Back grey_arrow_rt.gif
 
 
Diabetes Linked to Alzheimer's Risk
 
 
  Download the PDF here

from Jules: if these findings hold true HIV+ individuals may be at greater risk for neurological problems mentioned here including dementia, alzheimers & vascular dementia. On another point when HIV researchers conduct studies to find rates of dementia in HIV_ individuals I don't think they control for these factors so in the future studies researchers ought to consider controlling for this factor.

Diabetes Linked to Alzheimer's Risk - "Our findings emphasize the need to consider diabetes as a potential risk factor for all-cause dementia, Alzheimer's disease, and probably vascular dementia,"........."Our findings suggest that diabetes is a significant risk factor for all-cause dementia, AD, and probably VaD. Moreover, 2-hour PG levels, but not FPG levels, are closely associated with increased risk of all-cause dementia, AD, and VaD. (vascular dementia)"................"Diabetes and prediabetes together appeared to account for 14.6% of all-cause dementia, 20.1% of Alzheimer's disease, and 17.0% of vascular dementia risk in the population studied." in Japanese community-dwelling dementia-free subjects aged ≥60 years

By Crystal Phend, Senior Staff Writer, MedPage Today
Published: September 20, 2011

Perhaps more interesting, though, was the strong risk prediction of postload glucose levels during the oral glucose tolerance test, mimicking how meals are metabolized, commented Richard Bergenstal, MD, of the International Diabetes Center at Park Nicollet in Minneapolis.

Action Points

· Explain that in this study, diabetes was associated with elevated all-cause dementia risk.

· Note that diabetes also was specifically associated with elevated Alzheimer's disease risk.

Higher two-hour postload glucose levels correlated with greater risk of developing dementia (P<0.001 for trend for all-cause dementia and Alzheimer's disease, P=0.02 for trend for vascular dementia).

After adjustment for age, sex, hypertension, electrocardiogram abnormalities, body mass index, waist-to-hip ratio, total cholesterol, prior stroke, education, smoking, alcohol intake, and physical activity in the multivariate analysis:

· Two-hour postload glucose levels of 7.8 (140 mg/l) to 11.0 mmol/L (198 mg/l) predicted 50% elevated risk of all-cause dementia and 87% elevated likelihood of Alzheimer's disease (both P=0.02).

· Two-hour postload glucose levels above 11.0 mmol/L (198 mg/l) predicted 2.47-fold higher risk of all-cause dementia and 3.42-fold elevated Alzheimer's risk (both P<0.001) and 2.66-fold elevated vascular dementia risk (P=0.01).

·

mmol/l to mg/dl conversion en.mte.cz/conversion.php

Conversion of glucose plasmatic level. mmol/l to mg/dl and mg/dl to mmol/l.


limite.gif

Those findings suggested "that postprandial glucose regulation is critical to prevent future dementia," Kiyohara's group argued.

Bergenstal, a past president of the American Diabetes Association, cautioned that it's not yet clear whether better control of mealtime insulin levels could impact risk for patients who have already developed diabetes.

Awareness of the link may be the bigger message, he suggested in an interview with Medpage Today.

"We need to understand it a lot better before we build this into our clinical practice," he said. "We don't know yet from these studies how to intervene."

Some prior epidemiologic studies have also implicated diabetes in dementia risk, though not consistently so, they noted.

"Our findings emphasize the need to consider diabetes as a potential risk factor for all-cause dementia, Alzheimer's disease, and probably vascular dementia," Kiyohara and colleagues agreed in the paper.

In the study, diabetes detected on an oral glucose tolerance test was associated with 2.07-fold higher age- and sex-adjusted risk of vascular dementia during follow-up compared with those with normal glucose tolerance (P=0.04). But the vascular dementia link lost statistical significance with further multivariable adjustment, though it still trended (HR 1.82, P=0.09).

This might have been due to small numbers of vascular dementia cases, or because diabetes impacts vascular dementia by mediating hypertension and other cardiovascular risk factors that were controlled for in the multivariate analysis, the researchers explained.

Hyperglycemia itself may have an impact on the brain through atherosclerosis, oxidative stress and accumulation of advanced protein glycation, and changes in insulin metabolism yielding distorted amyloid metabolism, they noted
.

Their prospective Hisayama Study included 1,017 community-dwelling older adults without baseline dementia in Japan who had an oral glucose tolerance test at age 60 or older and were followed for dementia status for 15 years.

Unlike oral glucose tolerance test results, fasting plasma glucose levels at baseline didn't correlate with dementia overall or by dementia subtype.

