Try the all-new QuickBooks Online for FREE.  No credit card required.

Diabetes and Dementia

Document Sample
Diabetes and Dementia Powered By Docstoc
					                                                                 Current Issues

                                           Diabetes and Dementia

                                                                                                                                              a report by
                                                                                                                             Geert Jan Biessels

                                                                          Neurologist, Department of Neurology, Rudolf Magnus Institute of Neurosciences,
                                                                                                                       University Medical Center Utrecht

                                           New views are emerging on the spectrum of                  We have recently performed a systematic review on
                                           neurological complications that are associated with        this topic.6 The review included 11 studies in which
                                           diabetes. In addition to somatic and autonomic             both diabetes and dementia were assessed in later life
                                           peripheral diabetic neuropathy, stroke and acute           (see Figure 1). The incidence of ‘any dementia’ was
                                           metabolic catastrophes, progressive end-organ              higher in diabetic patients in five of seven studies
                                           damage to the brain is now recognised as a long-term       reporting this aggregate outcome. Overall, the
                                           complication of diabetes. This cerebral damage is          incidence of dementia in diabetic patients was higher
                                           manifested in impaired cognitive performance and           by 50 to 100%, relative to non-diabetic individuals.
  Geert Jan Biessels is a Neurologist      subtle structural cerebral abnormalities. In addition,     This increased risk involved both Alzheimer’s disease
 at the Department of Neurology of         the risk of dementia is increased. This paper provides     and vascular dementia (seven of 11 studies, six of
       the Rudolf Magnus Institute for
        Neuroscience, at the University    a brief overview on the relationship between diabetes      seven studies respectively), with a ~50% to 100%
         Medical Center Utrecht in the     and dementia.                                              increased risk of Alzheimer’s disease and a ~100%
Netherlands. His research focuses on
 the effects of diabetes on cognitive
                                                                                                      increased risk of vascular dementia.
  functioning, and on other vascular       Risk of Dementia in Diabetes
      causes for cognitive decline and                                                                The review also included four studies that assessed
           dementia. He is one of the
  founders and project leader of the       Both type 1 and type 2 diabetes mellitus are               diabetes at mid-life (before the age of 60) and
      Utrecht Diabetic Encephalopathy      associated with cognitive performance impairments.         incident dementia at advanced age, after an extended
Study Group. This study group aims
                                           In type 1 diabetes mellitus this is reflected in a mild    follow-up of 25–35 years. The interpretation of the
 to elucidate the pathophysiology of
       accelerated cognitive decline in    to moderate slowing of mental speed and a                  findings from such studies requires some caution,
    diabetic patients through detailed     diminished mental flexibility.1 In type 2 diabetes         because the prolonged follow-up can lead to
     neuropsychological assessment, in
combination with brain imaging and         cognitive changes mainly affect learning and               substantial effects of attrition, and incident diabetes
         examination of metabolic and      memory, mental flexibility and mental speed.2,3            during follow-up can further confound the results.
   vascular risk factors. He is author                                                                Nevertheless, the incidence of any dementia was
         or co-author of 50 papers on
  peripheral and central neurological      Longitudinal population-based studies indicate that        shown to be increased in diabetic subjects in two out
 complications of diabetes, and is a       the rate of cognitive decline is accelerated in            of three of these studies and the incidence of
  board member of NEURODIAB, the
Diabetic Neuropathy Study Group of
                                           elderly subjects with type 2 diabetes.4 While the          Alzheimer’s disease in one of two studies. Two
     the European Association for the      association between diabetes and these modest              studies observed a modest, non-significant increase in
 Study of Diabetes. His PhD studies,       changes in cognition is now well established; the          the incidence of vascular dementia.
        completed in 1997, involved a
   series of experiments on cognition      relationship between diabetes and dementia has
              and synaptic plasticity in   been subject of debate.                                    The observed association between diabetes and
         experimental diabetes in rats.
                                                                                                      vascular dementia is not surprising, given the fact that
                                           Early studies on the relationship between diabetes         diabetes is a well-known risk factor for
                                           and dementia suggested that: “diabetes and                 cerebrovascular disease. The nature of the association
                                           Alzheimer’s disease may not co-exist”.5 This               between diabetes and Alzheimer’s disease is less clear.
                                           suggestion was based on the observation that the           Although experimental data link disturbances in
                                           prevalence of diabetes among patients with                 insulin and glucose metabolism to amyloid synthesis
                                           established Alzheimer’s disease was low. It is likely,     and breakdown,7 there is, as yet, no solid clinical
                                           however, that these early studies suffered from            neuropathological evidence that the deposition of
                                           methodological limitations such as survival bias.          amyloid plaques, the pathological hallmark of
                                           Moreover, the possible effects of Alzheimer’s disease      Alzheimer’s disease, is accelerated in the brains of
                                           itself on glucose metabolism may have confounded           diabetic patients.
                                           the results.
                                                                                                      An alternative view would be that Alzheimer’s-
                                           Over the past decade several large longitudinal            type pathology and diabetes-related pathology
                                           population-based studies have provided more reliable       occur independently from each other and that
                                           estimates of the risk of incident dementia in diabetes.    diabetes just reduces the threshold at which

