Colour-centred Release Hallucinations in a Patient with Early Parkinson’s Disease and Unrecognised Autosomal Dominant Optic Atrophy

ABSTRACT

Visual hallucinations in Parkinson’s disease (PD) are usually attributed to medications and dysfunction in higher order sensory processing as the disease progresses. However deficits in visual processing, including colour discrimination, have been reported in early, untreated PD and it is unclear how these, along with co-morbid conditions affecting vision, could contribute to hallucinations. This case describes a 66-year-old otherwise fully independent woman with early, mild PD who presented with discrete episodes of unusual vivid hallucinations centred on colour. She was later found to have a subclinical colour deficiency in excess of her PD and, after reporting a lifelong history of poor vision in her father, tested positive for autosomal dominant optic atrophy. This case illustrates how a lifelong extrinsic deficiency in colour vision can interact with the effects of visual changes in early stage PD and medication to provoke colour hallucinations. It therefore emphasises the importance of full ophthalmological work up in similar cases where hallucinations are atypical and unexpected for the severity and stage of PD.

Introduction

Visual illusions and hallucinations are estimated to affect up to 25% of patients with Parkinson’s disease (PD), and are also an important marker of morbidity, being associated with advanced illness duration, declining cognition, and sleep-wake cycle dysfunction.¹ Current models for these symptoms in PD focus on side-effects of medication and changes to attentional networks and thalamic processing as the disease progresses.¹ However, the mechanism underlying cortical release hallucinations suggests that any cause of visual input loss is clinically important, and to this end a range of visual processing deficits have been well described in early, untreated PD.^2–5 Whilst these alone are usually subtle and unlikely to be contributing to hallucinations as cortical release phenomena, they may decrease the threshold for other causes of visual pathology to cause symptoms. Unfortunately, as studies of hallucinations in PD have mostly excluded co-morbid ocular pathology, there are little data on this.¹

Here we present a case of early and unusual colour-centred hallucinations in PD, resulting from a new and important interaction between PD-related factors and a lifelong extrinsic optic nerve pathology affecting colour vision.

Case report

The case involves a 66-year-old woman with a background of early and mild idiopathic PD, first diagnosed 5 years previously, as well as a history of migraine with aura and gastro-oesophageal reflux. She was fully active and independent, attending a gym on a regular basis. She was a non-smoker, denied excess alcohol consumption, and ate a balanced diet. She used glasses for reading and denied any lifelong history of significant visual problems. Whilst she had more recently been made aware of occasional minor colour misperceptions, these never adversely affected her. She denied any history of problems with facial and objection recognition. From a motor perspective, her PD was well controlled on selegiline initially, followed by ropinirole with levodopa and entacapone introduced to good effect the following year. She occasionally used triptans for her migraines but used no other regular medications. There was no reported family history of PD or dementia. She did report a lifelong history of poor vision in her father, who still managed to continue work as a carpenter. Her father’s parents were first cousins. She has no children and has one brother who she did not suspect to have any visual problems.

She presented with new episodes of visual illusions involving altered colour perception as well as formed hallucinations. Her symptoms started 4 years into her PD diagnosis, when she complained of photopsia in her left hemifield that was initially attributed to her migraine. In the following year she developed discrete episodes where she saw random blocks of colour, for example where her partner’s face would turn blue or she would notice pink patches on the wall. Sometimes whilst knitting she noticed that running her finger along red fabric would leave a blue trail. These episodes would occur towards the end of an evening and last seconds to minutes. She also had more typical formed hallucinations, for example when she would see another person climbing on the sofa where she was sat, or noticed a spider in front of her cat for several minutes (supplementary audio file available online).

On examination, she saw 0.3 logMAR corrected with each eye, which did not improve with pinhole. Her pupils reacted normally with no relative afferent defect. Her intraocular pressures were normal. Ophthalmoscopy found no evidence of cataracts, but she had temporal pallor to both optic discs. 24–2 Humphrey perimetry revealed an enlarged blind spot bilaterally but with poor reliability indices. On Ishihara testing, she read the test plate only with each eye. On the Cortical Visual Screening Test (CORVIST)⁶ she gave incorrect responses to all four panels of the Hue-sorting test but otherwise scored 100% correct responses. Optical coherence tomography revealed severe retinal nerve fibre layer (RNFL) atrophy bilaterally (Figure 1). Magnetic resonance imaging (MRI) of her brain and orbits confirmed bilateral optic nerve thinning but revealed no other cause for this finding nor her altered colour perception. Vitamin B₁₂ and folate levels were normal. In view of the history of poor vision in her father, testing for the OPA1c gene was performed. After this returned positive, she was diagnosed with autosomal dominant optic atrophy (ADOA).

