When psychiatric symptoms reflect medical conditions

ABSTRACT
The brain dysfunction associated with certain medical and neurological conditions can produce essentially any psychiatric symptom. This means there is always a chance that presentations thought to be ‘psychiatric’ are actually explained by unidentified medical pathology. This paper aims to outline an approach to minimise these missed diagnoses.
Key points
Mental-state disturbance in medical inpatients is delirium until proved otherwise
Psychiatric conditions tend to develop insidiously rather than over hours to days
A full physical and neurological exam, basic cognitive assessment and routine ‘psychiatric’ blood screens should be undertaken in all patients presenting with new-onset psychiatric symptoms
Cognitive assessment is fundamental to the identification of delirium and/or encephalopathy, with impairment in tests of sustained attention having particular sensitivity
Limbic encephalitis, especially NMDA receptor antibody encephalitis, should be considered as a differential diagnosis in first presentations of psychosis
Introduction
The days of medical dualism are over. It is beyond rational argument that psychiatric conditions have biological underpinnings. Our understanding of the pathophysiology of, for example, schizophrenia now dwarfs that of many ‘medical’ conditions, such as migraine. However, it is still the case that psychiatric diagnoses are syndromal, determined by the presence or absence of specific symptoms. With some notable exceptions, diagnostic tests are elusive. Given that many medical and/or neurological disorders can produce mental-state disturbance, the potential for ‘medical’ conditions to masquerade as psychiatric syndromes remains.
Here, we provide guidance to help clinicians identify when mental-state disturbance is due to an underlying medical and/or neurological condition. Our focus is on presentations of agitation, emotional disturbance or psychotic symptoms. Guidance for the general physician on somatisation and functional neurological symptoms can be found elsewhere.1,2 We do not aim to provide an exhaustive list of medical differential diagnoses. Instead, we have outlined our general approach to such cases, highlighting when clinicians should be especially vigilant to potential underlying medical and/or neurological illness. We also do not discuss the generally poor healthcare received by psychiatric patients. The statistics are shocking; major psychiatric disorder is associated with a 10–15-year reduction in life expectancy, with excess mortality predominantly attributable to medical illness rather than to suicide.3 Although the care of patients with mental illness in general hospitals has recently received more attention,4 this moral emergency warrants its own paper.
How to avoid misdiagnosis
Missing medical causes of disturbed mental state can have catastrophic consequences, because the underlying medical condition goes untreated. Data on the frequency with which this occurs are limited, but Johnson’s case series reporting that 12% of consecutive psychiatric admissions had some (previously unidentified) physical illness that was judged to be aetiologically important to the presentation remains a salutary lesson.5 In the absence of reliable data, we feel that it reasonable to refer to clinical experience, which suggests that the following are crucial in preventing the erroneous attribution of symptoms to psychiatric aetiology.
Think delirium
In medical inpatients, delirium is the cause of mental-state disturbance until proved otherwise. However, it is often missed and its pleomorphic presentations overlooked completely or mistaken as a psychosis. In its most florid hyperactive form (of which delirium tremens is the archetype), patients are agitated, hallucinating and experiencing persecutory delusions. This can be mistaken for schizophrenia or, because sleep disturbance can be prominent, even mania. However, hypoactive delirium is in fact the more common presentation; these patients often go undetected, or lethargy and psychomotor retardation confused for the avolition and withdrawal of severe depression.6
Misdiagnosis can be avoided if it is remembered that delirium is characterised by an abrupt onset, altered conscious level and fluctuating course, features that also distinguish it from dementia. Impaired attention, with associated disorientation, is the key clinical finding. It can be identified through simple bedside tests. At a minimum, one should formally test orientation to time and place and sustained attention, with serial seven subtractions or months of the year backwards being useful tests of the latter. Additional disturbances in cognition (particularly memory, executive and visuospatial functions) are also often present, thinking is muddled, sleep fragmented, and perceptual disturbances, especially illusions and visual hallucinations, can occur. There are many validated screening tools (eg the 4AT, www.the4at.com) and they should be more widely used. Patients with schizophreniform or manic psychosis are generally orientated and have preserved recent memory (Table 1). Although often distractible, they will not have the gross attentional disturbance of delirium. In ‘psychiatric’ conditions, hallucinations are usually auditory rather than visual. Causes of delirium that are commonly missed are detailed in Table 2. Although it relies on expert interpretation, electroencephalography can be used to distinguish metabolic and other systemic disorders from intracranial pathologies, and provide useful evidence of delirium in challenging cases.7,8
Comparison of features of various psychiatric presentations and deliriumaa
Causes of delirium especially likely to be missed
The treatment priority in delirium is identifying and addressing precipitants and maintaining factors. Multifactorial causation is the norm, meaning that consideration of potential contributors should continue even after a putative precipitant is identified. The threshold for developing delirium in compromised brains (be it because of age, dementia, multiple sclerosis, Parkinson’s disease, or traumatic brain injury) is reduced. If very vulnerable, relevant precipitants (such as sleep disturbance, hunger or simply being in a strange environment) can appear trivial. This undermines the argument ‘it can’t be delirium because they are not unwell enough’.