Impaired glucose tolerance did show some trends and borderline significance in predicting dementia of any type and Alzheimer's.

Adding those prediabetes cases together with overt diabetes also significantly predicted both in the multivariate-adjusted model.

Diabetes and prediabetes together appeared to account for 14.6% of all-cause dementia, 20.1% of Alzheimer's disease, and 17.0% of vascular dementia risk in the population studied.

The researchers cautioned that generalizability to other ethnic populations needs confirmation in further studies.

Other limitations included a single baseline measurement of diabetes status and inability to account for changes participants made in their risk factors during follow-up.

----------------------------------------

Glucose tolerance status and risk of dementia in the community

The Hisayama Study


Abstract

Objective: We investigated the association between glucose tolerance status defined by a 75-g oral glucose tolerance test (OGTT) and the development of dementia.

Methods: A total of 1,017 community-dwelling dementia-free subjects aged ≥60 years who underwent the OGTT were followed up for 15 years. Outcome measure was clinically diagnosed dementia.

Results: The age- and sex-adjusted incidence of all-cause dementia, Alzheimer disease (AD), and vascular dementia (VaD) were significantly higher in subjects with diabetes than in those with normal glucose tolerance. These associations remained robust even after adjustment for confounding factors for all-cause dementia and AD, but not for VaD (all-cause dementia: adjusted hazard ratio [HR] = 1.74, 95% confidence interval [CI] = 1.19 to 2.53, p = 0.004; AD: adjusted HR = 2.05, 95% CI = 1.18 to 3.57, p = 0.01; VaD: adjusted HR = 1.82, 95% CI = 0.89 to 3.71, p = 0.09). Moreover, the risks of developing all-cause dementia, AD, and VaD significantly increased with elevated 2-hour postload glucose (PG) levels even after adjustment for covariates, but no such associations were observed for fasting plasma glucose (FPG) levels: compared with those with 2-hour PG levels of <6.7 mmol/L, the multivariable-adjusted HRs of all-cause dementia and AD significantly increased in subjects with 2-hour PG levels of 7.8 (140 mg/l) to 11.0 mmol/L (198 mg/l) or over, and the risk of VaD was significantly higher in subjects with levels of ≥11.1 mmol/L.

Conclusions: Our findings suggest that diabetes is a significant risk factor for all-cause dementia, AD, and probably VaD. Moreover, 2-hour PG levels, but not FPG levels, are closely associated with increased risk of all-cause dementia, AD, and VaD.

GLOSSARY


AD=Alzheimer disease;
CI=confidence interval;
DSM-III-R=Diagnostic and Statistical Manual of Mental Disorders, 3rd edition, revised;
FPG=fasting plasma glucose;
HR=hazard ratio;
IFG=impaired fasting glycemia;
IGT=impaired glucose tolerance;
NGT=normal glucose tolerance;

OGTT=oral glucose tolerance test;
PG=postload glucose;
VaD=vascular dementia.

Diabetes mellitus is one of the most common metabolic disorders, and its prevalence has risen globally in recent years. Some epidemiologic studies have reported that diabetes is independently implicated in the development of dementia.1,-,3 However, these findings are inconsistent for its subtypes; one study found an association between diabetes and the risk of both Alzheimer disease (AD) and vascular dementia (VaD),1 whereas other studies found an association with only AD2,3 or only VaD,4,-,7 and still others showed no association between diabetes and either condition.8,9 These conflicting results may have been related to differences in the study designs, including the defined criteria for diabetes and dementia subtypes, as well as in the regional characteristics and ethnicities of the settings and subjects. Thus, accurate definitions of diabetes and dementia subtypes are needed to ascertain the true associations between the two, and a 75-g oral glucose tolerance test (OGTT) and morphologic examination of the brain may meet this requirement. However, to date, very few cohort studies have had enough quality data to allow reliable diagnosis using these methods.

To resolve these issues, we performed a prospective cohort study of dementia in a Japanese community-dwelling population, all members of which underwent the OGTT. The most important feature of this study is that the subtypes of dementia were verified by detailed neurologic and morphologic examination, including neuroimaging and autopsy. Using data from this cohort study, we investigated the association between glucose tolerance levels defined by the OGTT and the development of dementia and its subtypes.

DISCUSSION

In a long-term prospective study of an elderly Japanese population, we demonstrated that diabetes that was assessed 15 years earlier was a significant risk factor for the development of all-cause dementia, AD, and VaD. Moreover, the risks of developing all-cause dementia and its subtypes progressively increased with elevating 2-hour PG levels.

