Research Emerging That Links Insulin Resistance to Cognitive Decline

Retrospective analyses1,2 of patient records have long

tended to find an increased risk of incidence of Alzheimer’s

disease among patients with type 2 diabetes

(T2D). In more recent years, as rates of T2D have soared, rates

of Alzheimer’s disease—although far lower, overall—have risen

at a similar rate, and a number of studies3,4 have found

evidence that suggests Alzheimer’s disease can be slowed or

even reversed by increasing either the amount of insulin in

the brain or the brain’s ability to respond to existing insulin.

This evidence has convinced many that diabetes can trigger

Alzheimer’s disease in many patients who would otherwise

have escaped it, or that Alzheimer’s disease is essentially

a type of diabetes that affects the brain, sometimes in

conjunction with T2D, sometimes on its own. Other observers

believe the evidence linking diabetes and Alzheimer’s

disease, to date, is interesting, but inconclusive—particularly

when one considers study results that find no connection between

the 2 conditions.

The symptoms of Alzheimer’s disease can be nearly identical

to the symptoms of vascular dementia, a form of cognitive

decline, that certainly can result directly from T2D’s effects

on the circulatory system or indirectly from T2D-related

strokes.5 Indeed, the only sure way to diagnose Alzheimer’s

disease is via autopsy, and studies that have looked for T2D

after confirming Alzheimer’s disease via autopsy, have not

found it nearly as frequently as population-based studies

would predict.6,7 According to Geert Jan Biessels, MD, PhD,

a professor of Neurology at the Rudolph Magnus Institute,

who spoke with Evidence-Based Diabetes Management (EBDM),

studies, that have used positron emission tomography (PET)

scans to compare amyloid protein levels in patients with and

without T2D, have not found diabetic patients to have the

sort of protein buildup that would indicate elevated risk of

Alzheimer’s disease.8

Many investigators are working to solve the apparent contradictions

by focusing their research on the pathways that

could connect the 2 diseases, and on the effort to see whether

any existing treatment for T2D can also treat Alzheimer’s disease.

9 There is hope, looking forward, that such treatments

could be a breakthrough in Alzheimer’s care.

That said, researchers have been investigating possible

links between T2D and Alzheimer’s disease for many years.

A review, published by Diabetes Care nearly 2 decades ago,10

gathered evidence from 19 controlled studies on patients with

T2D, and noted that 13 of those studies reported that diabetics

performed worse than controls on at least 1 aspect of cognitive

function. All the remaining studies, the review authors

noted, lacked the statistical power to find a relatively strong

gap in cognitive abilities.

That review did not find enough evidence to distinguish

among vascular dementia, Alzheimer’s disease, or other types

of cognitive impairment, but another review,11 published, just

2 years later by Diabetic Medicine, went further. “There is also

evidence for an elevated risk of both vascular dementia and

Alzheimer’s disease in Type 2 DM albeit with strong interaction

of other factors such as hypertension, dyslipidemia and

apolipoprotein E phenotype,” the authors wrote. “Both vascular

and nonvascular factors are likely to play a role in dementia

in diabetes.”

Such conclusions soon led other researchers to look at

health record databases in search of further evidence, and

their findings varied significantly. International Psychogeriatrics,

for example, published a 2002 analysis12 of records from 702

people aged 80 years or more. Among the 187 patients to receive

a dementia diagnosis at some point during the study period,

31 began the study with T2D. After adjusting for age, gender,

education, smoking habits, and circulatory diseases, the

investigators calculated that diabetics were more than twice

as likely as other people to develop vascular dementia, but

no more likely than others to develop Alzheimer’s disease. On

the other hand, a 2004 study that appeared in Archives of Neurology13,

followed 824 Catholic nuns, priests, and monks who

received thorough clinical evaluations each year, and found

that preexisting T2D did predict elevated risk of Alzheimer’s

disease. The 127 patients who began the study with T2D had a

65% greater risk of incidence of Alzheimer’s disease than nondiabetic

counterparts (95% CI, 1.10-2.47), even after adjusting

for age, sex, and education.

Several newer analyses have pulled their data from far

larger datasets in hopes of drawing stronger and more accurate

conclusions. A 2014 paper that appeared in PLoS One14

used information on 1 million randomly sampled people to

compare the 71,433 diabetics in the group with 71,311 nondiabetics

matched by age, sex, hypertension, hyperlipidemia

and previous stroke history. Incidence of Alzheimer’s disease

was rare in both groups (0.48% vs 0.37%, P <.001). Still, over a

follow-up period of up to 11 years, T2D patients were far more

likely than controls to develop Alzheimer&rsquo;s disease (hazard ratio

[HR], 1.76; 95% CI, 1.50-2.07; P <.001).

Another large study, published by Diabetes Care in 2015,6

concluded that although dementia risk is most elevated in

patients with longstanding cases of T2D, that risk begins to

go up as soon as patients start demonstrating resistance to

insulin. Investigators from Canada used records from Ontario&rsquo;s

health administration databases to match 225,045 seniors

with newly-diagnosed T2D and 668,070 controls, and followed

them up to 17 years. A fully-adjusted Cox model found the

risk of dementia to be significantly higher among diabetic patients

(HR, 1.16; 95% CI, 1.15-1.18). Risk of Alzheimer&rsquo;s disease

increased within a month of T2D diagnosis, and it kept rising

over time.

&ldquo;When testing the proportional hazards assumption, we

did detect a significant interaction between diabetes status

and time,&rdquo; the study authors wrote. &ldquo;The HR associated with

diabetes status was 1.19 (HR; 95% CI, 1.16-1.22), and this effect

increased by about 1% per year (HR for diabetes &times; year

interaction: 1.009; 95% CI, 1.004-1.014). Thus, after 10 years of

duration, diabetes was associated with a nearly 30% higher

incidence of dementia.&rdquo;

Collectively, the evidence from the individual population

studies has convinced the authors of several meta-studies

that T2D patients do face significantly elevated risk of Alzheimer&rsquo;s

disease. A review that appeared in The Lancet Neurology

in 2006, found that 8 of 13 included studies linked T2D

with Alzheimer&rsquo;s, while 6 of 9 included studies linked T2D

with vascular dementia.1 A 2012 meta-analysis that appeared

in Internal Medicine Journal,2 pooled the relative risks found in

19 studies with nearly 45,000 subjects and found that T2D

patients, compared to other patients, faced a higher risk of

Alzheimer&rsquo;s disease (relative risk [RR], 1.46; 95% CI, 1.20-1.77),