Introduction

Unprecedented numbers of people with intellectual disabilities (ID) are now living into middle and old age, with a concurrent increase in prevalence of general age-related health conditions (Haveman et al., 2009) and dementias (Strydom et al., 2010b) associated with increased morbidity and mortality (Coppus et al., 2008; Strydom et al., 2013b). A review of care of people with ID and dementia highlights concerns regarding pathways to assessment, changing care needs, transitions, end-of-life care, training, and support of care staff and families. Pharmacological and non-pharmacological management of behavioral and psychological symptoms of dementia (BPSD) is extrapolated from the general population (Courtenay et al., 2010). Whilst the cost of care increases with age, especially for those in residential services, dementia per se does not necessarily increase the overall care costs (Strydom et al., 2010a).

Diagnosing dementias in people with ID

The dementias are a meta-syndrome of acquired cognitive deficits across multiple domains, secondary to a range of brain disorders, resulting in loss of independence in daily functioning. The dementias encompass neurodegenerative disorders, vascular dementias, traumatic brain injury, alcohol-related dementia, and other brain insults. Historically, memory loss has been a fundamental of dementia diagnosis regardless of the severity of deficits in other cognitive domains. The Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5), whilst accommodating the continuing use of the term “dementia,” has introduced the category of Neurocognitive Disorder, with two levels of severity of acquired cognitive impairment (mild and major) across multiple domains, with or without memory impairment (American Psychiatric Association, 2013).

People with ID are a heterogeneous population with a multiplicity of neurodevelopmental disorders, severity and profile of pre-existing impairments in cognitive and adaptive functioning, and confounding multi-comorbidity, rendering the assessment and diagnosis of dementias a complex task (Torr, 2009). Standard diagnostic criteria for dementia have acceptable expert inter-rater reliability (0.68) and high specificity (95%); however, there is a risk for false-positive diagnoses, especially in those who are young, or have more severe ID and sensory impairments (Strydom et al., 2007, 2013a). Demonstrating decline in cognition and functioning from baseline is essential to a diagnosis of dementia. Furthermore, what degree of functional change warrants a diagnosis of dementia? Perhaps it is the increases in support needs requiring changes in the supports provided. The diagnosis of dementia also depends upon which criteria are used. Use of the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision (DSM-IV-TR) (American Psychiatric Association, 2000) criteria for dementia diagnosis is comparable with use of the Diagnostic Criteria for Psychiatric Disorders for Use with Adults with Learning Disabilities/Mental Retardation (DC-LD) (Royal College of Psychiatrists, 2001). However, dementia, even to a moderate degree, may be missed by the ICD-10 Diagnostic Criteria for Research (DCR-10) (World Health Organization, 1993; Strydom et al., 2007).

Lack of general and disorder-specific normative cognitive data requires the demonstration of changes from individual rather than group baselines. Baseline assessments in early adulthood are recommended for people with Down syndrome and longitudinal assessments for people with other ID when changes in cognition and/or functioning are noted. There is currently no standard protocol for the assessment of dementias in people with ID. A recent review identified 114 instruments; 35 informant-based, 79 direct assessments, and 4 test batteries (Zeilinger et al., 2013). For more detailed reviews of selected neuropsychological assessments, see Prasher (2009). Selected instruments should assess memory, executive functioning, visual spatial skills and language, attention, and speed of processing, if possible, as well as daily functioning. For people with severe cognitive impairment, instruments such as the Severe Impairment Battery (Saxton et al., 1993) have utility (Witts and Elders, 1998).

Deterioration in self-care and other activities of daily living are often reported by carers to precede memory and other cognitive impairments. Both retrospective carer report and prospective monitoring for functional decline have potential in the screening for dementia, especially for people with moderate and severe ID. Reports of memory change are better predictors of dementia than functional change for people with mild ID. Although emotional and behavioral changes also occur early on, these changes are non-specific and not predictive of dementia (Jamieson-Craig et al., 2010).

Sensory and motor impairments, medical and mental illnesses may mimic or be comorbid with dementia. The risk for delirium is high in people with ID given the higher rates of neurodevelopmental brain abnormalities, epilepsy, chest and urinary infections, and polypharmacy. Yet delirium is often overlooked. In a community cohort of 187 adults with Down syndrome, only one case of delirium was identified over two years (Mantry et al., 2008). Persistent subacute delirium can be mistaken for dementia. Long-term psychotropic medications, especially medications with anticholinergic effects, can result in persistent cognitive impairment (Gedye, 1998). Many people with ID do not have access to specialist clinical services (Torr et al., 2010a), and in practice there are real risks of false-positive and false-negative diagnoses of dementia.

