Diabetes mellitus in patients with dementia : clinical care and pharmacological treatment

Abstract

Diabetes mellitus (DM) and dementia are frequent chronic disorders in the older population, however their relationship is complex - while DM is an established risk factor for dementia, cognitive symptoms in dementia may hinder the self-management essential in DM care. Importantly, the co-occurrence of DM and dementia is common in clinical practice, however the research examining patients suffering from both disorders is scarce. This thesis analyzes the bidirectional associations between DM and dementia in patients with both disorders in Sweden, with specific focus on pharmacological care.

The thesis is based on the merged data from the Swedish Dementia Registry and the Swedish Prescribed Drug Register, Swedish National Patient Register, Swedish Cause of Death Register, Total Population Register and the Longitudinal Integrated Database for Health Insurance and Labour Market Studies. All included studies were observational, study 1 was cross-sectional and in studies 2-5 longitudinal open-cohort design was used.

Study 1 compares the characteristics of patients with DM and dementia to patients without DM. We show that DM is prevalent in 16.5% of patients with dementia, and that DM is associated with diagnosis at younger age, vascular dementia and mixed-pathology dementia (MixDem), and less frequent use of cholinesterase inhibitors (ChEI) and memantine. In study 2 we analyze the association between ChEI and mortality in patients with DM and Alzheimer’s disease (AD) or MixDem. We show that the initiation of ChEI class, donepezil and galantamine is associated with lower all-cause mortality, and the direction and strength of the association is comparable to DM-free patients. Study 3 explores the changes in long-term utilization of antidiabetic medication in patients with type 2 DM or other/unspecified DM with and without dementia. We conclude that utilization as well as new dispensation of insulin is significantly higher among patients with dementia, while the newer antidiabetic drugs are less commonly prescribed. Study 4 compares the mortality risk associated with six major antidiabetic drugs in patients with type 2 DM or other/unspecified DM and with and without dementia diagnosis. Overall, the initiation of insulin in patients with type 2 DM or other/unspecified DM is associated with higher mortality, regardless of dementia status. Additionally, we observe lower mortality in patients with dementia who used sodium-glucose cotransporter-2 inhibitors (SGLT-2i). Lastly, study 5 examines whether the use of antidiabetic medications is associated with longitudinal changes in Mini-Mental State Examination (MMSE) scores in patients with AD or MixDem. Importantly, we conclude slower decline in MMSE scores among users of metformin and dipeptidyl-peptidase-4 inhibitors.

In conclusion, the patients with DM and dementia constitute a unique cohort less likely to receive treatment with ChEI despite the observed lower mortality associated with ChEI in our and previous studies. Moreover, we describe higher utilization of insulin and lower use of newer antidiabetic medications in patients with dementia, reflecting the Swedish clinical approach. Furthermore, we suggest that antidiabetic medications may provide cognitive benefit in patients with AD or MixDem. Additional studies focusing on optimization of antidiabetic and dementia medication, glycemic control as well as cognitive changes are needed to disentangle the role of DM in patients with manifest dementia.