10                                                                                                                     EUROPEAN ENDOCRINE DISEASE 2006
                                                                                                      Diabetes and Dementia

Alzheimer-type pathology leads to clinically               Figure 1: The Risk of Dementia in Diabetes
manifest Alzheimer’s disease.
                                                                 Any dementia
Brain Imaging Studies
                                                                  Ott 1999
Neuroimaging studies may offer insight into the
nature and severity of cerebral damage in diabetes,                                                                                                                                n=461
                                                              Brayne 1998
and thus produce clues as to whether the increased
risk of dementia is the result of vascular                      Peila 2002
abnormalities, degenerative changes or both. On                                                                       n=9131
brain magnetic resonance imaging (MRI) ischaemic           MacKnight 2002

cerebrovascular disease is reflected in symptomatic or                                                                           n=1301
                                                                  Xu 2004
asymptomatic infarcts and in so-called white matter
                                                                                                                                 n=~75 000
hyperintensities (WMH).8 Degenerative changes are             Leibson 1997
reflected in focal or generalised atrophy.8 Infarcts,
WMH and atrophy are relatively common findings                Hassing 2002

in the elderly in the general population, but diabetes
may accentuate these lesions.
                                                                              0.5                 1                          2                                5                        10

In a recent systematic review of brain imaging
                                                                 Alzheimer’s disease
studies in diabetes the results of 55 studies that
compared MRI, magnetic resonance spectroscopy,
computed tomography, single photon emission                                                                                                         n=6370
                                                                  Ott 1999
computed tomography or positron emission                                                                                                                                   n=461
                                                              Brayne 1998
tomography of the brain between patients with                                                                                                                       n=828
                                                            Yoshitake 1995
diabetes and controls were summarised.9
Methodology regarding population selection,                     Peila 2002
diabetes assessment, neuroimaging rating methods           MacKnight 2002
and data analyses were heterogeneous.                                                                                                           n=1301
                                                                  Xu 2004
                                                                                                                                                  n=~75 000
                                                              Leibson 1997
Diabetes was associated with a modest degree of                                                                                                                     n=1786
                                                           Luchsinger 2005
cerebral atrophy in eight out of ten studies
reporting this outcome. The atrophic changes in            Arvanitakis 2004
diabetes appeared to be rather generalised, unlike           Katzman 1989
MRI findings in the early stages of Alzheimer’s                                                                                           n=702
                                                              Hassing 2002
disease, where atrophy predominantly affects the
medial temporal lobe.                                                         0.1                         0.5                    1                2                        5           10

Eight out of 19 studies reported an association
                                                                 Vascular dementia
between diabetes and lacunar infarcts. A meta-
analysis of these 19 studies showed odds ratios for
infarcts in diabetic patients relative to controls to be                                                                                                            n=6370
                                                                  Ott 1999
1.3 (95% CI 1.1–1.6) in population based study                                                                                                    n=828
cohorts and 2.2 (1.9–2.5) in cohorts of patients with       Yoshitake 1995

vascular disease. There was little evidence for an                                                                                                         n=3508
                                                                Peila 2002
association between diabetes and WMH, but the
rating methods that were used were often                   MacKnight 2002

inappropriate to detect subtle abnormalities.                                                                                                                     n=1301
                                                                  Xu 2004
Indeed, when more sensitive WMH rating scales
are used, WMH severity appears to be accentuated           Luchsinger 2001
in type 2 diabetes.10                                                                                                                                 n=702
                                                              Hassing 2002

Risk Factors
                                                                              0.5                 1                          2                                5                        10
Uncertainty still exists concerning the main risk
factors for dementia in diabetes. In fact, even the        Adapted from a systematic review by Biessels et al, 2006.6 The plots include point estimates for the relative risk of incident
                                                           dementia in individuals with diabetes, compared with those without, with 95% confidence intervals, adjusted for age and sex.
causality of the association between diabetes and          Individual studies are ranked according to methodological quality. The studies with the highest quality are at the top of each
dementia is still debated. Alternative options are that    plot. See the original review for full references and further details.6