Dementia in people with ID not due to Down syndrome

The Becoming Older with Learning Disability (BOLD) study is a community population longitudinal study of dementias in people with ID, excluding Down syndrome, aged 60 years plus (Strydom et al., 2007). Following screening, standardized clinical assessments were conducted, and diagnosis of dementias made by expert consensus according to DC-LD (Royal College of Psychiatrists, 2001), DSM-IV-TR (American Psychiatric Association, 2000), and DCR-10 (World Health Organization, 1993). Prevalence of any dementia was 13% for 60 years plus; 18% for 65 years plus; and 33% for 75 years plus (Strydom et al., 2007, 2009). This is consistent with a similar community-population study reporting overall prevalence of dementia of 22% for people with ID over the age of 65 years, whereby prevalence increased across age cohorts: 15% at 65–74 years; 24% at 65–84 years; and 70% at 85–94 years (Cooper, 1997) Although prevalence is related to age, when compared with the general population the Standardized Morbidity Ratio (SMR) for dementia is 2.77 for those aged ≥65 years. However the SMR diminishes with age from 7.7 for those aged 65–74 years to 2.7 for those ≥85 years, indicating differential risk for earlier onset of dementia in people with ID. (Strydom et al., 2009). Overall, there was no association between gender or severity of ID and prevalence of dementia (Strydom et al., 2007). The incidence of dementia was up to five times that of the general population, and estimated to be 55/1000 person-years in those over 60 years of age and 98/1000 person-years in the 70–74-year age group (Strydom et al., 2013b).

The most common subtype of dementia was Alzheimer’s disease (≥60 years, 8.6%; ≥65 years, 12%), followed by Lewy body dementia (≥60 years, 5.9%; ≥65 years, 7.7%), frontal type dementia (≥60 years, 3.2%; ≥65 years, 4.2%) and vascular dementia (≥60 years, 2.7%; ≥65 years, 3.5%; Strydom et al., 2007). Frontal type dementia was over-represented and vascular dementia is not as common as might have been expected. Lewy body dementia was “possible” rather than “probable” in the majority of cases, raising questions about delirium, which has common clinical features, as well as possible extra-pyramidal side effects from antipsychotic medications. The stability of diagnostic subtype was not reported at follow-up (Strydom et al., 2009). Very early-onset dementias have been described for some rare genetic intellectual and developmental disorders (Strydom et al., 2010b).

The minimal research into the risk factors for dementia in people with ID focuses on associations such as age, multiple physical problems, and poorly controlled epilepsy (Cooper, 1997). There are more hypotheses than evidence. For example, vitamin D deficiency is more common in people with ID (Frighi et al., 2014), and vitamin D deficiency is a risk factor for Alzheimer’s disease in the general population (Littlejohns et al., 2014). Repeated head trauma and head banging are associated with taupathy, atrophy, and cognitive impairment/dementia (Carlock et al., 1997). The big questions relate to the interplay between the complexities of neurodevelopment and neurodegeneration at the cellular level, and the impacts of environmental factors, and how a life is lived, upon these processes. In the absence of evidence to the contrary, it might be prudent to foster healthy lifestyles, with good nutrition, management of weight and vascular risk, regular exercise, educational and occupational endeavors, social engagement, and inclusion.

Down syndrome and Alzheimer’s disease

People with Down syndrome account for 15% plus of the identified population with ID. Life expectancy is now about 60 years. Reported prevalence varies, but overall about 50% of people with Down syndrome will develop the dementia of Alzheimer’s disease in the 6th decade, with an average age of diagnosis of Alzheimer’s disease in the mid 50s (Torr et al., 2010b). The reported incidence is 4.88 per 100 person–years during the 6th decade, increasing to 13.3 per 100 person-years in the 7th decade. Nonetheless, not all people with Down syndrome have dementia in older age and the prevalence falls to 25% in the over-60-year-old age group (Coppus et al., 2006). Emotional and behavioral changes precede frank cognitive decline, beginning with emerging executive dysfunction (Ball et al., 2006, 2008) before the development of impairments in memory, language, visual, spatial, and learnt motor skills.

Down syndrome is caused by trisomy 21 in 95% of cases, the remainder being due to mosaic or partial trisomy 21 and various translocations (Mulcahy, 1979; Stoll et al., 1998). Triplication of the amyloid precursor protein (APP) gene on chromosome 21 is a primary risk factor for the high rates of early-onset dementia of Alzheimer’s disease. Indeed, a 78-year-old woman, with partial trisomy 21, not involving the APP gene, showed no clinical or neuropathological evidence of Alzheimer’s disease (Prasher et al., 1998), and two women with Down syndrome and partial disomy for chromosome 21 living into their 70s and 80s showed no signs of dementia (Schupf and Sergievsky, 2002). Triplication of chromosome 21 results in complex whole of genome and epigenetic effects. Expression of the APP gene is 3–4 times higher than the expected 1.5-fold gene dosage effect. Moderate overexpression of ETS2, a transcription factor located on chromosome 21, increases transcription of both APP and presenelin-1 and β-amyloid production (Wolvetang et al., 2003) The gene for DYRK1A is also located on chromosome 21 and expression is increased further by β-amyloid. DYRK1A phosphorylates the tau protein, helping to drive the development of neurofibrillary tangles (Jones et al., 2008; Wegiel et al., 2008).

