Treatment of Mood Disorders: Diabetes Mellitus
The high prevalence of psychiatric illness in the diabetic population has been significantly evaluated only in recent years. In a study of 114 adult diabetic patients, 71% had a lifetime prevalence of a psychiatric disorder. The most common psychiatric diagnosis in these patients was depression along with anxiety disorder. Several studies have shown that the prevalence of depression in the adult population with diabetes mellitus ranges from 8.5% to 27.3%. Also, with comorbid depression comes evidence of poorer compliance, poorer glycemic control, and higher glycosylated hemoglobin values. Additionally, depression has been associated with an increased rate of complications that most probably were significantly affected by this poor control.

Previous studies have reviewed the different pharmacological agents available and recommended the use of SSRIs as the first-line choice of drug. The basis for this recommendation includes relative low risk for side effects and lack of drug interactions. SSRIs have been associated in some cases with a 30% decrease in fasting blood sugar as well as some anorectic effect leading to an average 2-pound decrease in weight. Several studies have examined the effectiveness of fluoxetine in use with diabetic patients with and without depression. _{Maheux et al. (1997}) demonstrated improved glucose control in nondepressed obese type 3 diabetic patients in a randomized, placebo-controlled trial. Insulin-mediated glucose disposal was assessed by a two-step euglycemic-hyperinsulinemic clamp technique. Compared with placebo, fluoxetine increased glucose disposal by 2.4-fold. Also of note was that the patients’ weight remained stable throughout the study. Goodnick demonstrated the effectiveness of sertraline in particular in type 2 diabetic patients in 1997 with overall reductions in Hamilton Rating Scale for Depression and Beck Depression Inventory scores in a 10-week placebo-controlled study. He and his coauthors were also able to show improved dietary compliance and reduction in glycosylated hemoglobin. Similar findings have been reported with the use of fluoxetine. _{Connolly et al. (1995}) showed, in a double-blind, placebo-controlled study, significant reductions in weight and in glycosylated hemoglobin levels in fluoxetine-treated subjects compared with control subjects. The advantages of sertraline may be its relatively low muscarinic blockade and the expected lower rates of the side effects of constipation, memory impairment, and dry mouth.

TCAs have been used extensively in patients with depression and diabetes despite evidence that TCAs may worsen some aspects of the patients’ metabolic control. Short-term use of TCAs can initially lower serum blood sugar in diabetic patients, but long-term use has been shown to induce hyperglycemia as well as increase glycosylated hemoglobin. TCAs are well known for their associated weight gain as well as the increase in carbohydrate craving. In addition, one should consider the increased prevalence of CAD in diabetic patients and the possibility of cardiac conduction influencing the TCAs. A wealth of research has shown significant cognitive worsening in diabetic patients with the use of TCAs and may be attributable to anticholinergic side effects.

Research on the use of alternative agents in diabetic patients is rather limited. MAOIs have been shown in small studies to exaggerate hypoglycemic responses believed to be a result of a drug interaction between insulin or oral sulfonylureas. Other potential hazards with the use of MAOIs in patients with diabetes include weight gain, which, for patients with type 2 diabetes, may be problematic. One literature source has also cited a nefazodone-induced hypoglycemia in a diabetic patient with major depression. Of the newer antidepressants, bupropion could become the preferred agent, especially with recent clinical data showing significant weight reduction in patients with depression.

Because of the potential for drug interactions and risk of worsening other medical conditions, behavioral therapy may be indicated for use in the diabetic patient. Evidence from a randomized clinical trial indicated improvement in symptomatic depression and increase in glycemic control using CBT. Lustman et al. found the CBT intervention over the 6-month study reduced Beck Depression Inventory scores of patients with depression significantly over scores of control subjects (P < 0.001) and decreased glycosylated hemoglobin levels, but not to a significant degree. Long-term studies confirming continued improvement in either depression or glycemic control have yet to be performed with either psychopharmacologically based or psychotherapy-based treatment modalities.

ECT is considered to be a valuable option in the treatment of depression in diabetic patients, despite some degree of controversy. In patients with severe insulin-dependent diabetes, ECT can transiently elevate blood glucose. In this population, additional and more frequent monitoring regimens should be followed. When necessary, modifications in the amount or timing of either insulin or hypoglycemics should be made to ensure patient compliance. Some authors have argued that ECT may lead to short-term hypoglycemia but long-term hyperglycemia in a mechanism similar to that of norepinephrine-based antidepressants.