Abstract
Charles Bonnet Syndrome describes the triad of complex visual hallucinations secondary to ophthalmic pathology in psychologically normal people. We present a rare case of acute-onset Charles Bonnet Syndrome following cardiac surgery that resulted in profound loss of visual acuity in both eyes with characteristic visual hallucinations that were initially mistaken for delirium. Computed tomography of the brain revealed bilateral occipital infarcts, providing the substrate for Charles Bonnet Syndrome. A high index of suspicion should be maintained in cognitively intact patients with visual loss who are also experiencing visual hallucinations to ensure prompt diagnosis and management of this often overlooked condition.
Introduction
Acute Stanford type A aortic dissection is the most common life-threatening disorder that affects the aorta. By definition, it must involve the ascending aorta, although the flap may extend to the aortic arch, the descending aorta and beyond. 1 It is a dramatic medical and surgical emergency with a high mortality rate (1%–2% per hour for the first 24 h).1,2 Instantaneous onset of severe chest (85%) and/or back pain (46%) are characteristic presenting symptoms (and may be described as ‘tearing’, ‘heart breaking’ or ‘crucifying’), but abdominal pain (22%) (due to extension of the dissection flap to the abdominal aorta causing abdominal visceral ischaemia), syncope (13%) and stroke (6%) are also common sequelae. 3 A diastolic murmur from aortic regurgitation is common if the aortic annulus is involved (40%–50% of Stanford type A) and pulsus paradoxus (tamponade), cardiogenic shock (tamponade, aortic regurgitation, coronary ostium occlusion) or hypovolaemic shock (aortic rupture) may also be evident. 4 Women with aortic dissection are more likely to present with atypical symptoms which often delays diagnosis and may lead to higher mortality. 4 Annual incidence is 3 per 100,000 in the UK. 5 Risk factors include hypertension, genetic disorders (e.g. Marfan’s Syndrome) and previous cardiac surgery (e.g. aortic valve replacement). 6
The precise trigger for aortic dissection remains unknown. Intimal tearing with propagation of blood into the media forcing the intima and adventitia apart, or rupture of the vasa vasorum (the small vessels that supply the aortic wall) which then weakens the aortic wall3,7 have both been suggested. Initial management involves close monitoring of pain, cardiac rhythm, blood pressure and urine output. Intravenous alpha/beta-blockers (e.g. labetalol) and vasodilators (e.g. nitroprusside) should be used to keep the systolic blood pressure <120 mmHg. However, swift open surgical intervention is required for Stanford type A dissections. This involves resection of the diseased portion of the aorta and replacement with a synthetic (Dacron) vascular prosthesis with concurrent coronary artery re-implantation (when the aortic root is involved). 4 Ischaemic stroke may occur in up to 14% of patients following corrective surgery. 8 In this case report, we present a rare case of acute onset Charles Bonnet Syndrome (CBS) as a likely result of ischaemic stroke.
Case report
A 66-year-old woman presented to the emergency department (ED) at a district general hospital complaining of sudden onset (10 out of 10 in severity at onset) ‘chest tightness’ which radiated to the jaw, down the left arm and through to the interscapular region. It began at rest, was associated with nausea and sweating and lasted for approximately 20 min until it was alleviated by IV morphine administered by the attending ambulance crew. The patient denied a ‘tearing pain’ sensation when specifically asked. There were no associated neurological symptoms, abdominal pain or syncope. Her medical history included hypertension and hypercholesterolaemia. Regular medications included a statin and a dihydropyridine calcium channel blocker. She was a lifelong non-smoker and had moderate alcohol intake of 1–2 units/night. On examination, her heart rate was 120 beats/min, blood pressure was 160/80 mmHg. There were no obvious murmurs or evidence of pulsus paradoxus. Her electrocardiogram showed sinus tachycardia without ischaemic changes, and chest X-ray was initially reported as ‘normal’ by the ED physician. Subsequent results revealed a raised troponin I (16 ng/mL) and D-dimer (941 ng/mL). Given the history of chest pain and elevated troponin I, she was treated with aspirin, clopidogrel, fondaparinux and metoprolol in the ED as per guideline for suspected acute non-ST elevation myocardial infarction (NSTEMI). Thereafter, she was admitted to the general medical ward where the pain recurred the following day. The medical team, having re-reviewed the Chest X-ray (CXR), felt that it demonstrated a widened mediastinum and given the clinical history, requested an urgent computed tomography aortogram (Figure 1). This demonstrated a 7–8 mm intramural haematoma of the ascending aorta, confirming an acute Stanford type A aortic dissection. She was then transferred to the Royal Infirmary of Edinburgh (RIE) for emergency surgical repair, replacing the ascending aorta with a 30-mm VascutekR interposition graft and repairing the right and non-coronary aortic root sinuses with a bovine pericardial patch.
