Diabetes, Depression, and Cognitive Disorders
The interactions between diabetes and the mind are complex; physical illness increases the risk of a number of psychiatric disorders, while mental illness and its treatment also alter the risks of diabetes and worsen both acute metabolic and long-term outcomes of diabetes.
The prevalence of depression is approximately 1.5–2-fold higher in people with diabetes compared with the general population. Approximately 10% of people with diabetes will have a formal diagnosis of depression and around a quarter have significant depressive symptoms. Microvascular and macrovascular complications and treatment with insulin are associated with higher rates of depressive symptoms. The underlying mechanisms are multifactorial and include genetic and environmental factors as well as disease and treatment effects. The presence of depression adversely affects diabetes outcomes; quality of life and glycemic control are worsened, while the rates of microvascular and macrovascular complications and mortality are increased in people with depression. Screening for depression in people with diabetes and prompt treatment, where necessary, are recommended.
Diabetes has modest effects on certain aspects of cognition, including general intelligence, psychomotor speed, and mental flexibility, particularly when diagnosed in children under the age of 7 years.
Diabetes increases the risk of vascular dementia and Alzheimer’s disease, even after adjustment for traditional cardiovascular risk factors. Approximately 1 in 15 cases of dementia is attributable to diabetes. Insulin directly affects β-amyloid formation. Dementia impedes the person with diabetes’ ability to self-manage their diabetes and mandates a change in glycemic targets and management strategies.
KeywordsDiabetes Depression Diabetes-related distress Cognitive function Dementia Alzheimer’s disease
An effect of diabetes on the mind and vice versa has been recognized for many centuries; in the seventeenth century, Thomas Willis discussed how “diabetes is a consequence of prolonged sorrow” (Willis 1675). As the brain is highly vascular and dependent on glucose for its normal functioning, it is perhaps unsurprising that diabetes affects cognitive function and the risk of mental illness. What is surprising is that clinicians looking after people with diabetes frequently ignore this association. Nevertheless, the effects of comorbid mental illness on someone with diabetes may be profound as the comorbidity worsens the clinical outcomes of both conditions. Quality of life across a broad range of domains is worsened, while the individual’s ability to self-manage their diabetes is impaired, ultimately leading to a higher incidence of complications and reduced life expectancy (Holt and Katon 2012).
Despite the pressing clinical need to consider the comorbidity, in many countries, mental and physical health services are not properly integrated; this leaves diabetes services poorly equipped and organized to address both the physical and psychological needs of patients in the same setting (Mitchell et al. 2009). Over the last decade, however, there have been increasing levels of interest in comorbidity from researchers, who have made considerable progress in understanding the epidemiology and underlying mechanisms explaining the association. This is beginning to change clinical practice with national and international guidelines highlighting the importance of assessing and treating the psychological sequelae of diabetes (International Diabetes Federation 2012; National Institute for Health and Care Excellence 2015a, b).
This chapter will first describe the complex relationship between diabetes and depression before considering the effects of diabetes on cognitive function, with particular reference to the association between diabetes and dementia.
Diabetes and Depression
Depression is a mood disorder, which is characterized by persistent low mood and loss of interest or pleasure in life. Other symptoms include weight loss or gain, change in sleep patterns, agitation or retardation, fatigue or loss of energy, feelings of worthlessness or guilt, diminished ability to think or concentrate, and recurrent thoughts of death including suicidal ideation, respcetively.
The Diagnostic and Statistical Manual of Mental Disorders, 5th Edition (DSM-5): definition of a “major” depressive episode
A. Five (or more) of the following symptoms have been present during the same 2-week period and represent a change from previous functioning; at least one of the symptoms is either (1) depressed mood or (2) loss of interest or pleasure
Depressed mood most of the day, nearly every day, as indicated by either subjective report (e.g., feels sad or empty) or observation made by others (e.g., appears tearful)
• Markedly diminished interest or pleasure in all, or almost all, activities most of the day, nearly every day (as indicated by either subjective account or observation made by others)
• Significant weight loss when not dieting or weight gain (e.g., a change of >5% of body weight in a month) or decrease or increase in appetite nearly every day
• Insomnia or hypersomnia nearly every day
• Psychomotor agitation or retardation nearly every day (observable by others, not merely subjective feelings of restlessness or being slowed down)
• Fatigue or loss of energy nearly every day
• Feelings of worthlessness or excessive or inappropriate guilt (which may be delusional) nearly every day (not merely self-reproach or guilt about being sick)
• Diminished ability to think or concentrate, or indecisiveness, nearly every day (either by subjective account or as observed by others)
• Recurrent thoughts of death (not just fear of dying), recurrent suicidal ideation without a specific plan, or a suicide attempt or a specific plan for committing suicide
B. The symptoms cause clinically significant distress or impairment in social, occupational, or other important areas of functioning
C. The symptoms are not due to the direct physiological effects of a substance (e.g., a drug of abuse, a medication) or a general medical condition (e.g., hypothyroidism)
Epidemiology of Diabetes and Depression
Depression within the general population is common and its prevalence is increasing. It is predicted to become the second leading global cause of disability after heart disease by 2020 (Murray and Lopez 1997). The lifetime prevalence varies widely across the globe from 3% in Japan to 17% in the USA, falling between 8% and 12% in most countries. At any one time, approximately 3–5% of men and 8–10% of women have depression. Given the high prevalence of diabetes and depression, one would expect a degree of comorbidity, but the current evidence suggests that depression occurs more frequently in people with diabetes and vice versa than would be expected by chance (Holt et al. 2014).
Recent meta-analyses have demonstrated that significant depressive symptoms affect approximately 1 in 3–4 adults with diabetes, while a formal diagnosis of depressive disorders is made in approximately 10–15%, equivalent to a 1.5–2-fold increased prevalence (Anderson et al. 2001; Ali et al. 2006). Longitudinal cohort studies report the incidence of depression to be 15–24% higher in people with diabetes compared with those without diabetes (Nouwen et al. 2010; Mezuk et al. 2008). On the other hand, the incidence of type 2 diabetes is also increased by 15–37% among people with depression (Mezuk et al. 2008; Knol et al. 2006), indicating the bidirectional nature of the relationship between these conditions. Episodes of depression appear to be more persistent and more likely to relapse among people with diabetes (Nefs et al. 2012), which may, in part, explain the discrepancy between the relative incidence and prevalence figures.
