What is it, why does it happen and what can we do about it?
Recent onset of fluctuating inattention and confusion, linked to a range of predisposing and precipitating factors
Delirium is the most common complication of hospital admission for older people, with occurrence rates of 10-40%. It is particularly common in the post-operative period (>50% of hip fracture patients), has a high mortality (15-30%) and morbidity, may not be entirely reversible and results in huge cost to the healthcare system ($16,000 – 60,000/patient). Despite this, the pathophysiology remains poorly understood and there is only a sparse evidence base to inform management.
Diagnosis remains largely clinical and can be challenging in patients on non-medical wards or in the intensive or palliative setting. At the heart of a diagnosis is recognition of an acute or sub-acute change in baseline cognition/ behaviour and so is dependent on a collateral history.
One of the hallmarks of delirium is “fluctuation”, and quick mono-disciplinary ward rounds, the norm in hospitals, are likely to miss this feature. The use of the Confusion Assessment Method (CAM) can be helpful in these circumstances (see appendix).
There are three behavioural subtypes
- hyperactive
- hypoactive
- mixed
Hyperactive delirium is easily spotted; patients are hypervigilant, restless, distressed and agitated. In addition they frequently experience multi-modal hallucinations (visual, tactile) and delusions.
Hypoactive delirium can often go undiagnosed; in this instance patients show paucity of movement, apathy, drowsiness or even unresponsiveness. Common to both hyper- and hypoactive delirium is disturbance of the sleep-wake cycle. Hypoactive delirium is more common in the elderly and carries a worse prognosis.
Assessment of delirium can be undertaken using the “Confusion Assessment Method” (CAMS).
Look for these key cognitive features:
Pathophysiology
The pathophysiology of delirium is not fully understood. It seems likely that a combination of factors is required including drug toxicity, systemic and central nervous system inflammation and acute stress responses, that lead to neurochemical imbalance and, ultimately, delirium.
The two largest risk factors for delirium are age and pre-existing cognitive impairment, suggesting that delirium requires a “primed” substrate in order to develop. Patients with Alzheimer’s disease (AD) and Parkinson’s disease (PD) seem to be particularly at risk, not just of delirium, but of permanent sequelae from what should be a transient illness.
Two key neurotransmitters are involved in the pathogenesis of delirium – acetylcholine and dopamine. Cholinergic deficits are evident in patients with delirium and anti- cholinergic drugs are a well-known precipitant of confusion in the elderly. Dopaminergic drugs can also cause hallucinations and dopamine antagonists (haloperidol) have long been utilised as anti-psychotic agents.
Interestingly, patients with PD dementia (PDD), a common complication of PD, demonstrate many of the clinical features of a “delirium” – albeit in a more chronic setting (see table). Neuroimaging demonstrates profound cholinergic deficits in the brainstem and cerebral cortex, perhaps explaining the inattention and visual hallucinations seen in PDD.
Cholinesterase inhibitors are effective in PDD, treating cognitive and behavioural symptoms by boosting levels of acetylcholine in the brain. In contrast, dopaminergic drugs can markedly worsen symptoms in some PD patients, leading to somnolence, inattention, hallucinations and psychosis – symptoms akin to an acute delirium.
In addition to an “at risk” brain, increasing evidence suggests that trau
ma, infection or surgery can lead to production of a range of pro- inflammatory cytokines. These not only influence neurotransmitter levels but, perhaps more importantly, are also directly neurotoxic.
Low-grade inflammation may be relevant to the neurodegeneration see
n in AD and delirium accelerates the rate o
f cognitive decline in patients with AD.
Thus it is a combination of predisposing and precipitating factors that places patients at risk of developing delirium.
Treatment
As ever, the best form of treatment is prevention.
There is some evidence to suggest that complex interventions, involving targeting of six key risk factors (cognitive impairment, sleep deprivation, immobility, visual impairment, hearing impairment, and dehydration) can reduce the incidence of delirium by up to 1⁄3. These involve the use of regular orientation strategies, therapeutic ward-based activities, early mobilization post-op, quiet ward environments, non-pharmacological sleep promotion, early treatment of dehydration and provision of hearing and vision aids.
Proactive geriatric consultation has been shown to reduce the incidence of post-hip fracture delirium and early home rehabilitation is an effective and popular intervention with patients and relatives. More generic ward education strategies can also have an impact, although this is less marked than more intensive interventions, and more variable in outcome.
There is little evidence base to inform pharmacological management decisions. In general, haloperidol is the most widely used anti-psychotic in delirium studies. There is weak evidence that prophylactic use reduces delirium after hip fracture surgery but, in practice, very few physicians are likely to adopt this policy.
Haloperidol (0.5 – 1.0 mg po/im 4 hourly) reduces symptom severity and shortens duration of delirium. This is, of course, limited by Parkinsonian side-effects and is not to be recommended in patients with PD. In addition, it can be fatal in dementia with Lewy bodies, a condition closely related to PDD.
Atypical anti-psychotics (quetiapine, olanzapine, risperidone) have also been put forward as alternatives to haloperidol. Whilst these are associated with fewer extrapyramidal complications, there efficacy is less well established and they are not without their own specific problems.
Cholinesterase inhibitors are an attractive option given the neurochemical deficits in delirium and their established efficacy in PD dementia. There is no evidence of benefit at present in small case series and one study has suggested higher mortality in the treatment arm, albeit in small numbers. At present, therefore, they cannot be recommended in the management of delirium. Benzodiazepines should be avoided if at all possible.
Further reading:
- Inouye et al. A multicomponent intervention to prevent delirium in hospitalized older patients. NEJM (1999).
- Fong et al. Delirium in elderly adults: diagnosis, prevention and treatment. Nat Rev Neurosci (2009).
- Young & Inouye. Delirium in older people. BMJ (2007).
- Delirium: Diagnosis, prevention and management. NICE. (July 2010.)