Reversible Dementias Michael Philpot

INTRODUCTION

The aim of this chapter is to review the causes of ‘reversible’ dementias, their prevalence, methods of investigation and response to treatment. Given the right circumstances, almost any illness can affect cognition, so the objective is to identify those treatable conditions that most commonly present in this way. The early literature on this subject included numerous case studies or series of ‘dementia’ that appeared to resolve completely following treatment of the associated physical or psychiatric condition. Undoubtedly many of the patients described were actually suffering from delirium or other organic brain syndromes¹. Also, the concept of ‘reversible’ dementia appears at odds with the term ‘dementia’, denoting an insidiously progressive disorder². This confusion extends to psychi-atric disorders such as depression and schizophrenia that may present as ‘pseudo-dementia’³. Lastly, many patients, indeed probably the majority, with ‘real’ dementias will have concomitant conditions that, while not being aetiologically related to the dementia, may reduce cognitive function.

AETIOLOGY

The timely investigation of patients presenting with symptoms of dementia is a keystone of many clinical protocols^4–6. Diagnosis of ‘probable’ Alzheimer’s disease is largely by exclusion of other causes of dementia – including those reckoned to be reversible. A list of what are considered the classic potentially reversible causes of dementia (PRD) is given in Table 61.1. Table 61.2 lists PRD that have been described in the past 10 years – usually in either single cases or small case series. The identification of the latter can depend crucially on how intensively a case is investigated. They include conditions sometimes grouped together as the ‘rapidly progressive’ dementias⁷.

PREVALENCE

Despite the extensive lists in Tables 61.1 and 61.2, the frequency with which PRD occur in everyday clinical practice is an important factor in determining the rigour with which contributory physical disorders or other causes are pursued. The overall prevalence iden-tified in case series or clinic surveys can vary depending on the setting (inpatient, outpatient or community), the medical specialty involved (geriatric psychiatry, geriatrics, neurology or neuropsychi-atry) and the disease pattern of the country in which the study is carried out (e.g. neurological infections are more prevalent in devel-oping countries). Early studies generally failed to provide follow-up data to confirm whether the identified PRD were actually reversible or not. Previous reviews^24,25 concluded that approximately 11–12% of patients presenting to a variety of specialist services with symp-toms of dementia during the 1970s and 1980s had PRD. Philpot and Burns²⁴ found that the rate was 18% in patients under the age of 65 years but only 5% in those over 65, an observation that has been supported^26,27. However, few of the early studies provided follow-up data to confirm recovery. Clarfield²⁸ has reviewed the literature for studies published between 1987 and 2000 and identified 39 that ful-filled meta-analysis criteria. He found that recent studies were more often community or outpatient based, but that only 9% of cases had PRD. These figures represented a fall in prevalence from a previous review examining studies published in the 1970s and early 1980s when 13% of dementias were potentially reversible²⁵.

Table 61.3 lists more recent studies of PRD since 2001. The fre-quency of each underlying cause is compared to Clarfield²⁸.Mental disorders (chiefly depression) comprise the most prevalent category, with alcohol-related disorders also being more prominent than in pre-vious reviews. Space-occupying lesions, endocrine/metabolic disease and drug toxicity appear less frequently.

INVESTIGATION PROTOCOLS

A number of organizations have published guidelines for routine investigations used in the assessment of dementia, and the recommen-dations of three such are listed in Table 61.4^4–6. Chui and Zhang³⁹ examined the added value of using investigations routinely. After a thorough clinical assessment, blood tests and neuroimaging results changed management in only 13% and 15%, of cases respectively. Van Crevel etal.⁴⁰ found that the number of patients requiring inves-tigation to find one case of PRD was approximately 100, and the financial saving on care costs was insignificant. Foster etal.⁴¹,in a detailed examination of the cost-effectiveness of brain computed tomography, concluded that scans should only be performed routinely if symptom duration was less than one year, symptoms were rapidly progressive, presentation was atypical or the patient was under 65 years. However, the hard facts of health economics need to be bal-anced against the imperative to do one’s best for the patient, so consensus protocols tend not to specify limiting investigations to particular age groups.

