DENVER – It would seem logical to loosen the reins and manage diabetes less intensively in patients with comorbid dementia. That’s what American Geriatrics Society guidelines recommend. After all, the cognitive impairment in such patients renders them less able to manage a complicated medical regimen, and less communicative in describing complications when they arise.

But is less-stringent metabolic control in diabetic patients with comorbid dementia actually what happens in clinical practice? Mostly not, according to Dr. Paul Tatum of the University of Missouri, Columbia.

He pointed to a massive cross-sectional study involving 497,900 veterans aged 65 and older with type 2 diabetes. The combined prevalence of dementia and cognitive impairment was 13.1% in those aged 65-74, climbing to 24.2% among those aged 75 and older.

The veterans with dementia/cognitive impairment had more intensive diabetes management than did patients with neither condition. Their mean HbA_1c value was lower. More of them were on insulin (30%, compared with 24% in elderly patients without dementia or cognitive impairment). And among all subjects on insulin, 26.5% of those with dementia and 19.5% with cognitive impairment experienced hypoglycemic episodes during the 2-year study period, significantly higher rates than the 14.4% among insulin-using diabetic patients without either condition.

The unadjusted odds ratio for hypoglycemia was 2.4 in veterans with dementia and 1.7 for those with cognitive impairment. After adjustment for other comorbid conditions, demographics, nursing home stays, and other potential confounders, the adjusted odds ratios for hypoglycemia were attenuated but still significant, at 1.58 for dementia and 1.13 for cognitive impairment (J. Am. Geriatr. Soc. 2011;59:2263-72).

"We can help change that paradigm," Dr. Tatum urged at the annual meeting of the American Academy of Hospice and Palliative Care Medicine.

The American Geriatrics Society guideline on diabetes can be an ally in this regard. It states that "for frail older adults, persons with life expectancy of less than 5 years, and other in whom the risks of intensive glycemic control appear to outweigh the benefits, a less stringent target such as 8% is appropriate" (J. Am. Geriatr. Soc. 2003;51[5 suppl. guidelines]:S265-80).

The average life expectancy from the initial diagnosis of Alzheimer’s dementia is 4.5 years (Ann. Intern. Med. 2004;140:501-9).

The American Geriatrics Society guideline also emphasizes the importance of screening for cognitive impairment in older patients with diabetes. This point was driven home in a large, longitudinal cohort study conducted at Kaiser Permanente in Northern California.

The investigators followed 16,667 patients with type 2 diabetes, no cognitive impairment at baseline, and a mean age of 65 years. Subjects who experienced one or more acute hypoglycemic episodes severe enough to require hospitalization during 1980-2002 turned out to have an increased risk of being diagnosed with dementia in 2003-2007.

The risk was graded. Patients with a single episode of severe hypoglycemia had a fully adjusted 1.26-fold greater likelihood of later being diagnosed with dementia, compared with patients who had no episodes. Those with two episodes had a 1.8-fold increased risk, and three or more episodes placed an individual at a 1.94-fold increased risk. A similar relationship held true for emergency department visits for hypoglycemia and subsequent dementia risk. These findings were independent of HbA_1c level, diabetes comorbidities, and diabetes medications (JAMA 2009;301:1565-72).

It’s unclear whether hypoglycemia increases the risk of dementia in older type 2 diabetes patients because severe hypoglycemia can cause lasting neurologic insult, or whether severe hypoglycemia is an early marker for cognitive decline.

For busy clinicians who say they simply can’t fit in cognitive assessments in older patients, diabetic or otherwise, Dr. Tatum recommended the Mini-Cog as a 3-minute test that approximates the Mini-Mental State Exam, which takes at least twice as long. The Mini-Cog involves a three-item recall test for memory, along with a simply scored clock-drawing test.

"It’s at least a start for that busy physician who says, ‘I just don’t have time,’ " according to Dr. Tatum.

The Mini-Mental State Exam is not a great instrument for assessing executive function, which often deteriorates well before memory. He recommended the Montreal Cognitive Assessment as a useful tool for a more detailed evaluation of executive function as part of cognitive testing.

Dr. Tatum said that the American Geriatrics Society recommendations regarding diabetes and dementia are consistent with the U.K. National Institute for Health and Clinical Excellence (NICE) guidelines on dementia, which he finds quite helpful. Among the best practices highlighted in the NICE guidelines: "Adopt a palliative care approach from diagnosis until death to support the quality of life of people with dementia and to enable them to die with dignity and in the place of their choosing."

Dr. Tatum reported having no relevant financial conflicts.

Diabetes Managed More Tightly in Demented Patients