Although the literature is consistent in showing an increased prevalence of depression in people with diabetes and vice versa, within each of the meta-analyses, there is considerable variation in risk estimates. This variation stems in part from the meaning of the word “depression,” which spans from relatively minor, occasional negative mood symptoms to life-threatening disabling conditions (Holt et al. 2014). More recently, papers have started to differentiate between “depressive symptoms” and “depression,” and this change in definition partly explains why current risk estimates of “depression” are lower than former ones. Another reason why earlier studies reported higher prevalence rates is because the studies recruited selected patient populations, often drawing from specialist diabetes clinics, where referral patterns and other differences in demographic characteristics, such as ethnicity, may increase the likelihood of depression.
The gold standard diagnostic procedure is a diagnostic interview, such as the Structured Clinical Interview for DSM-IV-TR Axis I Disorders (SCID) interview (First et al. 2002) or the Schedule for Clinical Assessment in Neuropsychiatry 2.1 (World Health Organisation 1999), but these are time-consuming and are unfeasible in most large epidemiological studies. Consequently, many studies have relied on self-rating scores, which tend to overestimate the true prevalence of depression and only provide an estimate of true caseness. These questionnaires may further exaggerate the prevalence of depression because of the overlap of the symptoms of diabetes and depression (Roy et al. 2012).
Depression or Distress
Worries about the future and the possibility of serious complications.
Feeling guilty or anxious when you get off track with your diabetes management.
Feeling scared when you think about living with diabetes.
Feeling discouraged with your diabetes regimen.
Feeling depressed when you think about living with diabetes.
These symptoms are recognized in up to 60% of people with type 1 diabetes or insulin-treated type 2 diabetes (Polonsky et al. 1995; Sturt et al. 2015) and are negatively associated with diabetes self-care and optimal glycemic control (Polonsky et al. 1995). Indeed HbA1c correlates more closely with diabetes-related distress than depression. These feelings are more likely to develop in those with long-standing diabetes and in those with recurrent severe hypoglycemia. Given the commonality of some symptoms, e.g., low mood and guilt, it is unsurprising that many people are reported to display both diabetes-related distress and depressive symptoms with ~30% overlapping variance. Nevertheless, as well as the distinct association with glycemic control, the association with self-management also differs between distress and depression, strengthening the view that these are two distinct entities.
Many studies examining the prevalence of depression in people with diabetes have not differentiated between the types of diabetes. This limitation is important because people with type 2 diabetes are generally older and depression prevalence varies with age, the rates of various diabetic complications and other comorbid conditions (e.g., obesity and heart disease) differ, and management strategies are different. Because the prevalence of type 1 diabetes is so much lower than type 2 diabetes, people with type 1 diabetes are underrepresented in depression association studies. One review of depression in type 1 diabetes (Barnard et al. 2006), however, reported that depression was present in 12%, compared with 3.2% in people without diabetes. However, if studies without control groups and interview ascertainment were excluded, the estimated prevalence fell to 7.8%, which was no longer statistically significantly different from people without diabetes (odds ratio [OR] 2.4, 95% CI −0.7 to 5.4). A recent study of 368 individuals with type 1 diabetes found an unexpectedly low rate of major depressive disorder (3.5%) and highlighted the marked difference in estimated prevalence rates using self-rated questionnaires (11.4%) compared with diagnostic interviews, again, perhaps, reflecting an effect of the overlap with diabetes-related distress (Fisher et al. 2016).
Increased risk of depressive episodes in people with and without diabetes in four continents after adjustment for age, sex, education, BMI, smoking, and physical activity (Mommersteeg et al. 2013)
Although there are few data, depression rates (9–26%) also appear elevated in children and adolescents with diabetes (Holt et al. 2014).
Etiology of Diabetes and Depression
Which People with Diabetes Are at Risk of Depression?
Female sex, marital status, childhood adversity, and social deprivation are all risk factors for depression in otherwise healthy individuals, and these appear to operate equally in people with diabetes. However, in addition, there are a number of diabetes specific and treatment risk factors associated with the development of depression.
Poor glycemic control and recurrent hypoglycemia are risk factors for depression, in part as a direct effect of hypoglycemia and hyperglycemia on brain function as well as the psychological effects of abnormal glucose levels. Animal models of diabetes have loss of hippocampal integrity and neurogenesis (Ho et al. 2013), while hippocampal atrophy has also been shown in MRI studies of people with diabetes (Lyoo et al. 2009). These structural changes are associated with neurotransmitter abnormalities, including increased prefrontal glutamate-glutamine-gamma-aminobutyric acid levels, which have been observed in people with type 1 diabetes in a manner that correlates with mild depressive symptoms.
Diabetes is not the only chronic physical condition associated with the development of depression, which also occurs more commonly in people with cardiovascular disease (Carney and Freedland 2003), cancer (Sellick and Crooks 1999), and inflammatory arthropathies among others (Matcham et al. 2013). As disease burden increases, so does the prevalence of depressive symptoms. It is, therefore, unsurprising that depressive symptoms are more common in people with diabetes who have developed either macrovascular or microvascular complications (de Groot et al. 2001). In a specialized outpatient clinic, people with two or more diabetes complications had twice the risk of depression, with neuropathy and nephropathy showing the strongest association with depression (van Steenbergen-Weijenburg et al. 2011). Sexual dysfunction and painful peripheral neuropathy also appear to be particularly associated with depression (de Groot et al. 2001).
People with insulin-treated type 2 diabetes have higher rates of depression compared to those treated with lifestyle interventions or noninsulin medications (Hermanns et al. 2005; Li et al. 2008). Exactly why this is the case is uncertain but probably has more to do with the increased treatment demands including intensive self-monitoring of blood glucose, longer duration of disease, and higher rates of diabetes complications than a direct effect of insulin per se (Li et al. 2008).