EUROPEAN ENDOCRINE DISEASE 2006                                                                                                                                                        11
                           Current Issues

     both diabetes and dementia are caused by shared risk        exploratory analyses into the modulating effects of
     factors, or that risk factors that are closely associated   hypertension and vascular disease. Overall,
     with diabetes, such as hypertension, account for the        adjustment of the relative risk of dementia in
     increased risk of dementia in diabetic subjects. It         diabetic patients for hypertension and other vascular
     should also be taken into account that risk factors         risk factors appeared to have limited effects,
     for Alzheimer’s disease and vascular dementia are           indicating that the association between diabetes and
     likely to differ.                                           dementia is at least in part independent of vascular
                                                                 disease.6 Unfortunately, the majority of studies did
     As epidemiological studies on the association               not collect data on diabetes-specific factors such as
     between diabetes and dementia mainly included               diabetes duration, haemoglobin A1c and
     individuals of advanced age, the vast majority of the       microvascular complications. Moreover, despite the
     patients involved will have had type 2 diabetes.            fact that the study populations often included
     Potential risk factors for dementia in these patients       thousands of subjects, the actual number of diabetic
     include demographic, socioeconomic or lifestyle             patients that developed dementia was mostly small,
     factors, that may predispose to both diabetes and           limiting statistical power for the identification of risk
     dementia, factors that are linked to the                    factors for dementia in this subgroup.
     pathophysiology of diabetes but that are not specific
     for diabetes (e.g. hypertension, dyslipidaemia),            Conclusion and Future Directions
     diabetes-specific factors (e.g. hyperglycaemia),
     factors related to diabetes treatment and conditions        Diabetes is associated with a 1.5 to 2 fold increased risk
     that may be the consequence of diabetes (e.g.               of dementia. Imaging studies show both modest
     microvascular complications, atherosclerosis, stroke,       generalised atrophy and vascular lesions. Possibly, this
     depression). As these factors are strongly interrelated     ‘mixed pathology’ reduces the threshold of developing
     and many of them not unique to diabetes per se, the         dementia either through additional vascular lesions, or
     identification of individual risk factors may prove to      through Alzheimer’s-type pathology.
     be a daunting task. In this respect, it could be helpful
     to try and cluster risk factors around certain              As diabetes is a common metabolic disorder in older
     pathophysiological        constructs        such      as    individuals, measures aimed at preventing dementia
     hyperglycaemia-mediated damage or atherosclerosis.          in diabetic patients can have considerable impact.
     The validity of such constructs could then be tested        The initiation of such preventive measures, however,
     in pathophysiological studies with a multimodal             is clearly hampered by the current lack of insight in
     approach that includes detailed cognitive testing and       the key risk factors and underlying mechanisms.
     brain imaging, vascular examinations and assessment         Filling in these gaps in our knowledge is the next
     of diabetic complications.                                  challenge for research in this field. It will require
                                                                 longitudinal studies that include detailed assessment
     According to the currently available epidemiological        of cognition in combination with brain imaging and
     studies, the association between diabetes and               detailed assessment of diabetes-related factors and co-
     dementia is unlikely to be due to the potential             morbid conditions. In the meantime, the most
     confounding effects of demographic variables such as        practical approach to prevent accelerated cognitive
     age or sex, educational level or socioeconomic              decline in individual diabetic patients is to optimise
     status.6 In-depth analyses of the involvement of            glycaemic control and manage vascular risk factors
     vascular risk factors or symptomatic or asymptomatic        rigorously. Although evidence that this approach will
     cerebrovascular disease have not yet been                   indeed prevent dementia is still lacking, it will at least
     performed, although several studies did provide             prevent stroke and other diabetic complications. ■


     1. Brands AMA, Biessels GJ, De Haan EHF, et al., Diabetes Care (2005) 28: pp. 726–735.
     2. Awad N, Gagnon M, Messier C, J Clin Exp Neuropsychol (2004) 26: pp. 1,044–1,080.
     3. Allen KV, Frier BM, Strachan MWJ, Eur J Pharmacol (2004) 490: pp. 169–175.
     4. Cukierman T, Gerstein HC, Williamson JD, Diabetologia (2005) 48: pp. 2,460–2,469.
     5. Bucht G, Adolfsson R, Lithner F, Winblad B, Acta Med Scand (1983) 213: pp. 387–392.
     6. Biessels GJ, Staekenborg S, Brunner E, et al., Lancet Neurol (2006) 5: pp. 64–74.
     7. Craft S, Watson GS, Lancet Neurol (2004) 3: pp. 169–178.
     8. Frisoni GB, Scheltens P, Galluzzi S, et al., J Neurol Neurosurg Psychiatry (2003) 74: pp. 1,371–1,381.
     9. van Harten B, de Leeuw FE, Weinstein HC, et al., Diabetes Care (2006) 29: pp. 2,539–2,548.
     10. Manschot SM, Brands AM, van der GJ, et al., Diabetes (2006) 55: pp. 1,106–1,113.

                                                                                 EUROPEAN ENDOCRINE DISEASE 2006