In prior prospective epidemiologic studies, there have been conflicting results regarding the associations between diabetes and incidences of all-cause dementia and AD, while the influence of diabetes on the risk of VaD has been positive in most studies.1,4,-,7 Cohort studies in which diabetes was defined by nonfasting blood glucose levels or clinical information did not reveal clear associations of diabetes with the development of all-cause dementia and AD,4,-,8 while the risks of dementia and its subtypes significantly increased in diabetes in some studies, most of which defined diabetes using the OGTT.1,-,3 The latter findings were in accord with ours. This fact suggests that differences in the methods used to define diabetes lead to a discrepancy in the association between diabetes and the risk of dementia, especially AD, and that an OGTT is essential for the definition of diabetes in epidemiologic studies on the diabetes-dementia association.

In our study, the incidence of VaD was significantly higher in subjects with IGT or diabetes than in those with NGT, but this association disappeared after adjustment for other covariates. This might occur due to the few VaD cases. In addition, since other known cardiovascular risk factors, such as hypertension, obesity, and dyslipidemia, accumulate under a prediabetic or diabetic state, as shown in our data (table 1), IGT and diabetes seem to increase the risk of VaD through mediation of these risk factors, especially hypertension.

In the present study, increased 2-hour PG levels including a prediabetic range were significantly linked to elevated risks of all-cause dementia, AD, and VaD, but no such associations were observed for FPG. The epidemiologic evidence from Asia has also indicated that 2-hour PG levels are better in detecting prediabetes and diabetes compared with FPG levels.19 However, very few prospective studies have investigated the associations between FPG as well as 2-hour PG levels and the risks of dementia and its subtypes. Only the Uppsala Longitudinal Study of Adult Men evaluated the associations of FPG levels with the risks of developing AD and VaD,20,21 and this study concluded that increased FPG levels were not risk factors for these subtypes of dementia. This is in good agreement with our findings. The Uppsala Study21 and the Honolulu-Asia Aging Study1 also found no clear associations between 2-hour PG levels and the risks of AD and VaD. These findings are inconsistent with ours. Our recent clinicopathologic study of deceased Hisayama residents revealed that higher levels of 2-hour PG but not of FPG were clearly associated with increased risk for formation of neuritic plaques even after adjustment for confounding factors.22 This evidence together with the findings of the present study suggests that elevated 2-hour PG levels play an important role in the formation of neuritic plaques, and thereby in the development of AD. Meanwhile, it is well known that increased 2-hour PG levels are closely associated with the development of stroke, which is well established as a main cause of VaD. Thus, it is reasonable to postulate a close association between 2-hour PG levels and the risk of VaD.

Possible pathophysiologic mechanisms through which diabetes or elevated blood glucose levels might affect the initiation and promotion of dementia have been extensively discussed in a number of studies.23 A recent review summarized 4 major pathways for hyperglycemia-induced dementia: namely, atherosclerosis, microvascular disease, glucose toxicity leading to the accumulation of advanced protein glycation and increased oxidative stress, and changes in insulin metabolism resulting in an insulin-resistant state and distorted amyloid metabolism in the brain.23 The former 2 pathways are considered to be involved in the development of VaD, while the latter 2 pathways may mainly contribute to the development of AD. Additionally, recent evidence has emerged to imply that vascular factors may be involved in AD.23 It is reported that 2-hour PG values can be a good marker of oxidative stress levels arising from hyperglycemia24,25 and correlate with insulin resistance.26 Higher oxidative stress and insulin resistance may precede the accumulation of amyloid-ß peptide and neurofibrillary tangles23,27 and accelerate arteriosclerosis in the brain,28 resulting in increased risk of AD and VaD. It is known that Asians have lower levels of insulin secretion compared with other ethnic groups29 and can develop diabetes, insulin resistance, and metabolic syndrome with lower body mass index levels.30 These findings suggest that hyperglycemia plays a larger role in the development of dementia compared with insulin resistance in Asians including Japanese. Further studies are needed to elucidate the pathogenesis of hyperglycemia and diabetes in the development of dementia.