Why Do People with Diabetes Develop Depression?
The traditional view is that people with diabetes develop depression because of the psychological response to living with a chronic condition that is associated with unpleasant consequences and treatment that place heavy behavioral demands on the individual. There is support for this hypothesis as a meta-analysis indicated that the rates of depression were only increased among people with diagnosed diabetes while those with undiagnosed diabetes or impaired glucose regulation had no difference in depressive symptoms compared to those with normal glucose metabolism (Nouwen et al. 2011). This finding is important for clinicians who have the responsibility of communicating the diagnosis and its implications to people with new onset diabetes in a sensitive and compassionate manner to help people adjust to the diagnosis.
According to a German study, adults with new onset type 1 diabetes were more than twice as likely to develop a major depressive episode (5.8% vs. 2.7%), although the difference was only statistically significant in women (Petrak et al. 2003). By contrast, the situation appears less clear cut in people with type 2 diabetes where several studies have shown that the diagnosis has little impact on well-being (Adriaanse et al. 2004; Pibernik-Okanovic et al. 1996); the higher rates of depression only start to appear as the disease moves from being asymptomatic to one where complications begin to occur and where treatment demands increase.
This psychological model does not preclude other biological mechanisms, and it is important to recognize that acute changes in glucose may lead to a change in mood (Hermanns et al. 2007). Whether long-term changes or glucose variability may trigger depression directly is uncertain but changes in brain structure and function have been seen in the areas responsible for mood in people with type 1 diabetes (Lyoo et al. 2009).
Why Do People with Depression Develop Diabetes?
The low mood and loss of interest in pleasurable activities may lead to changes in behavior that increase the risk of diabetes. People with depression tend to eat less prudent diets (comfort eating is a readily understandable concept), are less likely to undertake regular physical activity, and are more likely to be smokers, all of which increase the risk of diabetes (McMartin et al. 2013; Payne et al. 2012; Weyerer 1992).
Depression is associated with poorer self-care management. This has been studied in more depth in people with established diabetes where people with comorbid depression are more likely to miss medical appointments and are less likely to follow advice about medication use, glucose monitoring, and foot care (Gonzalez et al. 2008). In people with established diabetes, this is associated with poorer diabetes outcomes, but if similar patterns of behavior predated the diagnosis of diabetes, this may have contributed to its onset.
Several biological changes occur in depression, including alterations in the hypothalamic-pituitary-adrenal (HPA) axis and inflammatory markers that could lead to increased insulin resistance and consequently risk of diabetes (Champaneri et al. 2010). Subclinical hypercortisolism, blunted diurnal cortisol rhythm, and hypocortisolism with impaired glucocorticoid sensitivity have all been observed in depression, while inflammatory changes include elevated concentrations of C-reactive protein, tumor necrosis factor-α (TNF-α), and proinflammatory cytokines, which have been implicated in causing sickness behavior in animal models and depression in humans (Musselman et al. 2003; Dantzer et al. 2008). Disrupted sleep, which is common in depression, may be a further biological mechanism linking to diabetes as poor sleep quality and altered circadian rhythms are associated with an increased risk of diabetes through insulin resistances (Gangwisch 2009).
There have been concerns that the use of at least some antidepressants may worsen the risk of diabetes (Barnard et al. 2013) as substantial weight gain may occur with certain antidepressants, including mirtazapine, amitriptyline, and paroxetine (Serretti and Mandelli 2010). Case reports and early observational studies demonstrated a consistent association between antidepressant usage and risk of diabetes; however, more recent cohort studies have shown much lower odds ratios, to the point where the association could be explained by confounding, and the antidepressant use is merely a marker of individuals at high risk of diabetes. Randomized controlled trials have reported both hyperglycemic and hypoglycemic effects suggesting that not all antidepressants are alike (Barnard et al. 2013).
Do Common Antecedents Explain the Association Between Diabetes and Depression?
Both diabetes and depression occur more frequently among people from lower socioeconomic classes raising the possibility that both conditions occur because of shared environmental risk factors. The importance of social status was demonstrated in a recent study from Denmark, which showed that while all people with diabetes were more likely to develop depression than those without diabetes, those from lower employment and income groups were disproportionately affected (Cleal et al. 2017).
It is uncertain how the adult environment increases the risk of diabetes, but poor physical (e.g., traffic, noise, and decreased walkability) and social environments (e.g., lower social cohesion, increased violence, and decreased residential stability) are associated with worse diet and lower physical activity levels that predispose to obesity, diabetes, and depression (de Vet et al. 2011). Although it is impossible to determine causality from these observational studies, dysfunctional HPA axis activity and disruption of its normal circadian rhythm (i.e., blunted profile) (Skinner et al. 2011; Brenner et al. 2013; Karb et al. 2012; Do et al. 2011; Dulin-Keita et al. 2012) as well as enhanced inflammation have all been observed in people living in adverse neighborhood environments providing a potential biological mechanism to explain the association (Browning et al. 2012; Broyles et al. 2012). Similar mechanisms may also operate in childhood adversity.
An adverse fetal environment may also predispose an individual to both type 2 diabetes and depression. There is a J-shaped relationship between birth weight and plasma glucose, insulin concentrations, and type 2 diabetes, while some, but not all, studies have shown that fetal undernutrition is associated with adult depression (Holt et al. 2014). Again, programming of the HPA axis may be one biological mechanism to explain the association (Champaneri et al. 2010).