Pre-operative computed topography chest with contrast: dissection of the ascending aorta with intimal flap dividing false and true lumen.
Post-operatively, she was observed in the cardiothoracic intensive care unit and extubated on day 2. She was noted to be drowsy (not unexpected). Several hours later, she reported seeing animals on the ward treatment trolleys and by her bedside. The animals were seen with both eyes but only when they were open. She was suspected of suffering from delirium and commenced on haloperidol later that night as per unit protocol. Given that delirium has an incidence of 50% after cardiac surgery and ‘common things being common’, it was thought to be the most likely diagnosis. 9
On day 5, persisting visual hallucinations prompted a thorough history and neurological assessment. Although she denied any visual loss, she was found to have 6/60 vision in both eyes (having had normal visual acuity in both eyes previously). Visual fields were intact to crude finger movement testing in all four quadrants. However, she felt that the hallucinations were more prominent in the left visual field of both eyes. No other neurological or psychiatric signs were elicited. The patient was unperturbed by the animals she was seeing and had full insight that these were not genuine phenomena. The abbreviated mental state examination was 30/30.
The following day, a non-contrast and post-contrast computed tomography (CT) of the head (Figures 2 and 3) showed marked bilateral subacute occipital infarcts, with the left larger than the right (an unexpected finding giving her symptoms were worse in the left visual field). A post-operative echocardiogram did not demonstrate any intracardiac source of embolus. The infarcts were not in the typical distribution for a watershed stroke and were thought likely secondary to embolism from the aorta as a result of the surgery.
Non-contrast and post-contrast computed tomography of the head. This shows bilateral subacute occipital infarcts, larger on the left than the right. These are not in the typical distribution for watershed infarcts. There is an associated shallow subdural haematoma at the left occiput. No significant mass effect. CT venogram was performed which did not show any filling defect with patent sinuses. Unremarkable opacified cerebral arteries. The remainder of the brain shows normal grey/white differentiation. No extra-axial abnormality. Ventricular system and basal cisterns are normal. Normal skull, soft tissues and airspaces. Non-contrast and post-contrast computed tomography of the head.

Differential diagnosis
In addition to delirium, posterior reversible encephalopathy syndrome (PRES) was considered as a differential diagnosis due to the patient’s high blood pressure on presentation to the ED; however, she had no other features of PRES, e.g. altered mental status, seizure activity or any headaches.
Outcome and follow-up
She was referred for stroke rehabilitation at her local hospital and at six weeks’ follow-up, and she continued to experience visual hallucinations. She received appropriate education regarding the CBS such that although there is no proven cure, avoidance of sensory deprivation and sedation can reduce frequency of hallucinations. It does not represent a psychiatric diagnosis.
Discussion
CBS is an uncommon neuro-ophthalmic condition that mainly affects the elderly, with a mean age of 70–85 years. 10 It is often underreported and misdiagnosed. No official diagnostic criteria exist but certain features are generally considered essential including: (1) visual loss (visual fields and/or acuity); (2) complex visual hallucinations; (3) full or partial insight into the unreal nature of the hallucinations and (4) absence of delusions and hallucinations in other sensory systems. 11
Any aetiology that affects the visual pathway from the eye to the visual cortex may cause CBS. 12 It is postulated that these hallucinations are caused by visual sensory deafferentation leading to disinhibition of the visual cortical regions which then fire spontaneously, manifesting as hallucinations. 13 CBS is well documented in chronic ocular diseases such as glaucoma, diabetic retinopathy and age-related macular degeneration.14,15 However, CBS following acute ocular or cerebral injury, as was the case here, is an extremely rare phenomenon. 16
Conclusions
Up to 30% of patients eventually found to have an acute aortic dissection are initially misdiagnosed with another condition, e.g. NSTEMI.1,17,18 This is particularly relevant with female patients, who may present with atypical signs or symptoms of aortic dissection. Misdiagnosis, as occurred in this case, could lead to a fatal outcome. Fibrinolysis, which might be employed in the case of an ST-elevation myocardial infarction, is a potentially even more dangerous scenario. Assuming that ‘common things are common’ can delay the diagnosis of less common presentations (in our case, delirium and CBS, respectively). This highlights the value of a thorough clinical evaluation. This case highlights the difficulties involved with diagnosing acute aortic dissection and CBS, both of which require a high-index of suspicion in order to make a timely and yet valuable diagnostic intervention.
Footnotes
Author contributions
FB: first author, manuscript drafting, RM: manuscript drafting, SD: manuscript drafting and MK: senior author, corresponding author, manuscript drafting.
Patient consent
Not applicable. No patient identifiable details provided.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
