Depression as a Comorbidity to Diabetes: Implications for Management

Andrea A. Riley; Mindy L. McEntee; Linda Gerson; Cheryl R. Dennison


Journal for Nurse Practitioners. 2009;5(7):523-535. 

In This Article

Management of Depression in Diabetes

To optimize patient outcomes, PCPs should have a thorough understanding of research findings related to the treatment of depression in diabetes. While necessary, provider education and screening alone had little impact on patient outcomes and other barriers to management.[14,87,88] Depression remains inadequately treated in primary care,[2,13,31,89] with patients receiving inappropriately titrated doses of antidepressants[90] and up to 50% discontinuing use prior to achieving remission.[91] Both psychopharmacology[92–101] and psychotherapy[102–107] have been shown to effectively treat depression in diabetic patients, yet studies estimate only 33% to 52% of these patients were prescribed antidepressants,[2,32,108] while less than 20% completed 4 or more counseling sessions.[32]

Antidepressants are the most common pharmacological treatment option, although PCPs class selection must take into account a patient's presenting symptoms, comorbid medical conditions, drug interactions, and side effects. Because most antidepressants have little impact on the course of depression once discontinued and relapse following successful treatment is common,[89] a comprehensive treatment approach by the PCP should consider a referral for psychotherapy in addition to initial treatment and maintenance with medication. Both psychotherapy and medications are effective treating depression in diabetics, although the combination of these approaches has been shown to be more effective than either treatment alone in nondiabetic samples.[109] Monoamine oxidase inhibitors (MAOIs) are not recommended in diabetic patients due to associated dietary restrictions, weight gain, and the potential for sudden, severe hypoglycemia.[97,99] Research findings and considerations for the use of TCAs, SSRIs, newer antidepressants, and cognitive behavior therapy (CBT) are summarized below.

Tricyclic Antidepressants

TCAs are the oldest class of antidepressants and were commonly prescribed in primary care before the development of safer alternatives like SSRIs. Serious risks associated with the use of TCAs limit their current use, most notably due to cardiac effects[92,110–112] and lethality in overdose.[113] Weight gain,[92,97,110,114] hyperglycemia,[92,97] anticholinergic effects,[110] and impaired memory[97] also contribute to the limited use of these medications in patients with diabetes. In spite of their side effects, TCAs may aid in regulating sleep[110,112] and are well established as a treatment for diabetic neuropathy; both desipramine and amitriptyline were shown to be more effective treating neuropathy than the SSRI fluoxetine in a randomized, double-blind, placebo-controlled study.[115] While the TCA nortriptyline effectively reduced depressive symptoms among diabetics in a double-blind, placebo-controlled trial, it produced an adverse hyperglycemic effect unrelated to the drug's effect on weight. A similar increase in fasting blood glucose was found in nondiabetic patients following treatment with imipramine,[116] adding to the concern of TCAs in individuals with diabetes. Consequently, TCAs are not recommended as a first-line treatment for this population, and should only be considered when patients are unresponsive to trials of at least 2 newer antidepressants.

Selective Serotonin Reuptake Inhibitors

SSRIs are frequently a first-line choice of antidepressants in primary care, treating depression as effectively as TCAs with fewer adverse effects.[112,117,118] SSRIs have received a significant amount of study in diabetic patients, examining the effect of these drugs on depression as well as diabetes outcomes and self-care. Fluoxetine,[93,100,116] sertraline,[95,98,119] and paroxetine[100,101,120] were all effective treating depression in this population; fluoxetine and paroxetine were also associated with improved quality of life.[100] There was little evidence for superior effectiveness of one SSRI over another in any study, although patients treated with fluoxetine experienced a reduction in depression severity 2 weeks earlier than those receiving paroxetine.[100] Fluoxetine has also been associated with moderate weight loss,[121] though the effect of other SSRIs on weight has been inconsistent.[114] While SSRIs are associated with substantially fewer anticholinergic and cardiovascular effects, side effects such as sexual dysfunction, gastrointestinal distress, and agitation remain fairly common in diabetics and should be taken into consideration.[110]

Several randomized controlled trials have demonstrated a trend toward improvement in HgbA1c levels with SSRI treatment,[93,95,100,120] with statistically significant change in 2 open-label studies.[95,98] Musselman et al noted that the metabolic effect of SSRIs on certain oral antihyperglycemic agents may increase the risk of hypoglycemia, yet no clinical interaction of these drugs have been reported.[68] While the majority of treatment studies ranged from 8 to 12 weeks, a year-long maintenance regimen of sertraline following successful treatment found patients taking the drug were less likely to experience recurring depression than those receiving placebo.[95] Moreover, the initial improvement in glycemic control during treatment was sustained throughout the maintenance phase.[95] Based on this research, PCPs can prescribe an SSRI to treat depression with reasonable confidence that the medication will not worsen the course of diabetes and may even contribute to better glycemic control.

