WASHINGTON — Which comes first, diabetes or depression?

The data on this temporal relationship are mixed. Although findings from previous studies suggest that depression precedes diabetes, findings from another investigation, presented at the annual scientific sessions of the American Diabetes Association, propose that there is no time correlation between the two diagnoses.

This chicken-or-egg question is important because depression has a reported prevalence ranging from 11% to 33% in patients with diabetes, which is twice as high as it is in people without diabetes, said Dr. Lawrence S. Phillips, professor of medicine in the division of endocrinology and metabolism at Emory University, Atlanta.

Dr. Phillips and his colleagues conducted a cross-sectional study of 573 people (about half were white and half were African American) who said that they did not have diabetes. Each person received a 75-g oral glucose tolerance test after an 8-hour overnight fast, and screening for depression with a well-validated tool, the Patient Health Questionnaire.

Normal glucose tolerance (NGT) occurred in 65% of the participants, whereas 15% had impaired fasting glucose (IFG), 8% had impaired glucose tolerance (IGT), 8% had both IFG and IGT, and 4% had diabetes. Some participants had received (11%) or were currently receiving (12%) treatment for depression.

The PHQ scores rose in a statistically significant trend from being low among those who never underwent treatment for depression to being higher in people who received depression treatment in the past and highest in individuals who were currently receiving depression treatment.

But there was no relationship between the different categories of glucose tolerance and PHQ score, the prevalence of any depressive syndrome or major depressive disorder, or the severity of depression, Dr. Phillips said. In multivariate analyses, higher body mass index and current receipt of depression treatment significantly increased the risk of having any depressive syndrome, but this risk was not increased within any category of glucose tolerance.

An audience member asked Dr. Phillips how he viewed the results of his study given that a poster presented at last year's ADA meeting found that patients with IGT in the Diabetes Prevention Program had a significantly increased risk for depression, suggesting depression may have preceded IGT. “It's possible that among people who are depressed, there are neuroendocrine changes that lead to diabetes. There's room in human biology for both processes,” he said.

But he argued that the situation in which depression precedes the development of diabetes is unlikely given the lack of an association between depression and unrecognized IGT in his study and the fact that patients in the Diabetes Prevention Program were told that they had IGT and were at risk for diabetes, which could possibly have had a negative psychosocial effect.

Firm conclusions favoring either side of the issue, however, may have to wait for research into the dynamics of neurohormonal changes, which are believed to underlie some of the association between depression and the development of diabetes, said Dr. Sherita Hill Golden of the division of endocrinology and metabolism at Johns Hopkins University, Baltimore.

Melancholic depression increases the activation of the hypothalamic pituitary adrenal (HPA) axis, which leads to a simultaneous activation of the sympathetic nervous system, Dr. Golden said. In the tightly regulated feedback loop of the HPA axis, the hypothalamus produces corticotropin-releasing hormone (CRH) that stimulates the pituitary gland to release adrenocorticotropin hormone (ACTH), which stimulates the adrenal gland to release cortisol. Cortisol levels in turn regulate the production and release of CRH.

There is evidence to suggest that subclinical hypercortisolism, defined as having two of three abnormalities in HPA axis function (increased 24-hour urine free cortisol, failure of the dexamethasone suppression test, and decreased levels of ACTH), may contribute to the development of type 2 diabetes, Dr. Golden said.

A study of 12 patients with adrenal incidentalomas and subclinical hypercortisolism found that such patients had a higher prevalence of insulin resistance, impaired glucose tolerance, and type 2 diabetes, as well as greater central adiposity, than did 29 patients with a nonfunctioning adrenal incidentaloma and no subclinical hypercortisolism (J. Clin. Endocrinol. Metab. 2002;87:998–1003).

Dr. Golden's research has centered on determining which measures of neuroendocrine activity provide the most reliable results and on understanding how that activity correlates with metabolic parameters. Preliminary results of an extensive 3-day series of tests in 15 healthy African American women have indicated that static measures of HPA axis activity, such as salivary cortisol sampling, do not correlate well with more dynamic measurements, such as 24-hour urine-free cortisol levels. Other analyses in the patient group suggested CT scan measurements of adrenal gland volume and the results of dexamethasone suppression testing are correlated strongly with body mass index, high-density lipoprotein cholesterol, and systolic blood pressure.

The results of studies measuring the effect of neuroendocrine changes on metabolic parameters are beginning to suggest that “modification of the neurohormonal response may provide a novel approach to the primary prevention of type 2 diabetes,” Dr. Golden said, adding that such an approach “would be complementary to our already established measures.”

Diabetes, Depression Link Remains Muddled