Plain English - Type 2 diabetes linked to higher rate of Parkinson’s

Some of you may have already seen in the newspapers details of our recent study on the link between Type 2 Diabetes and Parkinson's disease... 

In our study, we used the Hospital Episode Statistics database and looked at a very large number of people (2,017,115) with a type 2 diabetes diagnosis during a hospital admission from 1999-2011, excluding people who already had a Parkinson’s diagnosis. We compared them to a group of over 6 million people from the same database without type 2 diabetes. Among the people with diabetes, 14,252 had a diagnosis of Parkinson’s during a later hospital admission, compared with 20,878 in the comparison group. After excluding people with some similar conditions, and controlling for age, sex and region of residence, we found that people with type 2 diabetes were 32 per cent more likely to have a subsequent diagnosis of Parkinson’s. The link was strongest among younger people (aged 25-44), and people with more complications from their diabetes.

Over the years, this link has proved less of a challenge to study and more of a challenge to interpret. For example, whilst most studies done on this subject suggest some sort of link between diabetes and risk of Parkinson's, the curious fact is that some suggest an increased risk of Parkinson's in those with diabetes, and some studies suggest a decreased risk. The difference may be accounted for by the design of the individual studies and requires further exploration. Our study adds to the literature and increases confidence that the association is such that type 2 diabetes increases risk, but whether this is due to common underlying mechanisms of disease or a true causal link between the two remains to be seen.

What is true, is that diabetes drugs may help treat Parkinson's, and we have already seen a great example of this in the last year. Exenetide appears to have a benefit on the symptoms of Parkinson's and furthermore may alter the underlying disease process. But before we can say that definitively, we need further studies.

- Alastair Noyce

Neurology
Eduardo De Pablo-Fernandez, Raph Goldacre, Julia Pakpoor, Alastair J. Noyce, Thomas T. Warner First published June 13, 2018, DOI: https://doi.org/10.1212/WNL.0000000000005771

http://n.neurology.org/content/early/2018/06/13/WNL.0000000000005771

Objective
To investigate the association between type 2 diabetes mellitus (T2DM) and subsequent Parkinson disease (PD). 

Methods
Linked English national Hospital Episode Statistics and mortality data (1999–2011) were used to conduct a retrospective cohort study. A cohort of individuals admitted for hospital care with a coded diagnosis of T2DM was constructed, and compared to a reference cohort. Subsequent PD risk was estimated using Cox regression models. Individuals with a coded diagnosis of cerebrovascular disease, vascular parkinsonism, drug-induced parkinsonism, and normal pressure hydrocephalus were excluded from the analysis.

Results
A total of 2,017,115 individuals entered the T2DM cohort and 6,173,208 entered the reference cohort. There were significantly elevated rates of PD following T2DM (hazard ratio [HR] 1.32, 95% confidence interval [CI] 1.29–1.35; p < 0.001). The relative increase was greater in those with complicated T2DM (HR 1.49, 95% CI 1.42–1.56) and when comparing younger individuals (HR 3.81, 95% CI 2.84–5.11 in age group 25–44 years).

Conclusions
We report an increased rate of subsequent PD following T2DM in this large cohort study. These findings may reflect shared genetic predisposition and/or disrupted shared pathogenic pathways with potential clinical and therapeutic implications.