Consequences of Depression in Diabetes
People with comorbid depression experience worsened diabetes outcomes and poorer quality of life (Goldney et al. 2004; Jacobson et al. 1997; Carper et al. 2014). While it is clear that those with microvascular complications are more likely to develop depression, recent work has also shown that those with depression are more likely to develop complications. In a 10-year cohort study of individuals with childhood-onset diabetes, in addition to longer duration of diabetes and poor glycemic control, the overall proportion of time that an individual was depressed predicted retinopathy severity (Kovacs et al. 1995). Similarly, in a longitudinal study of people with type 2 diabetes, progression to diabetic retinopathy and to proliferative diabetic retinopathy was more likely in those with high depressive symptom scores at both baseline and 6-year follow-up (Roy et al. 2007). Depression may worsen the pain experienced by those with painful peripheral neuropathy (Katona et al. 2005). Poor glycemic control is not the explanation for the increase in microvascular complications seen in people with comorbid depression as studies have not reported a consistent association between depressive symptoms and HbA1c (Lustman et al. 2000; Hislop et al. 2008; Aikens et al. 2008). Nevertheless, impaired self-care among people with depression may play a role.
Cardiovascular morbidity and mortality are increased in people with comorbid diabetes and depression (Park et al. 2013); in one study, those with diabetes and depression have an annual mortality rate of 8%, which was 2.5-fold higher than those without either condition (Egede et al. 2005).
Management of People with Diabetes and Depression
A greater awareness of the link between diabetes and depression has led several national and international guideline bodies to recommend action to improve the psychological well-being of people with diabetes (International Diabetes Federation 2012; National institute for Health and Care Excellence 2015a, b). There is a responsibility for healthcare professionals to identify depression when it occurs and then institute prompt treatment in order to reduce depressive symptoms and to improve self-care, glycemic control, and diabetes outcomes, respectively (Petrak et al. 2015).
Screening and Diagnosis of Depression
It goes without saying that before treatment can be offered, depression must be recognized and diagnosed. A formal diagnosis of depression requires a time-consuming validated interview, and so this is impractical for routine clinical care. Consequently, quicker and cheaper screening methods are needed to identify those in primary and secondary care with depressive symptoms who should then go forward for a diagnostic interview (Holt and van der Feltz-Cornelis 2012).
Many easy-to-use questionnaires that can be self-completed have been developed for routine use in clinical care, but because of the overlap between symptoms of diabetes and depression, only validated questionnaires should be used for people with diabetes (Roy et al. 2012). The Patient Health Questionnaire-9 (PHQ-9), which contains nine questions, is the most widely used and validated questionnaire in type 2 diabetes, but even this overestimates the prevalence of depression (Fisher et al. 2016). A score of ≥10 reliably identifies those with major depression is in community populations, but a higher cut-off of ≥12 has been suggested for people with diabetes in order to improve the discrimination between diabetes-related symptoms and depressive symptoms (van Steenbergen-Weijenburg et al. 2010). The Beck Depression Inventory (BDI), the Centre for Epidemiologic Studies Depression Scale (CES-D), and the Hospital Anxiety and Depression Scale (HADS) are other examples of well-validated questionnaires for people with diabetes.
During the past month, have you been bothered by having little interest or pleasure in doing things?
During the past month, have you been bothered by feeling down, depressed, or hopeless?
If the answer to either is “yes,” the healthcare professional should ask the patient if they want help with this problem. If the answer to this is also “yes,” a diagnostic interview should be undertaken followed by appropriate referral and treatment.
Although diagnosis of depression is necessary to instigate treatment, there is a debate as to whether screening for depression should be undertaken (Holt and van der Feltz-Cornelis 2012). Depression screening in the general population has little or no impact on the detection and management of depression if used alone, and robust clinical pathways are essential to ensure that appropriate treatment can be offered if a diagnosis is made (Gilbody et al. 2008). The importance of this in the context of diabetes was demonstrated in a Dutch randomized controlled trial, which investigated the benefits of depression screening in people with type 2 diabetes (Pouwer et al. 2011). Following screening, although written feedback was provided to both participant and doctor, neither utilization of mental health services nor depression scores improved. A further study from the USA also failed to demonstrate any improvement in depressive symptoms despite a modest increase in mental healthcare utilization (Scollan-Koliopoulos et al. 2012). Low acceptance of screening and subsequent referral to further care by people with diabetes, failure to screen those at highest risk of depression, reluctance by healthcare professionals to undertake screening and treatment, and generally poor quality of depression care in primary care systems may all contribute to these findings (Petrak et al. 2015).
Thus, while there is a strong imperative to identify people with depression, better integration with care pathways is needed, before screening can be wholeheartedly adopted, not least because screening without appropriate follow-up could lead to harm, by increasing the stigma and discrimination associated with depression and the risk of labeling transient distress as illness (Petrak et al. 2015).
Treatment of Depression in People with Diabetes
As depression adversely affects psychological well-being and diabetes outcomes, the best treatment approaches focus on both improving depressive symptoms and diabetes self-management. The aim of depression treatment is to achieve complete remission of symptoms with the further goals of improving health-related quality of life and psychosocial functioning (Petrak et al. 2015). Over the last decade, a number of randomized controlled trials have demonstrated the effectiveness of both psychological and pharmacological treatments of depression in people with diabetes (Petrak et al. 2015). Most of these trials have been undertaken in people with type 2 diabetes, and so, there is still a paucity of evidence for type 1 diabetes.
Psychological interventions are heterogeneous incorporating various techniques (e.g., cognitive behavioral therapy, problem-solving, and psychodynamic), different settings (primary and secondary care), and media (face-to-face, group, web-based, and telephone contacts) (Petrak et al. 2015). Given this level of variability, it is perhaps unsurprising that the effectiveness of interventions differs and comparisons between trials are challenging. Nevertheless, meta-analyses suggest psychological interventions improve depressive symptoms, with a moderate-to-large effect size (standardized mean difference [SMD] ranging from −0.14 to −1.47). The effect on glycemic control, however, is more modest with one systematic review reporting a reduced HbA1c of ~0.6% (6 mmol/mol) (Ismail et al. 2004) and another indicating a nonsignificant improvement (SMD from 0.40 to −1.40) (Baumeister et al. 2012). Four recent trials on psychological interventions found an improvement in glycemic control (SMD from −0.25 to −0.68) (Petrak et al. 2015). Web-based psychological therapies appear less effective than face-to-face contact, particularly for glycemic outcomes (van der Feltz-Cornelis 2013). The most effective psychological interventions combine diabetes self-management education with psychological support (van der Feltz-Cornelis et al. 2010).
