: Pravastatin, lovastatin, rosuvastatin, fluvastatin and cerivastatin	11 years	Statin users (n = 54) vs. non-users (n = 243) Visual symptoms: n = 14 (26.92%) vs. n = 68 (28.22%), p = .85

Key: CIC = cranial ischaemic complications, RR = relative risk.

The adjunctive use of statins with corticosteroids does not impact disease course of GCA or time to achieve a maintenance low dose of corticosteroids.^87,98,99 One longitudinal analysis of patients with GCA reported no statistically significant association between statin exposure and the ability to achieve a corticosteroid dose of <10 mg/day (HR 0.87, 95% CI 0.61–1.26, p = .87).⁹⁸ Similarly, a French population-based cohort study (n = 103) also did not find any influence of statin therapy on overall cumulative prednisone requirement or occurrence of GCA.¹⁰² It did, however, demonstrate a modest effect of baseline statin exposure up to 20 months on the likelihood of achieving remission maintenance with low-dose prednisone (HR 1.9).

In summary, statin medications are not considered requisite adjuncts in the management of GCA.⁸⁴ There is also insufficient evidence at this time to indicate that statin usage in GCA is harmful.

Ocular ischaemia

Retinal artery occlusion

Currently, there is a paucity in the literature with respect to the role of statins in central retinal artery occlusion (CRAO). The mechanisms by which statins may positively affect the retinal circulation in hypercholesterolaemia and ischaemia have been reviewed previously by the authors.⁵ Fasting lipid levels have been advocated for in CRAO patients with respect to risk for stroke.¹⁰³ In a meta-analysis of the risk of stroke in CRAO patients it was concluded that 30% of patients with acute CRAO and 29% of patients with acute branch retinal artery occlusion presented with acute cerebral ischaemia on MRI but with the majority silent.¹⁰⁴

Non-arteritic anterior ischaemic optic neuropathy

In non-arteritic anterior ischaemic neuropathy (NAION) the preventative role of statins is currently not clear. Given that NAION is mostly caused by arterial hypoperfusion of the posterior ciliary artery supplying the optic nerve head it may be considered different to a stroke, which is a thromboembolic phenomenon.¹⁰⁵ Although NAION and stroke share common risk factors, including hyperlipidaemia, the association of NAION and the risk for subsequent stroke has been reviewed as controversial with the latest population study demonstrating no increased stroke risk post-NAION in a Korean population.¹⁰⁵ There is also no link between hyperlipidaemia and the risk of fellow eye involvement in NAION.¹⁰⁶

Cerebral ischaemia

The role of statins in stroke has been well characterised in several recent reviews. Statins have a role in primary and secondary prevention of stroke as well as in the acute phase of ischaemic stroke and transient ischaemic attack (TIA).¹⁰⁷ Decreased cerebral perfusion can result in decreased visual acuity, visual field loss, ocular motility abnormalities and visuospatial perception defects. Prechiasmal vision loss can be caused by retinal ischaemia secondary to occlusion within the ophthalmic artery vascular supply, which is included in the definition of TIA.^108,109 Retinal ischaemia can occur either transiently as amaurosis fugax or permanently due to a branch or CRAO or more rarely an ophthalmic artery occlusion.¹¹⁰ Preceding symptoms of non-visual strokes often include transient monocular vision loss.¹⁰⁸ A retrospective study has concluded that acute brain infarction occurred in 23% of 213 patients with branch or CRAO, a significant finding given that vision loss was the only presenting symptom in 90% of these patients.¹⁰⁸ Transient binocular visual loss due to vertebro-basilar ischaemia is a warning sign for stroke.¹¹⁰ Isolated third, fourth or sixth cranial nerve palsy patients are also at increased risk for ischaemic stroke, suggesting a commonality of microvascular disease which needs further confirmation and investigation.¹¹¹

In the amaurosis fugax and TIA literature, there is one population study that suggests that in patients without cardiovascular disease, atorvastatin 10 or 20 mg was associated with a significantly lower risk of cardiovascular events compared with simvastatin 20 or 40 mg. Further mechanistic studies were suggested to establish whether atorvastatin should be used instead of simvastatin in this setting.¹¹²

