Experts address mechanisms, link between cognitive decline and diabetes

Patients with diabetes have a greater risk for developing depression, mild cognitive impairment, dementia and Alzheimer’s disease, recent epidemiologic studies suggest.

According to researchers, the prevalence of diabetes has reached epidemic proportions, and neurodegenerative diseases also are on the rise. Endocrine Today spoke with neuroscientists and other experts about their research, which focuses on the potential mechanisms behind this deleterious link to develop preventive or restorative therapies.

In a paper by David S. Knopman, MD, of the department of neurology, and Rosebud O. Roberts, MBChB, of the division of epidemiology in the department of health sciences research, both at Mayo Clinic in Rochester, Minn., and the Alzheimer’s Disease Research Center, the role of cognitive disorders in older patients with diabetes, hypertension and obesity was reviewed.

Middle-age and geriatrics

Roberts and Knopman said the associations of hypertension and obesity with cognitive impairment in late-life are attenuated or reversed, and the role of vascular risk factors in midlife should be the focus of public health efforts to reduce the burden of late-life cognitive impairment. With type 2 diabetes, however, the association with late-life brain pathology or cognitive impairment persisted with onset either in mid-life or late-life.

“One of the things that has come to the attention of researchers with regard to vascular risk factors and cognitive impairment is that certain vascular risk factors behave differently in their association with late-life cognitive impairment when you measure them in mid-life vs. late-life,” Roberts told Endocrine Today. “Researchers have found that high BP, BMI and lipid levels between the ages of 40 through 65 years are associated with an increased risk for dementia or cognitive decline later in life.” However, high levels measured in late-life are not associated with cognitive impairment.

According to Roberts, there could be a number of reasons for why these levels and their connection to cognitive decline in late-life could vary. “We don’t fully understand what is going on in the brain, but there are a few hypotheses,” she said.

For example, according to Roberts, hypertension in late-life could have a positive effect on the brain by allowing more blood flow to the brain cells, whereas lower BP later in life can reduce blood supply to brain cells. “In regard to lipids, high levels in late-life may simply reflect better health unrelated to cognition, or may play a role in maintaining brain cells.” she said.

“With obesity specifically, we’re not sure again what’s happening in late-life, but what people have found is that people (in particular women) who have unintended weight loss in late-life tend to develop dementia,” Roberts said. “With diabetes, our studies showed that a history of type 2 diabetes is associated with an increased risk for mild cognitive impairment (MCI), and the association was stronger for those with more severe diabetes, poor control of diabetes or longer duration of diabetes. This suggests that both prevention and control of type 2 diabetes are important measures for reducing the occurrence of cognitive impairment and dementia in late-life.”

In June, Kristine Yaffe, MD, of the University of California, and colleagues published a paper in Archives of Neurology. In their study, which began in 1997, researchers examined the association between cognitive decline and diabetes in community-dwelling white and black adults living in Memphis, Tenn., or Pittsburgh. The mean age of the patients at baseline was 74.2 years.

Of the 3,075 patients in the study, 717 (23.4%) had prevalent diabetes, which was associated with black males with less high school education, as well as a history of myocardial infarction or hypertension and a higher BMI. Moreover, 159 (5.2%) patients developed diabetes during follow-up, researchers said.

At baseline, patients with prevalent diabetes had lower unadjusted modified mini-mental state examination (3MS) and digit symbol substitution test (DSST) scores compared with patients without diabetes (P=.001). After adjustments, patients with prevalent diabetes compared with those without diabetes had a slightly lower mean 3MS baseline score (P=.01) and DSST score (P=.03).

After 9 years, patients with prevalent diabetes had a significant decline on both the 3MS (P=.008) and the DSST (P=.001) compared with patients without diabetes, according to data. The decline was still significant after adjustments for age, race, sex and education (P=.001). HbA1c levels among patients with diabetes were associated with lower average cognitive scores on the 3MS (P=.003) and DSST (P=.04), even after multivariate adjustment, they wrote.

Yaffe and colleagues suggested further studies to determine whether early diagnosis and treatment of diabetes will decrease the risk for developing cognitive decline.