Table 61.1 Classification of potentially reversible dementias (after Lishman²)

Reversible dementias	Common clinical examples
Degenerative	Normal pressure hydrocephalus, Wilson’s disease
Space-occupying lesions	Cerebral tumour, subdural haematoma
Infection	Neurosyphilis, chronic meningitis and encephalitis, HIV-associated dementia
Collagen-vascular	Cerebral vasculitis systemic lupus erythematosus, sarcoidosis
Metabolic/endocrine	Liver disease, uraemia, remote effects of carcinoma, hypo-and hyperthyroidism, hypopituitarism, hypo-and hyperparathyroidism, hypoglycaemia, Cushing’s disease, Addison’s disease
Toxic	Primary alcoholic dementia, Wernicke-Korsakoff syndrome, chronic intoxication with sedative drugs, heavy metals
Anoxic	Chronic obstructive airways disease, sleep apnoea syndrome, chronic carbon monoxide poisoning, post-cardiac arrest, congestive cardiac failure
Nutritional	Anaemia, deficiencies of vitamin B12, folate, thiamine
Psychiatric	Depression, late-onset schizophrenia

Table 61.2 Recently reported reversible dementias

Reversible dementias	Common clinical examples
Autoimmune encephalopathies	Anti-NMDA receptor⁸, anti-neuronal antibodies (limbic and extralimbic paraneoplastic syndromes)⁷, potassium channel antibody-associated⁹, non-vasculitic meningoencephalitis¹⁰, Hashimoto’s disease^11,12
Space-occupying lesions	Dural arteriovenous fistulae¹³
Infection	Neurocysticercosis¹⁴, cryptococcal meningitis¹⁵, enteroviral meningo-encephalitis¹⁶, Whipple’s disease¹⁷
Collagen-vascular	Cerebral amyloid inflammatory vasculopathy¹⁸, antiphospholipid syndrome¹⁹,other CNS vasculitides (multiple causes)⁷
Metabolic/endocrine	Paget’s disease²⁰, Cushing’s syndrome²¹
Toxic	Steroids²²
Anoxic	Spontaneous intracranial hypotension²³

RESPONSE TO TREATMENT IN ORGANIC DISORDERS

A minority of studies identifying the prevalence of PRD actually include follow-up data reporting the response to treatment. Individ-ual retrospective studies from the 1980s (e.g. Byrne¹) found some improvement in two thirds of patients and complete recovery in 40%. However, Clarfield^25,28 suggested a much less optimistic outcome: 15% of PRD showed some response to treatment, 8% with full recov-ery, the remainder only partial recovery. Overall, some degree of recovery was shown in only 0.6% of patients with dementia²⁸.In more recent studies listed in Table 61.3, treatment outcome was only reported in two studies. Taka da etal.³³ found that 55% showed some degree of reversibility. Of the 10 patients with neurosyphilis, only one fully recovered: full recovery was also seen in a patient with normal pressure hydrocephalus (NPH) and partial recovery in a patient with a subdural haematoma (SDH). All the potentially reversible cases identified in Sheng etal.³⁸ showed some degree of recovery. How-ever, 63% had functional psychiatric disorders, mainly depression. Clearly, the lack of reversibility revealed by these studies suggests that most of the potentially reversible disorders are actually concomi-tantwith real dementia⁴². Treatment of concomitant disorders may have some clinical benefits, improving overall function while not eliminating the underlying cause. Some problems may also be sec-ondary to the dementia itself, e.g. anaemia as a result of malnutrition. The response rates of a selection of disorders are now discussed in more detail.

Vitamin B12 Deficiency

Strachan and Henderson⁴³ first described dementia in association with B12 deficiency. Treatment with cobalamin brought about complete remission of cognitive impairment. However, there have been mixed results in prospective studies of replacement in patients with dementia and B12 deficiency, some showing improvement^44,45 and others showing little benefit^46–48. Patients with mild cognitive impairment and B12 deficiency fare better^44,47,48.

Thyroid Dysfunction

The propensity for thyroid deficiency to lead to memory impair-ment and psychomotor slowing has been known since the nineteenth century⁴⁹ and hypo-and hyperthyroidism are among the traditionally accepted causes of PRD². However, a clear reversibility of cognitive impairment sufficient to be considered dementia is rare²⁸. Hypo-and hyperthyroidism are found in approximately 0.4% and 0.3% of older people, but subclinical thyroid dysfunction is found five or six times more often^50,51. There is a complex relationship between this condition and incident dementia. An increased risk of demen-tia has been associated with subclinical hyperthyroidism^52,53 and hypothyroidism⁵⁴, while a U-shaped relationship with thyroid func-tion has also been demonstrated⁵⁵. However, the association found between high free T4, T3 levels and atrophy in the hippocampus and amygdala in otherwise healthy elderly people might suggest a preclinical phase of cognitive impairment⁵⁶.

Only gold members can continue reading. Log In or Register to continue