Primary stroke prevention

In the primary prevention of stroke, in populations at higher risk of stroke, such as those with established coronary heart disease, the majority of relevant studies have shown a beneficial effect of statins with respect to their lipid-lowering in ischaemic stroke.¹⁰⁷ The observational Multiple Risk Factor Intervention Trial (MRFIT) in 350,977 men demonstrated that blood cholesterol levels have a positive correlation with death from non-haemorrhagic stroke.¹¹³ In patients with risk factors but no established cardiovascular disease, potent statins such as atorvastatin and rosuvastatin have shown some benefits, but the clinical relevance is reportedly questionable.¹⁰⁷ There are also increased risks of myopathy and hepatic dysfunction but atorvastatin had the best safety profile as reported in a large meta-analysis, which supported the efficacy of atorvastatin and rosuvastatin.¹¹⁴

Acute ischaemic stroke

In acute ischaemic stroke, a review of the literature has concluded that in those studies with a large sample size (≥1000), most found that statin use was associated with a good functional outcome.¹¹⁵ It has been reviewed that statin use pre-stroke is also associated with beneficial effects in terms of severity as well as short- and long-term mortality.¹¹⁵ Benefit was increased with good adherence and withdrawal associated with worse functional outcomes.¹¹⁵ It has been reported that pre-stroke statin use in patients is associated with a smaller infarction size, more collaterals and greater early reperfusion.^116–118 In one study more pronounced benefit was seen with high-dose statins (atorvastatin 80 mg or rosuvastatin 40 mg daily).¹¹⁹

Secondary stroke prevention

Secondary prevention of stroke is crucial as patients with a history of a cerebrovascular event (stroke or TIA) have a significant risk of recurrent stroke as well as other cardiovascular events.¹⁰⁷ The beneficial effects of statin therapy commenced after ischaemic stroke during hospitalisation have been reviewed with good functional outcomes demonstrated in several studies but the need for a definitive prospective RCT has been emphasised.¹¹⁵ In regards to survival, several observational studies have been found to demonstrate improvement at discharge or at 90 days after stroke.¹¹⁵ This has been confirmed with meta-analyses.¹¹⁵ In one systematic review and network meta-analysis of nine trials it was concluded that differences in effects among statins were modest, signalling potential therapeutic equivalence.¹²⁰

Haemorrhagic stroke

With respect to haemorrhagic stroke, statins have been reviewed as previously being documented with an increased risk of intracerebral bleeding potentially due to their pleiotropic anti-thrombotic activity through inhibition of platelet aggregation.^107,115 Furthermore, it has been reported that statin use is associated with an increased intracranial haemorrhage (ICH) volume in those with a spontaneous bleed.¹²¹ Several meta-analyses, however, have been recorded as being not demonstrative of an increased risk with statins.¹¹⁵ Additionally, in one study the risk was deemed to be counterbalanced by the major and significant reduction in the risk of ischaemic stroke.¹²² Growing evidence suggests that statin use before ICH, at the onset of ICH or after ICH maybe associated with favourable functional outcomes for ICH at up to six months.¹¹⁵ The corollary is that discontinuation of statin use has been associated with worsened outcomes.¹¹⁵

Mechanisms of action of the statins in stroke reduction extend beyond their ability to lower low-density-lipoprotein-cholesterol as set out in recent reviews.^1,123 These pleiotropic effects, which are isoprenoid inhibition derived, include upregulation of endothelial nitric oxide expression, atherosclerotic plaque stabilisation, reduction of platelet activation and recruitment, pro-inflammatory cytokine as well as matrix metalloproteinase and reactive oxygen species reduction, and possible minimisation of cardiac fibrosis and hypertrophy development.^1,123 Recently, it has been suggested that statins also have anti-arrhythmic effects.¹²³

Statin treatment is effective in both primary and secondary prevention of cerebrovascular events. Patients should receive statins to avoid cerebrovascular events, but they should be used with caution in the elderly who may present with side effects more often than a younger cohort because of the possibility of multiple co-morbidities and drug interactions, as per a recent review paper.¹²⁴

Further studies are needed to investigate the usage of statins for retinal artery occlusion and NAION patients, where appropriate, in terms of risk reduction for stroke because of their additional pleiotropic effects and not just on their effects on dyslipidaemia.