Importance of diet, exercise

“Dietary factors also play a role in many of the vascular risk factors because our diet has an important impact on whether we are going to be obese, diabetic or hypertensive,” Roberts said.

In a study by Geda and colleagues, also at the Mayo Clinic, caloric intake and its effect on MCI were examined among the Mayo Clinic Study of Aging in Olmsted County, Minn. Participants aged 70 to 92 years were included (1,072 were labeled cognitively normal, and 161 MCI). According to data, analyses were completed in tertiles of caloric intake: 600 kcal to <1,526 kcal per day (reference group); 1,526 kcal to 2,143 kcal per day (moderate caloric intake group); and >2,143 kcal per day (high caloric intake group).

They found that high caloric intake was associated with nearly twofold increased odds of having MCI (OR=1.96; 95% CI, 1.26-3.06) compared with the reference group. However, this was not statistically significant among the moderate caloric intake group.

“If we can get people to have a healthy diet, it could have long-term implications for vascular disease as well as for cognitive impairment. In addition to a healthy diet, there is a need for regular exercise, which reduces the risk of obesity, diabetes, and hypertension. These two things, diet and exercise, may be key to our well-being, and somehow we need to get that message across to people,” Roberts said.

Adolescence, diabetes and depression

“We have plenty of evidence to know that we need to be doing more for our patients with diabetes, particularly from a pediatric standpoint because they’re at higher risk for depression and anxiety, and their parents are as well,” clinical and health psychologist Korey K. Hood, PhD, associate professor of pediatrics at the University of California, San Francisco, told Endocrine Today. “There’s a bit of work that’s been done on ‘diabetes burnout.’ It’s not going to end up in any diagnostic manual, but it looks a lot like depression but is largely the result of diabetes.”

Hood’s research interests include resilience promotion in teenagers with type 1 diabetes, improving self-management of type 1 diabetes and enhancing adolescent and parent involvement in diabetes adherence problem-solving.

Hood and colleagues published a study in Diabetes Care that suggests that depression reported by youth with diabetes is partially associated with metabolic abnormalities and systemic inflammation. They utilized data from the Search for Diabetes in Youth Study (SEARCH), an observational study of US children aged younger than 20 years who were diagnosed with diabetes.

They found that six of eight markers were linked to depression in young patients with diabetes (P<.006), according to bivariate analyses. Higher levels of depression were linked to indicators of worse metabolic or inflammatory functioning in general, they wrote.

“We need to do more work on screening and identifying problems. Programs like systematic screening for depression and other psychological factors are important areas in both the patient and the parent,” Hood said. “There’s a large literature on effectively treating depression and anxiety in adolescents, but specifically within adolescents with diabetes, there’s very little. We don’t know the best interventions; we need more research to understand what is best.”

Evidence-based data

Hood and colleagues are joined by other researchers who have examined this bidirectional association between depression and diabetes during the past 10 years. In 2003, findings from the British Women’s Heart and Health Study suggested an association between insulin resistance and depression. And in 2008, Sherita Hill Golden, MD, MHS, of Johns Hopkins University School of Medicine, and colleagues used the Multi-Ethnic Study of Atherosclerosis (MESA) to examine the link. They found that impaired fasting glucose and untreated type 2 diabetes were inversely associated with incident depressive symptoms.

More recently, Pei-Chun Chen, PhD, of the Institute of Epidemiology and Preventive Medicine at the National Taiwan University College of Public Health, and colleagues examined two cohorts using the National Health Insurance claims in 2000. According to data, the first cohort analysis suggested an incidence density of 7.03 per 1,000 person-years for depression in those with diabetes, and 5.04 per 1,000 person-years for those without diabetes (HR=1.43; 95% CI, 1.16-1.77).

The second analysis suggested an incidence density of 27.59 per 1,000 person-years for diabetes in those with depression, and 9.22 per 1,000 person-years for diabetes in those without depression. The researchers wrote that the covariate-adjusted HR was greater for incident diabetes linked to baseline depression (HR=2.02; 95% CI, 1.80-2.27).