Adequate history, neurological exam and cognitive assessment
Psychiatric symptoms reflect brain dysfunction. When consequent to medical and/or neurological conditions, additional evidence of nervous system dysfunction is likely; this can manifest as motor (eg dysarthria or altered gait), sensory (eg visual field deficits or peripheral neuropathy), cognitive or language disturbance. Given that the ability of the patient to provide a reliable history could be compromised, collateral history should complement the physical, neurological and cognitive examination. This can also identify symptoms that patients are unaware of or reluctant to reveal; examples include the apathy and social inappropriateness accompanying degenerative conditions, such as frontotemporal dementia (FTD), episodes of unresponsiveness with muscle twitching suggestive of complex partial seizures, substance misuse, or the social withdrawal and bizarre preoccupations of a guarded patient developing schizophrenia. Extrapyramidal side effects (ie rigidity or tremor) are common in patients treated with antipsychotic medication, but localising neurological signs are not in keeping with a psychiatric diagnosis and imaging is indicated. Global or focal cognitive deficits might be apparent from patient or informant history, but might only be elicited by formal assessment (eg naming difficulties in semantic dementia or difficulty reading in posterior cortical atrophy). The Addenbrookes Cognitive Examination Version III provides a brief but impressively comprehensive means to do this, and it (or an equivalent) should be undertaken in all patients in whom cerebral pathology is suspected. It is available online along with guidance for use.9 Drug screens are important, but will not detect novel psychoactive substances.
Abnormal findings overlooked
When medical pathology is missed, case review often highlights clues that were there, but overlooked once it was decided that the presentation was psychiatric. Classic examples are pulmonary embolism attributed to a panic attack or co-occurring psychosis and movement disorder not prompting consideration of Huntington’s disease. It is of course the case that patients with psychiatric conditions can have abnormal findings on physical or laboratory investigation that would not justify an ‘organic’ diagnosis. Examples include sympathetic overactivity and hyperreflexia in anxiety or mildly elevated C-reactive protein (CRP) in depression. It is often prudent to recheck such markers, with reassurance provided when they settle as expected (eg tachycardia normalising when a panic attack resolves).
Be cognisant of how psychiatric disorders present
Just because a presentation is odd does not make it psychiatric. If in doubt, ask a psychiatrist because they are usually best placed to judge whether a presentation is indeed compatible with a psychiatric syndrome (but see Table 1); useful features include psychiatric conditions generally having an insidious rather than acute onset (although mania can be subacute) and tending to exhibit some consistency in dominant symptoms (rather than the fleeting and changeable symptoms of delirium). Schizophrenia and bipolar affective disorder usually first present in adolescence and/or young adulthood, being less likely explanations for new-onset psychotic symptoms in later life. Psychotic depression can first present at older ages; hallucinations and delusions are mood congruent, and generally focused on guilt, death and decay. Concrete guidance on who to refer to psychiatric and/or liaison psychiatry services is difficult to provide. However, it is the case that, as well as assisting with diagnostic clarity, psychiatric input is also often helpful in advising on management of agitation, risk management, capacity issues, and when use of the mental health act is indicated.