There are several classes of effective and well-tolerated antidepressants, and these form an integral component of depression management. Randomized clinical trials in people with diabetes have been undertaken for only a relatively small group of antidepressants, including nortriptyline, fluoxetine, bupropion, sertraline, paroxetine, and citalopram (Petrak et al. 2015); however, these trials show that these antidepressants improve depressive symptoms to a similar extent in people with diabetes and the general population. All antidepressants studied appear to have similar efficacy, and as long as adequate doses are used, the effect sizes are −0.61 SMD (Baumeister et al. 2012). However, there are gaps in our evidence base, with regard to glycemic control and the medium- and long-term sustainability of the pharmacological interventions after treatment cessation. Furthermore, a number of new antidepressants have recently been approved, including vilazodone, vortioxetine, and levomilnacipran, which have not been formally assessed in people with diabetes.
Given the comparable effectiveness, the treatment of choice depends largely on the side effect profile, patient preference, and individual response. Selective serotonin reuptake inhibitors (SSRIs) are widely regarded as first choice agents, because they are less cardiotoxic than the older tricyclic antidepressants and are safer in overdose.
Some antidepressants, including mirtazapine, paroxetine, and some tricyclic antidepressants, may cause unwanted weight gain (Serretti and Mandelli 2010), while bupropion, which is available in the USA but not in Europe, is associated with weight loss.
Several antidepressants may interact with oral hypoglycemic agents through inhibition of the cytochrome P450 3A4 and 2C9 isoenzyme. For example, the use of fluoxetine may potentiate the effect of sulfonylureas precipitating hypoglycemia (Musselman et al. 2003).
Antidepressant treatment should be continued at an adequate dose for at least 4–6 months after complete remission of depressive symptoms to reduce the risk of relapse and recurrence. This is particularly important in people with diabetes, in whom the risk of relapse and persistence of symptoms is greater than the general population.
Clinical trials demonstrate that SSRIs lead to a modest improvement in glycemic control (SMD −0.38), but there is a mixed effect on glycemic control with other antidepressants ranging from hyperglycemic effects with tricyclic antidepressant medications to euglycemic or slightly hypoglycemic effects with serotonin-noradrenaline reuptake inhibitors. The diversity of effect implies that any finding of improved glycemic control with individual antidepressants should not be extrapolated to other untried antidepressants (Petrak et al. 2015).
The treatment of depression may lead to a change in the patient’s behavior and routine requiring a change in diabetes self-management. For example, should the patient’s appetite improve, more insulin may be required; by contrast, if the patient becomes more physically active, less may be needed.
Models of Care
Many healthcare systems are poorly equipped to manage comorbidity, particularly where this involves mental and physical illness. The Cartesian split of mind and body affects the delivery of care, leading healthcare professionals to consider one or other illness rather than making holistic decisions. In order to overcome this, the late Wayne Katon and colleagues in Seattle pioneered a model of care, known as “Collaborative Care” (Katon et al. 2010). This model encourages interdisciplinary cooperation between healthcare providers and shared decision making to facilitate appropriate provision of evidence-based treatment options, regular follow-up, self-management training, and support for people with comorbid diabetes and depression. The first study of this model showed improvements in depression symptoms but no change in glycemic control (Katon et al. 2004); however, in later studies, greater attention was paid to diabetes and blood pressure interventions, leading to improved biomedical outcomes as well as improved depressive symptoms (Katon et al. 2010). These models of care are also highly cost-effective (Simon et al. 2001, 2007).
Cognitive function can be defined as an intellectual process by which one becomes aware of, perceives, or comprehends ideas. It involves aspects of perception, thinking, reasoning, and remembering. Given the multifaceted nature of cognition, a full assessment of cognitive functioning requires a battery of psychometric tests to obtain an overview of an individual’s thinking skills. These tests assess how a person processes information, reasons, and learns in different ways. Cognitive function may be impaired globally or specific components of cognitive function may be affected.
Diabetes and Cognition
Cognitive Dysfunction in Children and Adolescents
The age of onset of diabetes is a major determining factor for its effect on cognitive function in children. In those diagnosed before the age of 7 years, there is an increased risk of developing cognitive impairment across a range of cognitive domains, including attention, mental flexibility, psychomotor efficiency, learning, memory, problem-solving, and overall intelligence, and which start to be apparent within 2–3 years of diagnosis (Biessels et al. 2008; Gaudieri et al. 2008; Northam et al. 2001, 2006; Hershey et al. 2004). For children diagnosed after this age, the effect is much more modest and is confined primarily to overall intelligence and speed-related tasks, particularly those with a visual-perceptual aspect (Gaudieri et al. 2008). The cognitive impairment seen in children diagnosed before the age of 7 years persists into adulthood and manifests as lower IQ scores and slower information processing (Ferguson et al. 2005).
Academic achievement is lower in children with diabetes, irrespective of the age of diagnosis, but this could relate to school absences or hypoglycemia interfering with learning as much as a direct effect on the developing brain (McCarthy et al. 2003). However, there is evidence from older studies that severe hypoglycemia causes neuropsychological deficits, particularly in children whose onset of diabetes occurred below the age of 6 years (Ryan et al. 1985; Rovet and Ehrlich 1999). Part of the problem is that younger children may not be able to describe their hypoglycemic symptoms, thereby increasing the likelihood of prolonged and severe hypoglycemia. Some support for the hypoglycemia hypothesis comes from a meta-analysis of data from 441 children with recurrent severe hypoglycemia and 560 children without recurrent severe hypoglycemia (Blasetti et al. 2011). This study found that those with recurrent severe hypoglycemia had a modestly reduced performance in the domains of intelligence, language, and memory and learning but motor speed was unaffected (Blasetti et al. 2011). This effect appears to be limited to children, because there was no difference in the cognitive function in adults with type 1 diabetes in the Diabetes Control and Complications Trial with and without a history of severe hypoglycemia after an 18-year follow-up period (Jacobson et al. 2007). Similarly, the Action to Control Cardiovascular Risk in Diabetes-Memory in Diabetes (ACCORD-MIND) study, which recruited adults with type 2 diabetes, found no difference in cognitive function after 20-month and 40-month follow-up, respectively, between those treated intensively compared with standard care despite a threefold increase in the rates of hypoglycemia (Launer et al. 2011).
