Headache,Diplopia and Hallucinosis: A Unique Diagnostic Challenge

Painful ophthalmoplegia is an important presenting problem to neurologists and ophthalmologists. The differential diagnosis is extensive and frequently difficult. If, in addition, the patient presents with complex visual hallucinations the challenge becomes even greater.

Case report

A 46-year-old woman presented with double vision. Three weeks previously she had woken with a mild, persistent, clawing, left frontal headache, which for 1 week she controlled with simple analgesia. Thereafter the pain had become explosive, accompanied by nausea, vomiting and tearing of the left eye, with little relief from analgesics. A few days later the left eye had begun to shut and she noted that, if maintaining the eyelid elevated, she saw things initially blurred and then obviously duplicated.

Mild hypertension and dyslipidaemia were the only diseases in her past medical history; both were under good pharmacological control.

At examination she had left partial ptosis with reactive pupillary mydriasis and limitation of both vertical and adductive eye movements. There was no other cranial nerve abnormality: abduction, as well as intorsion attempted down-gaze, were present in the left eye; the eyelid and all movements in the right eye were also normal.

Magnetic resonance imaging (MRI) with angiographic study was normal. Cerebrospinal fluid (CSF) glucose, cell and protein content were normal, as were microbiological and serological [venereal disease research laboratory (VDRL), Borrelia] studies. Herpes viruses and enterovirus were not detected by polymerase chain reaction. Extensive laboratory screening, which included erythrocyte sedimentation rate, serum angiotensin-converting enzyme, antinuclear antibodies, anti-neutrophil cytoplasmic antibodies, anti-dsDNA, thyroid, liver and renal function, glucose and lipid profiles, glycosylated haemoglobin, HIV1 and -2, rheumatoid factor, serum VDRL and fluorescent treponemal antibodies, was unremarkable. Blood pressure never exceeded 145/80 mmHg.

She was empirically placed on prednisolone 60 mg daily, with rapid, although not complete, symptomatic relief. After 4 days the dosage was tapered over another 4 days.

On day 3 she started complaining of hallucinations, seeing in front of her what appeared to be a television screen, with images ‘like if they were taken out of my past’, scenes from The Jungle Book, men with red equipment running, orange monkeys in the lap of bigger orange monkeys, and fragments of one of her favourite films, Schindler's List. Some of these images were accompanied by sound. She maintained full insight, being amused by these figures, which she knew were unreal. The hallucinations persisted even if she closed her eyes. No relation to the amount of light or time of the day was seen. There was no interference with her sleep or dreaming activity.

At this point diffusion-weighted MRI was used; it showed no alterations. An EEG, done in the hallucinatory period, was also normal.

Neurophthalmological and psychiatric assessments showed no abnormality.

She improved over the following days, being discharged on day 9 with no complaints or abnormalities on examination.

Over a 2-year follow-up she had two more headache episodes. The headaches were again in the left frontal region, boring in nature, not pulsating, with nausea. Each headache lasted 2 weeks and subsided spontaneously. There were no hallucinations or double vision.

Discussion

The principal differential diagnoses in a painful ophthalmoplegia, if traumatic, vascular, tumoral and infective causes can be excluded, are Tolosa–Hunt syndrome (THS) and ophthalmoplegic migraine (OM) (1, 2).

THS results from a non-specific cavernous sinus or superior orbital fissure inflammation, clinically resulting in a steady, boring, mostly unilateral pain, lasting for days to weeks and accompanied by third (but also sometimes a second, fourth or fifth) nerve palsy (3). Attacks recur at monthly to yearly periods and the diagnosis implies exclusion of other causes by careful investigation (3). With the exception of nausea and vomiting, no other neurological (or systemic) involvement has been reported in this condition (3). MRI usually discloses abnormal signal and mass lesions in the cavernous sinus ipsilateral to clinical symptoms (4, 5).

In this patient, the normal MRI study and the initial presence of pain without ophthalmoplegia make this an unlikely diagnosis.

OM is a rather rare condition, which almost always begins in childhood, characterized by repeated episodes of unilateral headache associated with a third (but also sometimes a fourth or sixth) nerve palsy, the headache subsiding before the nerve palsy, that can persist for several weeks (6). Acute cranial nerve transitory enhancement on MRI is noted in most cases (7, 8), supporting the theory of recurrent inflammatory cranial neuropathy (9).

In this patient the age at first presentation and the lack of MRI nerve enhancement also make this diagnosis unlikely.

At this point we were already intrigued with the patient, but what to say when she told us that she had started seeing orange monkeys?

Complex visual hallucinations can occur in several medical conditions. When they are vivid, multiple, well formed and colourful, with patient's full insight, the syndrome is referred to as lhermitte peduncular hallucinosis (10, 11). It has been best described in vascular injuries to the dorsal mesencephalon or thalamus. The patient usually starts hallucinating a few days after the stroke, gets worse at the evening and can have disturbed consciousness and sleep (10).

Other conditions with similar manifestations are hypnagogic hallucination, Charles–Bonnet syndrome, migrainous coma, and some substance use (e.g. lysergic acid, levodopa or mescaline) (10, 11). Schizophrenia, occipital epilepsy and occipital vascular disease can all produce complex visual hallucinations, but there are usually other specific clinical findings (11).

Our patient underwent MRI and EEG in the symptomatic phase, but both were completely normal, suggesting that peduncular hallucinosis or epilepsy are very unlikely. Neurophthalmological and psychiatric assessments were also normal, excluding afferent visual defect and overt psychosis. As the patient never had any signs of disturbed consciousness and had no familial history, migraine coma seems to be unlikely. Finally, drug exposure was limited to corticosteroids (prednisolone 1 mg kg⁻¹ day⁻¹), which has never been linked with such full-insight complex hallucinations, although it can cause psychotic episodes (12, 13).

As both hallucinosis and third nerve palsy can occur in mesencephalic or thalamic lesions, the possibility of a transitory functional disturbance in these areas seems attractive, and functional imaging might have revealed some abnormalities. However, the headache prominence, the absence of nuclear tegmental signs (the right eye was absolutely normal) and the prolonged symptomatic phase make this hypothesis less likely. Fascicular third nerve selective involvement by an inflammatory lesion in the mesencephalon could explain the third nerve palsy, and perhaps both hallucinosis and normal imaging, but again the prominence of the headache and the CSF normality argue against this hypothesis.

After reviewing this case, we reach a point where we have no diagnosis for this intriguing triad: unilateral headache followed by a third nerve palsy followed by hallucinosis. Indeed, we are convinced that we will never know why this woman, during those few days, saw in double what was never there.