After adjusted for myriad confounders, people with uncontrolled HbA1c (defined as greater than 10%) had a 23% greater risk of developing dementia than people with an HbA1c of less than 6%, said Aidin Rawshani, MD, of the National Diabetes Register and Institute of Medicine, Gothenburg, Sweden.
In his presentation at the annual meeting of the European Association for the Study of Diabetes, Rawshani reported that his results came from the study of the heath records of 353,214 individuals with a mean follow-up of 4.8 years, translating to 1.7 million person-years.
Of the total cohort, 13,159 were diagnosed with some form of dementia – the most frequent being an unknown diagnosis. About 3,499 patients were diagnosed with Alzheimer's disease; 3,377 had vascular dementia.
"Higher HbA1c levels are associated with an increased risk of dementia among persons with type 2 diabetes," Rawshani said during his talk.
He said that one finding of his study was that a diagnosis of dementia carried with it a poor prognosis. About 60% of the patients who were diagnosed with dementia had died within 10 years, he said. That compared with a mortality of 30% of patients who were not diagnosed with dementia.
The risk of dementia when compared with various HbA1c levels followed a J-shaped curve. With HbA1c of less than 6% as the reference, Rawshani reported comparative risks for higher HbA1c categories:
- 6%-7%: 14% reduced dementia risk (P<.001)
- 7%-8%: 21% reduced risk (P<.001)
- 8%-9%: 10% reduced risk (P=0.002)
- 9%-10%: 3% reduced risk (P=0.558)
- Greater than 10%: 23% increased risk [hazard ratio 1.23, 95% CI 1.11-1.35, P<.001)
The researchers identified all patients with type 2 diabetes and no known hospitalization for dementia who were registered in the Swedish National Diabetes Registry from 2003 through 2012. These patients were followed up until hospital admission for dementia, death, or end of follow-up on Dec. 31, 2012. Computer modeling was used to calculate the link between HbA1c, and dementia, which also adjusted for variables including age, sex, duration of diabetes, marital status, income, education, smoking status, systolic blood pressure, body mass index, estimated kidney function, statins, levels of urine in protein, type of diabetes medication, atrial fibrillation, stroke, and blood pressure medications.
Previous stroke in these patients made them 40% more likely to develop dementia than those without stroke, Rawshani said.
"The positive association between HbA1c and risk of dementia in fairly young patients with type 2 diabetes indicates a potential for prevention of dementia with improved blood sugar control," he said.
In commenting on the study, session co-moderator Nita Forouhi, MD, head of the unit of nutritional epidemiology at the Medical Research Council at University of Cambridge in England, told MedPage Today, "I think this is good preliminary data. It gives us a good hypothesis. He certainly presented results on an association on dementia risk – but at this point all it is is an association."
She said the research would give impetus to understanding "the particular balance between episodes of hypoglycemia and increased levels of glucose – hyperglycemia.
"I think it is important that we study not only the link to dementia but to cognitive decline because I think that can be informative, too," she said.
Forouhi said that if further studies show the link is real, then clinicians might more to earlier testing of patients with diabetes for the signs of cognitive decline and potentially treat these patients earlier.
From the podium, Forouhi had asked whether the researchers had assessed alcohol consumption in their extensive list of confounders, but the research did not include that information. "I think alcohol intake is an important confounder because we know there is an association between alcohol intake and dementia risk."
In the study, Rawshani said that older patients – those about 67 and 68 years old – appeared to have better glycemic control. Those with the highest HbA1c were about 64 years old. About 54%-55% of the patients with good glycemic control were men; about 60% of the patients with poor control were men. Patients at the ends of the HbA1c spectrum had been diagnosed the shorted period of time – 4.22 years for the lowest mean HbA1c; 5.5 years for those with the highest mean levels. Patients with levels between 7% and 10% had been diagnosed for 8-9 years.
Rawshani and Forouhi disclosed no relevant relationship to industry.
Reviewed by F. Perry Wilson, MD, MSCE Assistant Professor, Section of Nephrology, Yale School of Medicine and Dorothy Caputo, MA, BSN, RN, Nurse Planner
Primary Source: European Association for the Study of Diabetes, 51st annual meeting
This article was first published on MedPage Today and reprinted with permission. Free registration is required.