Research examining the use of SSRIs in the treatment of diabetic neuropathy is limited. Randomized, double-blind, placebo-controlled trials have demonstrated the effectiveness of paroxetine[122] and citalopram,[123] although a number needed to treat analysis suggests they were less effective than TCAs.[124] Two 8-week open trials found sertraline improved neuropathic symptoms, yet its effectiveness has not been compared to other treatments.[125] Fluoxetine was the only SSRI studied in the treatment of depressed patients with neuropathy; although the drug was associated with improved neuropathic symptoms, its effects were no different than placebo.[115] Thus, while SSRIs are better tolerated than TCAs and have been well established to effectively treat depression, they are not deemed a sufficient choice for neuropathy.[126]

Newer Antidepressants

Several other new antidepressants are widespread in use, including serotonin-norepinephrine reuptake inhibitors (SNRIs), norepinephrine-dopamine reuptake inhibitors (NDRIs), and noradrenergic and specific serotonergic antidepressants (NASSAs). Like SSRIs, antidepressants in these classes are generally considered safer alternatives to TCAs for the treatment of depression. Many of these newer drugs may have indications for use either as a first-line treatment or when patients do not respond to SSRIs, although research on their use in diabetic patients is extremely limited.

SNRIs have been shown to effectively treat diabetic neuropathy in addition to depression. Duloxetine was approved by the FDA for this purpose in 2004 and is generally well-tolerated, with fewer adverse events than routine care.[127,128] While associated with a modest increase in fasting glucose and A1C, long-term studies have found no adverse effects of duloxetine on lipid profiles or weight.[127–129] The relative efficacy of another SNRI, venlafaxine, was comparable to TCAs and gabapentin for the treatment of neuropathy,[130] although this drug was associated with higher rates of adverse serotonergic effects and dose-dependent hypertension than duloxetine.[131]

Other side effects are similar to SSRIs and remain more conducive to use in diabetic patients than TCAs,[132] although these drugs should not be abruptly discontinued due to risk of serious withdrawal symptoms. These findings suggest that SNRIs may be a viable treatment option for depressed patients with diabetic neuropathy, although these drugs have yet to be studied in patients with comorbid diabetes and depression.

The NDRI bupropion is the only newer antidepressant to be studied in diabetic patients with depression. Unlike SSRIs, bupropion effectively treated depression without sexual side effects[96,133] or concern of altering the metabolism of oral hypoglycemic medications,[68] although increased anxiety, nausea, dizziness, and skin irritation were reported. Implications for long-term use of bupropion were promising: patients who stayed on maintenance treatment of the drug had no recurring depression.[96] Bupropion has also been shown to be useful for smoking cessation in diabetics and has minimal impact on weight, ranging from no change to modest loss.[96,133] Only one study has examined the effects of this NDRI on neuropathy, although the drug was more effective than placebo with nearly 70% of patients reporting improvement.[134] Like SNRIs, the NDRI bupropion may be an alternative to SSRIs in diabetic patients with neuropathy, although it should be noted that this is the only antidepressant not effective in relieving symptoms of anxiety.[135]

The NASSA mirtazapine has received the least study in the diabetic population. A randomized double-blind trial found mirtazapine relieved depressive symptoms significantly earlier than fluoxetine and was equally tolerable without sexual side effects or other adverse events related to serotonin.[136] Commonly reported side effects of somnolence, hyperphagia, and weight gain have led to some concern using mirtazapine in patients with diabetes; although patients taking the drug have experienced significant weight gain, glycemic control actually improved.[137,138] While no research to date has examined the use of mirtazapine in diabetic neuropathy, the drug has been shown to increase pain tolerance in healthy individuals[139] and provide significant relief to depressed patients with chronic pain.[140]

Cognitive Behavior Therapy

Several studies have shown CBT and its variants to be successful in relieving symptoms of depression in diabetics.[102–106,141] Unlike antidepressants, there is evidence that the beneficial effects of CBT are maintained for up to a year past the end of treatment.[103,105] Therapy has also been shown to improve patients' self-efficacy[104] and diabetes management,[102] while glycemic control improved in some[102,105,106,141] but not all[103–105,141] studies. Psychotherapeutic approaches are beneficial in that they lack the side effects of medication, but require active patient participation and may not be well received in all patients. Poor integration with specialty mental health care and the stigma associated with mental illness are also barriers to this form of treatment.[29,89] Providers should discuss the option of psychotherapy with depressed diabetic patients as a part of a comprehensive treatment plan, but be aware of potential barriers to this course of treatment.