The strengths of our study include its longitudinal population-based study design, use of OGTT for determination of glucose tolerance levels in all subjects, long duration of follow-up, perfect follow-up of subjects, and morphologic examination of the brains of most dementia cases with autopsy and neuroimaging. Several limitations of our study should be noted. First, the diagnosis of glucose tolerance status was based on a single measurement of glucose levels at baseline, as was the case in most other epidemiologic studies. During the follow-up, risk factor levels were changed due to modifications in lifestyle or medication especially in subjects with diabetes, and misclassification of glucose tolerance categories was possible. This could have weakened the association found in this study, biasing the results toward the null hypothesis. Therefore, the true association may be stronger than that shown here. Second, some subjects (n = 33 to 65) did not participate in the follow-up surveys of cognitive function performed in 1992, 1998, and 2005, and their cognitive conditions were evaluated only by mail or telephone. This might have resulted in failure to detect dementia cases. However, we also collected information on the development of dementia in another way, namely through the daily monitoring system established in the town. Thus, we believe that we detected almost all dementia cases, and this bias did not affect our findings. Third, the diagnosis of dementia was verified by autopsy only in 50.9% of dementia cases, resulting in a certain degree of subtype misclassification; agreement rate between clinical diagnosis and neuropathologic diagnosis was not high (64.4%) in our autopsy cases of dementia. However, a sensitivity analysis using only definite cases of dementia determined by brain autopsy did not make any material difference in our findings.

Our findings emphasize the need to consider diabetes as a potential risk factor for all-cause dementia, AD, and probably VaD. The other main finding, that elevated 2-hour PG levels are closely associated with increased risks of all-cause dementia and its subtypes, supports the view that postprandial glucose regulation is critical to prevent future dementia. Further investigations are required to clarify the associations between 2-hour PG levels by the OGTT and subtypes of dementia in other ethnic populations.

RESULTS

Table 1 shows the age- and sex-adjusted mean values or frequencies of risk factors for dementia by the WHO criteria at baseline. Compared with those with NGT, the mean values of systolic and diastolic blood pressures, body mass index, waist to hip ratio, and total cholesterol, and the frequencies of hypertension and alcohol intake, were higher in subjects with IFG, IGT; or diabetes.

The age- and sex-adjusted incidences and adjusted HRs of all-cause dementia and its subtypes according to glucose tolerance status defined by the WHO criteria are shown in table 2. Compared with those with NGT, the age- and sex-adjusted incidence and HR of all-cause dementia were significantly higher in subjects with IGT as well as those with diabetes. This association remained unchanged in subjects with diabetes even after adjustment for age, sex, hypertension, EKG abnormalities, body mass index, waist to hip ratio, total cholesterol, history of stroke at entry, education, smoking habits, alcohol intake, and physical activity. In regard to subtypes of dementia, the age- and sex-adjusted incidence and adjusted HRs of AD were significantly higher in subjects with diabetes than in those with NGT. The age- and sex-adjusted incidence and HR of VaD were significantly increased in subjects with IGT or diabetes compared with those with NGT; however, these associations were not significant after multivariable adjustment. No significant associations were observed between glucose tolerance levels and the risk of other dementia. When IGT and diabetes were brought together in one category, this category also had the significantly higher risks of all-cause dementia, AD, and VaD in the age- and sex-adjusted analysis, and these associations remained significant for all-cause dementia and AD even after adjustment for other possible risk factors. The population attributable fraction of this combined category was 14.6% for all-cause dementia, 20.1% for AD, and 17.0% for VaD.

Table 3 presents the associations between FPG levels and adjusted risks of all-cause dementia and its subtypes. The age- and sex-adjusted incidences and HRs of all-cause dementia and any of the dementia subtypes did not differ among FPG levels. This tendency was unchanged even in the multivariate analysis. Conversely, as shown in table 4, the age- and sex-adjusted incidences and HRs of all-cause dementia, AD, and VaD significantly increased with rising 2-hour PG levels. Compared with those with 2-hour PG levels of <6.7 mmol/L, the age- and sex-adjusted incidences and HRs of all-cause dementia, AD, and VaD were marginally or significantly higher in subjects with 2-hour PG levels of 7.8 to 11.0 mmol/L and significantly higher in subjects with 2-hour PG levels of ≥11.1 mmol/L. These associations remained robust even after multivariable adjustment; the risks of all-cause dementia and AD were significantly increased in subjects with 2-hour PG levels of 7.8 to 11.0 mmol/L and over, and the risk of VaD was significantly higher in those with 2-hour PG levels of ≥11.1 mmol/L.

Sensitivity analysis in which only definite cases of dementia determined by brain autopsy were used as event cases did not make any material difference in these findings, except with respect to VaD, for which the significant association disappeared, probably due to the few event cases (table 5). When only clinical diagnoses were used for cases with both clinical and neuropathologic diagnoses, the findings were substantially unchanged, though the HRs became slightly lower probably due to the decreased accuracy of diagnosis (tables e-1, e-2, and e-3).

 
 
 
  icon paper stack View Older Articles   Back to Top   www.natap.org