Figure 1.

Optical coherence tomography demonstrating severe bilateral thinning of the retinal nerve fibre layer, supportive of an optic neuropathy

Discussion

Visual hallucinations in PD are typically complex and involve the presence or passage of people, animals and objects, most often in darker lighting, Usually, they are not bizarre or distressing, and as insight is often preserved, they go unreported by patients.¹ Interestingly, release hallucinations seen in many conditions involving significant loss of visual input have a similar phenomenology.⁷ They are also often complex, non-threatening hallucinations of people and animals however they may also involve simple patterns on the wall or carpet when the patient knows these are plain.⁷ Whereas hallucinations in PD are often associated with declining cognition, release hallucinations are defined by the preservation of cognition, as is the case here. In both, vivid abstract colour hallucinations are highly unusual. Although in this case some of the formed hallucinations were reminiscent of PD, the nature of the colour hallucinations, preserved cognition and relatively early stage of PD prompted the neuro-ophthalmological referral.

As the cause of significant optic nerve disease in this patient, ADOA is alone a possible but unlikely cause of release hallucinations; indeed, no cases have previously been reported. It usually presents as mild-to-moderate central visual loss in the first decades of life, and is typically so slowly progressive that patients often adapt well and some can live normal lifestyles. This is in contrast to more common and usually more deleterious causes of visual loss which present later in life and which are well-known to lead to release hallucinations, for example macular degeneration, dense cataract, and glaucoma.⁷ Indeed, this patient’s visual acuity was reasonably preserved for her age and the main feature was a subclinical colour deficiency. As a lifelong, benign deficit, it was unlikely alone to account for new-onset colour hallucinations a few years into her PD diagnosis.

Nonetheless, if these symptoms were predominantly cortical in nature, as expected in PD, facial perception deficits would likely be present alongside other cognitive difficulties. This is because hallucinations involving colour correlate with functional MRI changes in specific areas of the ventral visual cortex,⁸ an area which is also known to be involved in facial recognition.⁸ The patient’s CORVIST score and preserved cognition suggest instead that the predominant cause of her symptoms is a deprivation of colour input to this region of cortex, leading to cortical release of colour phenomena. Altogether, it is likely that an interaction between ADOA and intrinsic PD-related visual processing deficits is responsible, as her symptoms are not typical for either condition alone.

There are many aspects of visual dysfunction in PD,² but of particular interest in this case are deficits in colour processing. Alongside other defects in acuity and contrast discrimination,^4,5 these have been demonstrated in early, untreated PD and have recently been proposed as a diagnostic marker.³ Advanced psychophysical tasks involving fitting discrimination ellipses have demonstrated the colour deficits to be independent of ageing as they are present in different visual processing pathways.⁹ Current theories implicate the retina, as suggested by the finding that the RNFL is thinner in PD,¹⁰ and that it contains alpha-synuclein deposits.¹¹ This may involve a retinal dopamine deficiency, based on neurochemical and electrophysiological evidence from studies using animal models of PD, as well post mortem studies.¹² Although subtle, these PD-related colour deficits have been independently associated with visual hallucinations in patients with otherwise normal visual acuity.¹³ This suggests that they might be a facilitating factor in certain PD patients, according to current models,¹ by synergising with the effects of medication and other deficits at all levels of the visual processing pathway, including with extrinsic comorbidities affecting vision. In turn, this may worsen or alter the nature of hallucinations, or provoke them earlier in the disease.

There have been reports of release hallucinations in PD patients with other conditions depriving visual input, for example bilateral posterior capsule opacification.¹⁴ However, these have involved late-onset visual loss and involve typical formed hallucinations. To our knowledge, this case is the first description of colour-centred hallucinations triggered by early-stage PD in a cognitively intact patient with a lifelong, progressive deficiency in colour vision.

Conclusion

This case highlights a new interaction of factors producing unusual colour-centred visual symptoms. It emphasises the importance of searching for co-morbid pathology by thorough neuro-ophthalmological work up, including testing colour discrimination, in similar cases where visual symptoms are out of proportion to the stage of PD. Future research is now needed to probe how visual processing deficits in PD and other neurodegenerative conditions interact with common ocular problems to produce hallucinations and worsen morbidity.

Declaration of interest statement

The authors report no conflicts of interest.