Cognitive Dysfunction in Adults with Diabetes
Adults with type 1 diabetes show modest nonprogressive cognitive deficits in measures of intelligence, attention, psychomotor speed, cognitive flexibility, and visual perception, but measures of language, learning, and memory are unaffected despite long duration of diabetes (Brands et al. 2005, 2006). With the exception of “crystallized intelligence,” the affected domains require a rapid response indicating that diabetes affects mental agility rather than accuracy.
These changes are accompanied by structural changes that are characterized by reduced gray matter volumes in the frontal lobe and the adjacent supramarginal and postcentral gyri (Hughes et al. 2013). These MRI alterations have been linked to disrupted integrity of fiber tracts connecting the main cortical areas of the brain (Koekkoek et al. 2015). Functional studies have demonstrated altered cerebral perfusion with both decreased and increased blood flow (Kodl et al. 2008; Franc et al. 2011; van Duinkerken et al. 2012).
Adults with type 2 diabetes also show mild cognitive deficits affecting memory, processing speed, and executive function, which may lead to the individual to be less able to process unstructured information (van den Berg et al. 2009; Palta et al. 2014). Interestingly, similar deficits are also present in people with newly diagnosed type 2 diabetes and impaired glucose regulation as well as those with features of the metabolic syndrome (Crichton et al. 2012; Lamport et al. 2009; Ruis et al. 2009). Type 2 diabetes is also associated with structural changes in the brain, characterized by a loss of gray matter in the medial temporal, anterior cingulate, and medial frontal lobes, while white matter is lost in the frontal and temporal regions, respectively (Moran et al. 2013).
The mechanisms underlying these changes are not fully understood although studies have not consistently demonstrated an association with cerebral small vessel disease (Koekkoek et al. 2015). The link between glycemic control and cognitive dysfunction in adults with diabetes is less obvious than for children. As previously described, hypoglycemia does not appear to be a major risk factor for cognitive decline in young- and middle-aged adults with diabetes, but hyperglycemia appears to be more important, at least in the case of type 1 diabetes. A recent meta-analysis found a weak negative association between HbA1c and cognitive function, with HbA1c explaining at most 10–15% of the variance in cognitive function (Geijselaers et al. 2015).
The presence of microvascular complications, especially retinopathy and nephropathy, is associated with accelerated cognitive decline (Ryan et al. 2003; Jacobson et al. 2011). Cardiovascular disease and its risk factors are also associated with cognitive decline as they are in the general population, but whether there is any specific interaction with diabetes is uncertain (Koekkoek et al. 2015; Ryan et al. 2003).
Causes of dementia in people with diabetes
Lewy body dementia
Diabetes and Dementia
The rate of cognitive decline in older individuals with type 2 diabetes appears to be up to twofold quicker than the general population, and a number of studies have indicated that the risk of dementia is increased by approximately 50% in people with diabetes (Cheng et al. 2012). The commonest cause of dementia in people with diabetes is Alzheimer’s disease, which is increased by 46%, while vascular dementia is increased 2.38-fold (Cheng et al. 2012). It is estimated that the diabetes-attributable risk of dementia is 6–7%; in other words, 1 in 15 cases of dementia is attributable to diabetes (Koekkoek et al. 2015). An increased risk of dementia has also been reported in people with prediabetes and metabolic syndrome (Crichton et al. 2012). Diabetes worsens the outcome for people with dementia and is associated with 90% increase in mortality compared with those without diabetes (Zilkens et al. 2013).
Risk factors for Alzheimer’s disease and vascular dementia
Previous coronary heart disease, stroke of transient ischemic event
Unlike in younger adults, recurrent hypoglycemia is an important risk factor for cognitive decline in patients with dementia (Whitmer et al. 2009; Punthakee et al. 2012). In a retrospective study of 16,667 older people with diabetes, a history of one, two, and three or more severe hypoglycemic episodes increased the risk of dementia by 26%, 80%, and 94%, respectively, independent of glycemic control, medications, and other comorbidities (Whitmer et al. 2009). The relationship, however, appears to be bidirectional as cognitive dysfunction increases risk of hypoglycemia, thereby, creating a vicious cycle.
Impaired insulin signaling in the brain may also play a role in the development of Alzheimer’s disease (Li et al. 2015). Briefly, the pathology of Alzheimer’s disease is characterized by the deposition of β-amyloid plaques and tangles in the brain (Turner et al. 2003; Niedowicz et al. 2011). β-amyloid is derived from the cleavage of amyloid precursor protein, an extracellular protein which is critical to neuron growth, survival, and post-injury repair. The formation of β-amyloid is regulated with several genes but is increased in aging and by obesity, and less prudent diet (Niedowicz et al. 2011). Insulin inhibits the production and accumulation of β-amyloid, and so in situations where there is insulin deficiency or impaired action, β-amyloid accumulates (Li et al. 2015). Supporting this hypothesis is the observation of reduced brain insulin and its receptor in people with Alzheimer’s disease.
Clinical Implications of Dementia in People with Diabetes
Prevention of Dementia
Several large randomized controlled trials in both type 1 diabetes and type 2 diabetes have investigated whether improved glycemic control can reduce the risk of dementia. To date, no difference in the rate of cognitive decline, cognitive performance, or incidence of dementia has been seen in those with tighter glycemic control (Jacobson et al. 2007; Launer et al. 2011; Koekkoek et al. 2012; de Galan et al. 2009). The ACCORD-MIND study also assessed whether better blood pressure and lipid control could improve cognition but again, no benefit was seen in those treated with a combination of statin and fibrate while intensive blood pressure lowering actually accelerated brain atrophy (Williamson et al. 2014).