Alzheimer’s disease

Alzheimer’s disease (AD) is an age-related neurodegenerative disorder characterised by progressive memory loss, inability to perform activities of daily living and personality changes.¹²⁵ Examination of the visual system in AD may reveal visual field deficits, prolonged VEP, depressed contrast sensitivities, problems with colour and stereoacuity as well as abnormal eye movement recordings.¹²⁶ Complex visual disturbances include constructional and visuoperceptual abnormalities, spatial agnosia and Balint’s syndrome, environmental disorientation, visual agnosia, facial identification problems, and visual hallucinations.^127,128 Hypercholesterolaemia is postulated to increase the risk of developing AD in midlife and has therefore been examined as a potentially modifiable risk factor.^129–131

Deposition of extracellular plaques containing amyloid-β (Aβ) and the accumulation of intracellular neurofibrillary tangles containing hyperphosphorylated tau protein in the brain parenchyma are pathological hallmarks of AD.¹³² Drusen in ARMD also contain Aβ. Amyloid-β is metabolised from the amyloid precursor protein predominantly into isoforms Aβ40 and Aβ42, which are frequently measured as biomarkers of AD in serum and cerebrospinal fluid (CSF). Serum Aβ1-42 and the Aβ1-42/Aβ1-40 ratio are also sensitive predictors of progression of mild cognitive impairment to AD in an at-risk person.¹³³ Plasma Aβ, however, is not supported as a prognostic factor or correlate of cognitive change.¹³⁴ Both AD and ARMD are also associated with an upregulated inflammatory response, as can be evidenced by the high concentrations of complement factors in both pathologies which are being targeted therapeutically.¹³⁵

Clinical trials demonstrate that serum and CSF levels of Aβ and phosphorylated tau are reduced with statins. A dose-dependent and statistically significant reduction of serum Aβ was demonstrated with lovastatin (p = .0348).¹³⁶ In patients with mild-moderate AD, high dose (80 mg) simvastatin slightly decreased CSF Aβ40.¹³⁷ However, in patients with late-onset AD, probable AD or normal cognition, lower dose (20–40 mg) simvastatin had no significant effect upon CSF levels of Aβ40 or Aβ42 or total tau.^138–141 In non-demented, moderately hypercholesterolaemic adult subjects, low dose simvastatin but not pravastatin, modestly reduced CSF levels of phospho-tau181, likely due to its greater penetrance of the CNS.¹⁴⁰ Lipophilic statins (lovastatin and simvastatin) are more likely to cross the blood brain barrier than non-lipophilic statins (pravastatin).⁴⁵

Statins may therefore modulate the phosphorylation of tau and levels of Aβ in humans and this effect may depend on the dosage and CNS availability of the statin. Evidence of statin use in patients with AD is nonetheless varied and requires further investigation.

The effect of statins upon the visual system in patients with AD has only been studied in a population of at-risk individuals. One four-month, double-blind RCT (n = 57) evaluated the effects of simvastatin 40 mg on visual-motor function in asymptomatic middle-aged adult children of persons with AD.¹³⁸ There were no significant differences found in performance in the Rey complex figure copy test measuring visuoconstructional ability (p = .065) or grooved pegboard test measuring visuomotor coordination (p = .427) between simvastatin and placebo groups when compared to baseline.

Statins may also reduce the risk of subsequent dementia in patients with late-onset depression as evidenced by a retrospective cohort study.¹⁴² Serum inflammatory markers have been demonstrated to be increased in severe depression patients with coronary artery disease, including IL-1β levels. Statins have been suggested to function as an anti-inflammatory for depression in these patients by downregulation of IL-1β.⁹⁶

To summarise, it is not clear whether the purported effectiveness of statins in AD is directly related to cholesterol-lowering effects of these agents or rather their pleiotropic functions.¹⁴³ Certainly, evidence for a visual-motor function improvement with statins is lacking. Further RCTs are required to establish the effect of statin therapy upon the visual and neuropsychological functions of patients with AD.

Parkinson’s disease

Parkinson’s disease (PD) is a common age-related neurodegenerative disorder characterised by dopaminergic neuron death in the substantia nigra pars compacta. As with AD, there is also a documented higher risk of PD in ARMD patients.¹⁴⁴ Visual symptoms form part of the sensory dysfunction in PD and are likely to be a combination of motor disorders of ocular motility with non-motor disorders of visual perception and dopaminergic denervation of the amacrine retinal cells.¹⁴⁵ Visual hallucinations, a predictor of increased morbidity and mortality, may also be found in up to 37% of PD patients.^146,147 There is also significantly increased frequency of seborrhoeic blepharitis, meibomian gland disease and dry eye syndrome, nuclear and posterior subcapsular cataract and normal tension glaucoma in PD patients compared with healthy individuals (p < .05).¹⁴⁸