There has been some evidence that maternal obesity and diabetes is associated with autism and other neurodevelopmental disorders. According to an article published in Pediatrics, researchers from the University of California Davis Health System examined whether metabolic conditions during pregnancy (ie, diabetes, hypertension and obesity) are linked to autism spectrum disorder, developmental delays or impaired development. Children aged 2 to 5 years (n=517, autism spectrum disorder; n=172, developmental delays; and n=315, controls) were included in the Childhood Autism Risks from Genetics and the Environment (CHARGE) study that took place between January 2003 and June 2010.

Researchers found that maternal metabolic conditions could be broadly linked to neurodevelopmental problems in children. Data from the study indicated diabetes, hypertension and obesity were linked to a higher likelihood of autism spectrum disorder (OR=1.61; 95% CI, 1.10-2.37) and developmental delays (OR=2.35; 95% CI, 1.43-3.88), relative to controls. These findings, coupled with the increased prevalence of diabetes and obesity, are evidence for concern, researchers wrote.

Components of decline

In a review published in the Journal of Alzheimer’s Disease, José A. Luchsinger, MD, MPH, of Columbia University College of Physicians and Surgeons, and the Joseph P. Mailman School of Public Health at Columbia University, addressed the mechanisms behind the connection between type 2 diabetes and cognitive impairment.

Luchsinger wrote that there is little clarification to whether type 2 diabetes can cause late-onset Alzheimer’s disease, but the link between vascular cognitive impairment and late-onset Alzheimer’s disease is much greater.

“Type 2 diabetes and its related conditions are known to cause cerebrovascular disease. Elevated adiposity, hyperinsulinemia, type 2 diabetes and their clustering with other vascular risk factors are risk factors for stroke,” he wrote. “In addition, insulin or diabetes related by-products may affect the amyloid cascade. Thus, we classify the mechanisms linking this continuum with cognitive impairment as cerebrovascular and non-cerebrovascular.”

Cerebrovascular mechanisms include brain infarcts such as strokes that could lower the threshold of amyloid in the brain that is necessary to cause dementia, Luchsinger wrote. Another cerebrovascular mechanism is white matter disease, or white matter hyperintensities.

Although the nature of this disease is controversial, Luchsinger wrote that some white matter hyperintensities could be related to amyloid burden and late-onset Alzheimer’s disease. Non-cerebrovascular mechanisms include hyperinsulinemia, advanced glycosylation end products and lipoprotein receptor-related proteins, he added. Furthermore, Luchsinger wrote that type 2 diabetes has been related to a twofold higher risk for developing MCI among postmenopausal women.

According to a more recent paper by Limor Raz, PhD, and colleagues at the Mayo Clinic in Rochester, Minn., data from the Kronos Early Estrogen Prevention Study (KEEPS) demonstrated that in vivo platelet activation could play a role in events that lead to the development of white matter hyperintensity in recently menopausal women. The research was published in Neurology.

Other literature has suggested that cognitive dysfunction in patients with diabetes could be caused by a number of mechanisms, including absence of C-peptide, insulin resistance, hyperglycemia-induced end organ damage (or microvascular disease), macrovascular disease, hypoglycemia and/or the absence of the apolipoprotein E4 allele.

Diabetes therapies for MCI

Kelly Dineley, PhD, associate professor in the department of neurology, and Larry Denner, PhD, professor in the division of endocrinology, department of internal medicine, and associate director of research at the Nelda C. and H.J. Lutcher Stark Diabetes Center and director of the Miriam and Emmett McCoy Diabetes Mass Spectrometry Laboratory, both at the University of Texas Medical Branch in Galveston, have been working with genetically engineered mice as models for Alzheimer’s disease. The researchers spoke with Endocrine 
Today about their research on the insulin sensitizer rosiglitazone (Avandia, GlaxoSmithKline), a thiazolidinedione (TZD) that is an agonist of the peroxisome proliferator-activated receptor gamma (PPARγ) initially used to treat insulin resistance in patients with diabetes. This transcription factor is a validated, druggable target in Alzheimer’s disease.

Dineley and Denner said rosiglitazone enhanced learning and memory, reducing the negative effects of the activity of the extracellular signal-regulated kinase (ERK) on Alzheimer’s disease. ERK, which is essential to learning and memory, became hyperactive both in the brain of Alzheimer’s patients and in the mice at a disease stage correlating to MCI in humans.

