WASHINGTON – A history of severe hypoglycemic episodes was associated with an increased risk for dementia in a longitudinal cohort study involving 16,667 older patients with type 2 diabetes.
“Severe hypoglycemic episodes might be associated with a neurological consequence in a population that is already at greater risk for dementia. …This study also adds to the evidence base that balance of glycemic control is a critical issue, particularly for the elderly,” Rachel A. Whitmer, Ph.D., said at a press briefing timed to coincide with publication of the special diabetes edition of the Journal of the American Medical Association.
The study, which retrospectively analyzed data from the Kaiser Permanente Northern California Diabetes Registry, identified a 2.39% increase in absolute risk of dementia per year of follow-up for patients with a history of hypoglycemia that resulted in hospitalization or an emergency department visit, compared with type 2 diabetic patients without such a history.
“Trying to aim for a very low glycemic target might not be beneficial and might even be harmful. We know that high blood sugar isn't good, but I think the message here is also that very low levels aren't good,” added Dr. Whitmer of Kaiser Permanente's division of research.
Because the data analyses included two methods by which the hypoglycemia events were separated in time from the onset of dementia, the study supports the direction of causality that the hypoglycemia preceded the dementia, rather than the other way around. Moreover, “Our findings were independent of glycemic control as assessed by levels of [hemoglobin A1c], type of diabetes treatment, and diabetes comorbidities,” wrote Dr. Whitmer and her associates in the published report (JAMA 2009;301:1565-72).
The patients were all aged 55 or older on Jan. 1, 2003, with no signs of dementia at that time. A total of 11% (1,822) patients were diagnosed with dementia during a mean follow-up of 3.8 years and a median follow-up of 4.8 years, and a total of 8.8% (1,465) had at least one episode of severe hypoglycemia during 1980-2002. Of those 1,465, 68.5% had one such episode, 1% had two, and 13.5% had three or more.
Age-adjusted incidence rates of dementia by frequency of severe hypoglycemic episodes were significantly higher among those with at least one episode, compared with those with no such episodes (566.8 vs. 327.6 per 10,000 person-years), with an attributable risk of 2.4% per year.
After adjustment for age, body mass index, race/ethnicity, education, sex, and diabetes duration, the hazard ratios for dementia, compared with patients who had no severe hypoglycemic episodes were 1.7 for those with at least one episode, 2.2 for two or more, and 2.6 for three or more episodes. Further adjustment for diabetes-related comorbidity, HbA1c level, diabetes treatment, and years of insulin use modestly attenuated the effect but it remained “statistically significant and clinically relevant” with hazard ratios of 1.3, 1.8, and 1.9, respectively, Dr. Whitmer and her associates said.
Trends were similar when only the incident dementia cases diagnosed between Jan. 1, 2005, and Jan. 15, 2007, were considered after adjustment for all the above-mentioned factors, with hazard ratios of 1.2 for at least one severe hypoglycemia episode, 1.7 for two or more episodes, and 2.1 for three or more episodes, compared with patients who had no such episodes. Another “backward lag model” analysis that examined only the impact of hypoglycemic events occurring from 1980 through 1985 also identified the same trend, with a hazard ratio of 1.3 for one or more episodes.
Other analyses, including adjustment for other variables indicative of diabetes severity, length of health plan membership, time since initial diabetes diagnosis, and medical utilization rate also yielded similar results, although there was some mild attenuation for patients with three or more episodes, the investigators said.
Possible mechanisms by which hypoglycemia might increase the risk of subsequent dementia in older individuals include neuronal death and/or increased platelet aggregation/fibrinogen formation. Cerebrovascular disease is another possibility, even though another analysis adjusting for acute stroke and transient cerebral ischemia in this study population did not fully account for the effect of hypoglycemia. Given evidence from animals, “cerebrovascular damage is likely one of the mechanisms,” they noted.
Moreover, while hypoglycemia was not found to be associated with higher risk of subsequent impairment among the young adults with type 1 diabetes in the Diabetes Control and Complications Trial, “older individuals are thought to have less brain reserve or brain plasticity and therefore may be unable to recover from neurologic insult as well as younger individuals are able to,” Dr. Whitmer and her associates said.