Diagnosis of Dementia
Once dementia develops, the person’s ability to self-manage their diabetes progressively deteriorates. Exercise and diet appear to be particularly affected, but the real danger lies in the increased risk of adverse events in people taking hypoglycemic drugs. The risk of hypoglycemia is exacerbated further because dementia may impair language and lead to disorientation and personality changes, all of which may mimic the symptoms of hypoglycemia (Sinclair et al. 2010). When lack of engagement with self-management occurs in older people, clinicians should consider cognitive dysfunction as a cause.
A risk score which includes age, microvascular disease, diabetic foot, cerebrovascular disease, cardiovascular disease, acute metabolic events, depression, and education has been developed to help predict the risk of developing dementia. In a prospective study over 10 years, the risk of developing dementia was 5.3% in those with the lowest score compared with 73.3% for the top scores (Exalto et al. 2013). Other authors have advocated the use of the easy-to-perform Mini-Cog test as a simple screening tool for dementia. This test has a sensitivity of 86.4% and a specificity of 91.1% (Sinclair et al. 2013).
A formal diagnosis of dementia involves a combination of history, examination, including tests of mental investigation, and investigation. It is important to obtain corroborating information from a friend or relative who knows the patient well. Unfortunately, the diagnosis is not always straightforward particularly in the early stages, and time may be an important diagnostic tool.
Treatment of Dementia
A detailed description of the treatment of Alzheimer’s disease is beyond the scope of this chapter as treatment is usually initiated in specialist memory clinics. The most commonly used drugs are the acetylcholinesterase inhibitors, such as donepezil, which are licensed for the treatment of mild-to-moderate Alzheimer’s disease (Cummings et al. 2015). This class of agents can cause bradycardia and exacerbate symptoms of gastric and duodenal ulcers, precipitate bronchospasm, and may cause convulsions. Dizziness, headache, and nausea are the most common side effects with abdominal disturbance and nightmares being also reported. Because of the cardiovascular effects, they should be used with caution in those whose baseline pulse is <60 beats per minute. An ECG is recommended if the pulse is <70 beats per minute or if it is irregular.
Given the relationship between cerebral insulin action and Alzheimer’s disease, the use of several antidiabetes drugs has been examined in people with dementia (Koekkoek et al. 2015; Li et al. 2015). Small studies have suggested that intranasal insulin improves delayed memory and cognitive function (Reger et al. 2008; Craft et al. 2012), but the results of larger ongoing trials of people with Alzheimer’s disease, with and without diabetes, are awaited (Koekkoek et al. 2015). Studies of thiazolidinediones have shown mixed effects on cognitive function but a possible benefit has been reported in those whose genotype is APOE-ε4-negative (Watson et al. 2005; Risner et al. 2006; Gold et al. 2010). Besides insulin, glucagon-like peptide 1 (GLP-1) also appears to play an important role in the control of synaptic plasticity and in some forms of memory formation, and two trials of GLP-1 receptor agonists are ongoing (Li et al. 2015).
Management of Diabetes in Someone with Dementia
As dementia affects self-management, it is important that glycemic targets are adjusted appropriately taking into account overall health and life expectancy. Several guidelines now recommend HbA1c targets of <8.5% (69 mmol/mol) for older, dependent people with dementia (Sinclair et al. 2012). In these individuals, avoidance of symptomatic hypoglycemia is more important than tight glycemic control, which may reduce the quality of life. Glycemic targets should be regularly reviewed and medications should be prescribed as appropriate.
Many older people with diabetes and dementia are overtreated with multiple drugs. In a retrospective cohort study of 15,880 people with type 2 diabetes and dementia from the Veterans Affairs Healthcare System, 52% of participants had an HbA1c <7% (53 mmol/mol), and within this group, 75% were treated either with sulfonylureas, insulin, or both placing them at a high risk of hypoglycemia (Thorpe et al. 2015). Declining weight, malnutrition, and frailty may lead to a reduced need for antidiabetes medications, and these have been safely withdrawn in several studies (Sjoblom et al. 2008) without a deterioration of their glycemic control (Abdelhafiz et al. 2014). Drugs with a lower risk of hypoglycemia should be preferentially used.
This chapter has highlighted some of the many and various ways in which diabetes interacts with the brain. It is clear that these connections can have a major impact on diabetes outcomes, and health professionals who work with people with diabetes, require good knowledge and awareness of these issues to be able to provide optimal care. There is, clearly, also a great need for closer working between diabetes services and mental health services while further research on these topics is also required.
Richard Holt has received fees for lecturing and consultancy from the following companies: Eli Lilly, Janssen, Lundbeck, Novo Nordisk, Novartis, Otsuka, Sanofi, Sunovion, Takeda, and MSD.
- American Psychiatric Association. Diagnostic and statistical manual of mental disorders. 4th ed. Washington, DC: American Psychiatric Association; 2005.Google Scholar
- de Galan BE, Zoungas S, Chalmers J, Anderson C, Dufouil C, Pillai A, et al. Cognitive function and risks of cardiovascular disease and hypoglycaemia in patients with type 2 diabetes: the Action in Diabetes and Vascular Disease: Preterax and Diamicron Modified Release Controlled Evaluation (ADVANCE) trial. Diabetologia. 2009;52(11):2328–36.PubMedCrossRefPubMedCentralGoogle Scholar
- First MB, Spitzer RL, Gibbon M, Williams JB. Structured Clinical Interview for DSM-IV-TR Axis I Disorders, research version, patient edition. (SCID-I/P). New York: Biometrics Research, New York State Psychiatric Institute; 2002.Google Scholar
- Gold M, Alderton C, Zvartau-Hind M, Egginton S, Saunders AM, Irizarry M, et al. Rosiglitazone monotherapy in mild-to-moderate Alzheimer’s disease: results from a randomized, double-blind, placebo-controlled phase III study. Dement Geriatr Cogn Disord. 2010;30(2):131–46.PubMedPubMedCentralCrossRefGoogle Scholar
- International Diabetes Federation. Global guideline for type 2 diabetes. 2012. http://www.idf.org/guideline-type-2-diabetes. Last accessed 17 Sept 2016.