Cardiovascular risk factors are associated with worse motor and cognitive phenotypes in PD.¹⁴⁹ Statins, which are primarily indicated in cardiovascular disease prevention, have therefore been reviewed in PD as possible neuroprotectants, given their beneficial role in the attenuation of inflammatory responses which includes the production of TNF-α, nitric oxide and superoxide through peroxisome proliferator-activated receptor-alpha activation, the reduction in the accumulation of α-synuclein protein, and alteration of dopamine D1/D2 receptor modulation as well as dopamine D1/D2 upregulation.^150–152

In a recent meta-analysis, most statins demonstrated a reduction in risk of PD (relative risk [RR] 0.81, 95% CI 0.71–0.92).¹⁵³ Only pravastatin, due to its weak lipophilicity and difficulty crossing the blood brain barrier, non-significantly increased the risk of PD (RR 1.35, 95% CI 0.58–3.10). This property-dependent effect was also observed in persons free of PD, where continuation of lipophilic but not hydrophilic statin therapy was associated with a decreased risk of PD (HR 0.42, CI 0.27–0.64), especially in females.¹⁵⁴ Data also suggest that statins delay disease onset by as much as nine years and slow progression.¹⁵⁵ However, more recent evidence from a large Korean population-based study found any short-term (<one year) use of statin was significantly associated with a higher risk of PD incidence (adjusted HR 1.28, 95% CI 1.12–1.46) when compared with statin non-use.¹⁵⁶ Statin use greater than one year was not, however, significantly associated with an increased risk of PD.¹⁵⁶

The effect of statins on visual symptoms in patients with PD has not yet been studied. However, the use of statins in ophthalmological disorders associated with PD has been studied in other cohorts. Recent data from a population-based study of dry eye disease have suggested that oral statins at conventional dosage are correlated with increased reported symptoms of dry eye. It was postulated that this may reflect an underlying disordered cholesterol synthesis in Meibomian gland dysfunction uninfluenced by systemic statin administration at conventional dosing. It is possible, however, that high dose oral statins may decrease cholesterol contamination of the lipid layer of the tear film by reducing local increased cholesterol output from the meibomian glands, glands of Zeis and pilosebaceous glands thereby promoting tear film stabilisation.⁶ Indeed a topical statin in the form of atorvastatin eye drops has been shown to reduce the signs and symptoms associated with blepharitis-associated dry eye thereby potentially having a local effect on local oil gland structures.¹⁵⁷

While not specific to PD patients, a meta-analysis has found that statin use is associated with a 19% decrease in the risk of cataract (OR 0.81, 95% CI 0.72–0.92, p = .0009) despite the requirement of cholesterol for maintenance of lens transparency.^158,159 The association with cataract however, is controversial and studies have also found an increase in cataract incidence.⁵ In regards to the risk for normal tension glaucoma, there are effects on the trabecular meshwork increasing aqueous humour outflow facility, which may be beneficial in terms of intraocular pressure lowering, as well as on ocular blood flow through to the optic nerve head.⁵ Although the clinical evidence is conflicted, the majority of studies reviewed have demonstrated a reduction in the incidence of open angle glaucoma in the normal population.⁵

In summary, whilst statins may have a role in reducing the risk of PD, there is a lack of evidence suggesting a benefit on visual symptoms associated with PD. Further RCTs would be warranted to confirm this.

Conclusion

Statins are effective and well-tolerated hypolipidaemic agents highly utilised worldwide for the preventative therapy of cardiovascular and atherosclerotic disorders related to hypercholesterolaemia. Increasing evidence, albeit limited, from a range of clinical and experimental studies suggest statins may confer benefits to certain but not all neuro-ophthalmological conditions. Few of these studies report the effect upon vision-related outcomes and the effect of high dose statins is not yet firmly established. Various theories have linked the observed benefits to the pleiotropic immunomodulatory and anti-inflammatory properties of statins demonstrated in a range of biochemical and animal-model studies. Clinicians should be aware that when using statins for cardiovascular disease there may be a benefit to a co-existent neuro-ophthalmological condition. Clinicians should be cautious where prescribing statins in patients who have pre-existing, or have risk factors for, neuromuscular disorders given the risk of potential aggravation or induction due to the ability of statins to induce mitochondrial dysfunction or as a result of toxicity causing neuromuscular-related adverse drug events. Further prospective clinical trials assessing the effect of statins upon vision-related outcomes in the range of neuro-ophthalmological conditions are required to determine if they are effective in neuroprotection.

Funding Statement

SLW is supported by a Sydney Medical School Foundation fellowship.

Declaration of interest statement

KGO and SLW have a patent on topical Atorvastatin as a novel tear film stabiliser.