There is an age-related continuum from precocious and excessive cerebral amyloid deposition to the progressive development of amyloid plaques, followed by intracellular accumulation of neurofibrillary tangles, proceeding to neurodegeneration and dementia. Deposition of amyloid plaques may occur in the 2nd and 3rd decades, and the number of number of plaques and tangles increases greatly from the age of 30 years. By 40 years of age the concentration of plaques and tangle meet the neuropathological criteria for Alzheimer’s disease (Wisniewski et al., 1985). Positron emission tomographic (PET) methodologies also demonstrate increasing amyloid and tau concentration with age (Nelson et al., 2011). Amyloid plaques alone are not associated with significant cognitive change (Jones et al., 2009; Hartley et al., 2014). The development of neurofibrillary tangles occurs later, and behavioral and early cognitive changes begin with increasing concentration of neurofibrillary tangles (Nelson et al., 2011; Koran et al., 2014). Neurodegeneration progresses with age, with loss of total brain volume as well as orbitofrontal, parietal, and hippocampal atrophy, and increase in ventricular volume, and is associated with ongoing cognitive decline and dementia (Strydom et al., 2002; Koran et al., 2014). Clinically, serial neuroimaging may be required to delineate atrophy from pre-existing brain differences (Prasher et al., 1996; Strydom et al., 2002).

A number of factors modulate individual risk for the age of onset of dementia in people with Down syndrome. Increased risk is associated with the apolipoprotein E4 allele (Apo E4) whilst the apolipoprotein E2 allele is protective (Schupf and Sergievsky, 2002). Increased risk is also associated with female gender, lower age of menopause, lower weight, certain polymorphisms of CYP17 and CYP19, which are involved in synthesis of estrogen, and reduced bioavailability of estradial (Chace et al., 2012) and high total cholesterol, although statins reduced this risk (Zigman et al., 2007). In one study, serum levels of both amyloid peptides Aβ1–42 β1–40 were increased in adults with Down syndrome. The apolipoprotein E4 allele is associated with increased levels of Aβ1–42, and levels of Aβ1–42 are selectively increased in those with dementia (Schupf et al., 2001). Other studies have found no relationship between Aβ1–42, Apo E4, or the age of dementia (Jones et al., 2009). However, earlier age of onset of dementia is associated with the extended tau haplotype (Jones et al., 2008). Randomized controlled trials with antioxidants (Tyrrell et al., 2001) and memantine (Hanney et al., 2012) have not been of preventive benefit. The trial numbers are small and participants are older adults, including those with dementia. It is likely that preventive interventions will need to begin in early adulthood if not before.

Cholinesterase inhibitors and memantine

People with Down syndrome are of smaller stature and have differences in brain structure, cardiac conduction, and drug metabolism; hence, caution is required in extrapolating evidence from general population research into clinical practice. Cochrane Systematic Reviews report insufficient research into the treatment of Alzheimer’s disease in Down syndrome with acetylcholinesterase inhibitors, donepezil, rivastigmine and galantamine, and memantine, a glutamate N-methyl-d-aspartate (NMDA)-receptor antagonist (Mohan et al., 2009a, 2009b, 2009c, 2009d). Only one study met the inclusion criteria – a 24-week randomized double-blind placebo-controlled trial of donepezil 5–10 mg in 30 patients with Down syndrome and Alzheimer’s disease (Prasher et al., 2002). Improvements were not statically significant; however, significant benefits were found in the treatment group in an open-label extension study (Prasher et al., 2003). Improvements, or less rapid deterioration in cognition and functioning, have also been reported in small case series and non-randomized trials for donepezil (Lott et al., 2002; Boada-Rovira et al., 2005; Kondoh et al., 2005a) and rivastigmine (Prasher et al., 2005). In the absence of bradycardia and cardiac abnormalities, that are common in Down syndrome, cholinesterase inhibitors seem to be safe but not well tolerated. For equal doses of donepezil, people with Down syndrome have higher serum levels, which are associated with more adverse effects (Kondoh et al., 2005b), including fatigue, anorexia, nausea, vomiting, abdominal discomfort, diarrhoea, incontinence, insomnia, agitation, and muscle weakness (Prasher, 2004). Commencing at lower doses, with slow titration and dose adjustments, is recommended (Prasher, 2004). Memantine in adults with Down syndrome aged 40 years plus, with and without dementia, found memantine to be safe but of no benefit (Hanney et al., 2012). However, benefits have only been demonstrated for moderate to severe Alzheimer’s disease in the general population (McShane et al., 2006).