- Kodl CT, Franc DT, Rao JP, Anderson FS, Thomas W, Mueller BA, et al. Diffusion tensor imaging identifies deficits in white matter microstructure in subjects with type 1 diabetes that correlate with reduced neurocognitive function. Diabetes. 2008;57(11):3083–9.PubMedPubMedCentralCrossRefGoogle Scholar
- Koekkoek PS, Ruis C, van den Donk M, Biessels GJ, Gorter KJ, Kappelle LJ, et al. Intensive multifactorial treatment and cognitive functioning in screen-detected type 2 diabetes – the ADDITION-Netherlands study: a cluster-randomized trial. J Neurol Sci. 2012;314(1–2):71–7.PubMedCrossRefPubMedCentralGoogle Scholar
- Launer LJ, Miller ME, Williamson JD, Lazar RM, Gerstein HC, Murray AM, et al. Effects of intensive glucose lowering on brain structure and function in people with type 2 diabetes (ACCORD MIND): a randomised open-label substudy. Lancet Neurol. 2011;10(11):969–77.PubMedPubMedCentralCrossRefGoogle Scholar
- Lyoo IK, Yoon SJ, Musen G, Simonson DC, Weinger K, Bolo N, et al. Altered prefrontal glutamate-glutamine-gamma-aminobutyric acid levels and relation to low cognitive performance and depressive symptoms in type 1 diabetes mellitus. Arch Gen Psychiatry. 2009;66(8):878–87.PubMedCrossRefPubMedCentralGoogle Scholar
- National Institute for Health and Care Excellence. Type 1 diabetes in adults: diagnosis and management, NICE guidelines (NG17). 2015a. https://www.nice.org.uk/guidance/ng17. Last accessed 17 Sept 2016.
- National Institute for Health and Care Excellence. Type 2 diabetes in adults: management. NICE guidelines (NG28). 2015b. https://www.nice.org.uk/guidance/ng28. Last accessed 17 Sept 2016.
- Nouwen A, Nefs G, Caramlau I, Connock M, Winkley K, Lloyd CE, et al. Prevalence of depression in individuals with impaired glucose metabolism or undiagnosed diabetes: a systematic review and meta-analysis of the European Depression in Diabetes (EDID) Research Consortium. Diabetes Care. 2011;34(3):752–62.PubMedPubMedCentralCrossRefGoogle Scholar
- Palta P, Schneider AL, Biessels GJ, Touradji P, Hill-Briggs F. Magnitude of cognitive dysfunction in adults with type 2 diabetes: a meta-analysis of six cognitive domains and the most frequently reported neuropsychological tests within domains. J Int Neuropsychol Soc. 2014;20(3):278–91.PubMedPubMedCentralCrossRefGoogle Scholar
- Sinclair A, Morley JE, Rodriguez-Manas L, Paolisso G, Bayer T, Zeyfang A, et al. Diabetes mellitus in older people: position statement on behalf of the International Association of Gerontology and Geriatrics (IAGG), the European Diabetes Working Party for Older People (EDWPOP), and the International Task Force of Experts in Diabetes. J Am Med Dir Assoc. 2012;13(6):497–502.PubMedCrossRefPubMedCentralGoogle Scholar
- Sjoblom P, Tengblad A, Lofgren UB, Lannering C, Anderberg N, Rosenqvist U, et al. Can diabetes medication be reduced in elderly patients? An observational study of diabetes drug withdrawal in nursing home patients with tight glycaemic control. Diabetes Res Clin Pract. 2008;82(2):197–202.PubMedCrossRefPubMedCentralGoogle Scholar
- van der Feltz-Cornelis CM, Nuyen J, Stoop C, Chan J, Jacobson AM, Katon W, et al. Effect of interventions for major depressive disorder and significant depressive symptoms in patients with diabetes mellitus: a systematic review and meta-analysis. Gen Hosp Psychiatry. 2010;32(4):380–95.PubMedCrossRefPubMedCentralGoogle Scholar
- van Duinkerken E, Schoonheim MM, Sanz-Arigita EJ, IJzerman RG, Moll AC, Snoek FJ, et al. Resting-state brain networks in type 1 diabetic patients with and without microangiopathy and their relation to cognitive functions and disease variables. Diabetes. 2012;61(7):1814–21.PubMedPubMedCentralCrossRefGoogle Scholar
- van Steenbergen-Weijenburg KM, van Puffelen AL, Horn EK, Nuyen J, van Dam PS, van Benthem TB, et al. More co-morbid depression in patients with type 2 diabetes with multiple complications. An observational study at a specialized outpatient clinic. Diabet Med. 2011;28(1):86–9.PubMedCrossRefPubMedCentralGoogle Scholar
- Watson GS, Cholerton BA, Reger MA, Baker LD, Plymate SR, Asthana S, et al. Preserved cognition in patients with early Alzheimer disease and amnestic mild cognitive impairment during treatment with rosiglitazone: a preliminary study. Am J Geriatr Psychiatry. 2005;13(11):950–8.PubMedPubMedCentralGoogle Scholar
- Williamson JD, Launer LJ, Bryan RN, Coker LH, Lazar RM, Gerstein HC, et al. Cognitive function and brain structure in persons with type 2 diabetes mellitus after intensive lowering of blood pressure and lipid levels: a randomized clinical trial. JAMA Intern Med. 2014;174(3):324–33.PubMedPubMedCentralCrossRefGoogle Scholar
- Willis T. Pharmaceutice rationalis sive diabtriba de medicamentorum operantionibus in humano corpore. Oxford; 1675.Google Scholar
- World Health Organisation. Schedule for Clinical Assessment in Neuropsychiatry 2.1. 1999. http://whoscan.org/wp-content/uploads/2014/10/xinterview.pdf. Last accessed 17 Sept 2016.