Further assistance in getting it right
Basic screening
Although over investigation is to be discouraged, basic screening should be undertaken in all cases of suspected psychiatric illness. As well as mental state, physical, neurological and cognitive exams, this will always include basic laboratory investigates (Table 2). Although scanning of all patients is not justified,10 imaging should be undertaken if the presentation is atypical for psychiatric illness or there are other ‘red flags’ (Table 3). Ordering of other investigations should be determined by individual presentations rather than by protocols, but guidance for investigation of some potential differential diagnoses are detailed in Table 4.
Situations in which brain imaging is mandatory, with magnetic resonance imaging generally being preferred modality
Limbic encephalitis and other recently characterised conditions in which psychiatric symptoms are prominent
Over the past decade, it has been recognised that various presentations can arise consequent to antibodies directed at neuronal cell surface antibodies that particularly target the limbic system. Although originally identified as paraneoplastic phenomena, it is now recognised that limbic encephalitis can arise in the absence of malignancy, often in young women. Most closely associated with psychiatric presentations is N-methyl-D-aspartate (NMDA) receptor antibody encephalitis, the early manifestations of which are often anxiety, dramatic expressions of distress (seeming ‘hysterical’), affective disturbance and psychosis. Delirium, seizures and severe autonomic disturbance can then supervene, although the range of presentation is wide. Other syndromes relating to other autoantibodies are increasingly recognised (Table 4).
These diagnoses should always be considered in the presence of movement disorder, seizures, prominent cognitive impairment, autonomic disturbance or treatment resistance. In many such cases, viral encephalitis will have already been a consideration and have prompted lumbar puncture (LP) and cerebrospinal fluid (CSF) analysis. However, it is conceivable that, in time, screening for these autoimmune conditions will be extended to all first presentations of psychosis. Diagnosis is based on detection of the relevant autoantibodies (presence in CSF being more specific than in serum), imaging (although magnetic resonance imaging [MRI] can be normal) and clinical judgement. Treatment centres on immunosuppression, supportive care and the exclusion and/or identification and treatment of any underlying tumour.
Limbic encephalitis is also a potential differential diagnosis in older first-presentations of psychosis, when an underlying malignancy is more likely. As age increases, the possibility that psychosis is consequent to an underlying neurodegenerative condition also increases. Visual hallucinations can suggest Lewy body dementia, which, given its fluctuating presentation, is easily mistaken for delirium. However, it is also increasingly recognised that certain genetic mutations giving rise to FTD can initially present with psychosis, which can take the more typically ‘psychiatric’ form of persecutory delusions and auditory hallucinations. The classic association is with the C9orf72 mutation (which can be tested for);11 one would expect poor performance on cognitive tests tapping into executive function (eg letter fluency), a collateral history of disinhibition and personality change, and an MRI scan showing localised frontal atrophy.
Conclusion
Here, we have focused on cases in which psychiatric symptoms arise from pathology that is already extant and detectable with a reasonable index of suspicion. However, there will always be some cases that only time will reveal; essentially, an underlying pathophysiological process has produced psychiatric symptoms early in its course, before other manifestations. It is often harsh to describe these as ‘missed diagnoses’, but it does emphasise the importance of a willingness to re-examine diagnoses if presentations change and unexpected symptoms appear. Equally, presentations highly suggestive of an ‘organic’ basis (rapid onset, disorientation or fluctuating presentation) can occur following psychological distress, sleep deprivation or sensory deprivation in the context of a ‘normal’ brain. Sometimes, only time and an open mind, on the part of both psychiatrists and physicians, will enable diagnostic clarity.
Declaration of interest
We confirm that there are no competing interests.
- © Royal College of Physicians 2018. All rights reserved.
References
- ↵
- Stone J.
- ↵
- Chitnis A
- ↵
- ↵
- Mental Health in General Hospitals
- ↵
- ↵
- O’Sullivan R
- ↵
- Obrecht R
- ↵
- Van Der Kooi Arendina W
- ↵
- Stringer S
- ↵
- ↵
- Kertesz A
Article Tools
Citation Manager Formats
Jump to section
Related Articles
- No related articles found.
Cited By...
- No citing articles found.