“We concluded that the diabetes drug treatment was actually reigning in a dysregulated ERK system to allow the animals to function more like a normal wild-type litter mate,” Dineley said.

Rosiglitazone brought ERK back to an appropriate level by activating the peroxisome proliferator-activated receptor gamma (PPAR-gamma) pathway, according to Dineley, who discovered ERK dysfunction in the Alzheimer’s mouse model several years ago. Besides rosiglitazone, Dineley and Denner also have studied the effects of another insulin sensitizing drug, metformin, in enhancing memory by activating the AMPK pathway. Using systems biology proteomics, genomics and bioinformatics approaches, the researchers have discovered the emergent properties of how the ERK and PPARγ pathways converge in memory enhancement through the transcription regulatory factor cyclic AMP response element binding protein (CBP). The researchers are now using systems biology to discover the emergent properties of next generation PPARγ agonists that lack the adverse effects profile that have limited the clinical utility of TZDs.

“Intervening in early stage disease like MCI gives a much better therapeutic opportunity to at least stall, if not, ideally, reverse the cognitive consequences,” Denner said.

GH for improved cognitive function

Laura D. Baker, PhD, associate professor of internal medicine and public health science at Wake Forest School of Medicine in Winston-Salem, N.C. and colleagues at the University of Washington in Seattle have examined the effects of growth hormone-releasing hormone (GHRH) on cognitive function among healthy older adults and in adults with MCI.

“We’ve been studying the link between metabolic dysfunction associated with type 2 diabetes and cognitive decline now for many years. In particular, we have looked at the link between metabolic dysfunction associated with type 2 diabetes and the development of Alzheimer’s disease. Our primary outcome when we study this relationship is cognition,” Baker told Endocrine Today.

In a randomized, double blind, placebo-controlled trial of growth hormone releasing hormone (GHRH), a key regulator of insulin-like growth factor-1 (IGF-1) (similar in molecular structure to insulin), Baker and colleagues enrolled 152 adults (66 with MCI) aged 55 to 87 years (mean age, 68 years); 137 (76 healthy, 61 with MCI) completed the trial. According to data, patients self-administered once-daily subcutaneous injections of tesamorelin (Egrifta, Theratechnologies, Inc.), a stabilized analogue of human GHRH (1 mg per day), or placebo 30 minutes before bedtime for 20 weeks.

“This study is the second to look at GHRH effects on cognitive enhancement. The first study was conducted by Michael Vitiello, PhD, of the University of Washington, which was published in 2006. The goal of that study was to look at how changing growth hormone levels in older adults could improve the quality of sleep,” Baker said.

Baker and colleagues wrote that the intent-to-treat analysis indicated a beneficial effect of GHRH on cognition (P=.03), which was comparable for adults with MCI and healthy older adults. Follow-up analyses to examine GHRH effects on specific cognitive abilities indicated benefits in executive function (P=.005) and verbal memory (P=.08). Circulating levels of IGF-1 increased by 117% (P<.001) as a result of GHRH treatment, and reduced body fat by 7.4% (P<.001), they wrote.

“The adverse events were very mild. The worst adverse events people experienced included water retention and carpal tunnel symptoms, and most of these events were reported by the women – a sex difference that may reflect unequal dose relative to lean muscle mass,” Baker said.

She said that these findings are encouraging but still need to be replicated in larger trials. Baker also said that one downside of GHRH administration is the expense; 1 daily dose can cost as much as $45.

Future directions

“With regard to the use of GHRH administration to improve cognition, I plan to continue this work in hopes of identifying an efficacious intervention strategy for older adults at risk for cognitive decline,” Baker said.

Although research is ongoing, Shai Efrati, MD, of Tel Aviv University’s Sackler Faculty of Medicine, has looked into hyperbaric treatment as a restorative option for patients with a chronically damaged brain tissue, according to a press release.

Efrati and colleagues enrolled post-stroke patients (n=74) whose permanent disabilities did not show signs of improvement. They categorized the patients into two groups: hyperbaric oxygen therapy (HBOT) and a control group. After 40 2-hour sessions five times per week in high pressure chambers, results indicated HBOT led to significant improvements in brain function.

“It is now understood that many brain disorders are related to inefficient energy supply to the brain,” Efrati said in the press release. “HBOT treatment could right such metabolic abnormalities before the onset of full dementia, where there is still potential for recovery.”

To develop future therapies in an increasingly vulnerable population, Luchsinger recommends clarifying whether type 2 diabetes is related to amyloid-driven cognitive impairment or to cerebrovascular disease.

“It’s important to educate patients and the general population about risk factors for cognitive impairment in late-life, and help them figure out ways in which they can alter their cognitive trajectory,” Roberts said. – by Samantha Costa

References:

Are neuropsychological tests for cognitive decline warranted in patients with prediabetes and diabetes?

Yes.

The prevalence of prediabetes is increasing dramatically worldwide, although most cases are preventable with healthy lifestyle changes and some can even be reversed. Diabetes and prediabetes states have consistently been shown to increase risk factors for cognitive decline, mild cognitive impairment and dementia (eg, vascular dementia and Alzheimer’s disease). In fact, fluctuations in the amount of glucose reaching the brain, a situation closely associated with impaired insulin signaling; cause structural, functional and biochemical changes in the brain. Because neuropsychological tests may detect even a mild cognitive impairment, neuropsychological testing can have an important role in identifying individuals at risk of developing dementia. This information is important for the physician to provide the most appropriate treatment for the patient to prevent dementia.

Paula I. Moreira, PhD, is an assistant professor of physiology in the faculty of medicine at the University of Coimbra, and principal investigator at the Center for Neuroscience and Cell Biology at the University of Coimbra, Portugal. She reports no relevant financial disclosures. She can be reached at Institute of Physiology Faculty of Medicine, Center for Neuroscience and Cell Biology, University of Coimbra, 3000-354 Coimbra, Portugal; email: venta@ci.uc.pt.

Yes, but with some reservations.

Older adults with type 2 diabetes have a greatly increased risk for manifesting mild cognitive dysfunction, which is most often characterized by mental slowing, difficulty learning new information and recalling that information after a brief delay. Because these are exactly the same types of cognitive problems seen in patients in the early stages of dementia, it may make sense to conduct a comprehensive neuropsychological assessment — but only if certain conditions are met.

First, there should be evidence from the patient (and family members) that cognition seems to have changed appreciably in the recent past. Second, the patient should have a history of relevant diabetes-related risk factors. A long history of poor metabolic control has been linked most strongly to cognitive impairment, but several other predictors of cognitive decline have been identified. This includes multiple episodes of moderately severe hypoglycemia, hypertension, hypercholesterolemia, obesity, poor sleep quality and depression.

If these conditions are met and a neuropsychological assessment is conducted, what should one do with the results when mild cognitive impairment is detected? One approach is to use these findings to counsel patients to improve their health status. We know from several smaller studies that improvements in metabolic control are associated with improved cognition, and we also know from a growing body of research that reductions in BMI and increased physical activity can lead to marked improvements in both glycemic control and brain function. By monitoring cognition over time (perhaps with repeated cognitive assessments), one could thereby determine the extent to which any given intervention is effective. As a researcher, this approach makes much sense to me. However, it may not be realistic clinically, given the out-of-pocket costs that may be associated with repeated cognitive evaluations.

The downside of ordering a neuropsychological assessment is the very real possibility that the test results may be misinterpreted as evidence that the patient is on their way to developing clinically significant Alzheimer’s disease (AD). Unfortunately, at the present time, neuropsychologists are not able to identify a unique diabetes-associated neurocognitive phenotype that can be readily differentiate from what is seen in people with early AD. Being absent of an effective treatment for AD (or other degenerative dementias), this type of diagnosis can have horrific consequences for patients and their families; especially if the diagnosis turns out to be a false-positive.

Christopher M. Ryan, PhD, is professor of psychiatry at the University Of Pittsburgh School Of Medicine. He can be reached at the Western Psychiatric Institute and Clinic of UPMC, Department of Psychiatry, 3811 O’Hara St., Pittsburgh, PA 15213; email: ryancm@upmc.edu. He reports no relevant financial disclosures.