Early Combination Therapy for Type 2 Diabetes Mellitus and Common Comorbid Mental Disorders

Brain tissue is highly dependent on glucose metabolism. Neuronal activity and biochemical processes require a steady supply of glucose. However, the central nervous system (CNS) has a minimum storage capacity, with low levels of glycogen or lipids. Thus, any changes in metabolic pathways may contribute to the pathophysiology of schizophrenia. A positron emission tomography study showed significantly decreased glucose uptake in all areas of the brain in patients diagnosed as having schizophrenia with negative symptoms. On the other hand, participants with a positive type of schizophrenia demonstrated markedly increased glucose metabolism in the parahippocampal region, striatum, and left thalamic area (4). These findings support the results of Dean et al.’s study, in which subjects given a diagnosis of schizophrenia had higher levels of pyruvate and acetyl-CoA in the striatum (5). Human studies of T2DM provide evidence that impaired glucose metabolism contributes to CNS damage, particularly within the hippocampus. Changes in hippocampal morphology in both, patients diagnosed as having schizophrenia and patients with T2DM, may contribute to cognitive deficits, mainly in immediate memory and attention. Research has shown that local delivery of insulin to the hippocampal area improves cognitive function (6). Collectively, these findings suggest that schizophrenia with comorbid diabetes results in more serious cognitive dysfunction than does schizophrenia alone (7). Several other authors came to the same conclusion regarding cognition of these patients. Takayanagi et al. (8) and Bora et al. (9) reported that patients with schizophrenia and comorbid T2DM demonstrated much worse cognitive performance than did patients with schizophrenia who did not have diabetes.

The mechanism of schizophrenia has been well defined in the literature. It is thought that dopamine, serotonin, and glutamate formulate the neurotransmitter hypothesis of schizophrenia, the same neurotransmitters found to interfere with glucose metabolism. This might partially explain the increased frequency of obesity, diabetes, and metabolic syndrome in individuals diagnosed as having schizophrenia. Dopamine has been shown to inhibit glucose-stimulated insulin production in the pancreas (10). Other studies demonstrated that insulin controls the brain’s supply of dopamine. Disrupted insulin signaling results in an excessive level of norepinephrine transporter protein. It would potentially lead to reduced dopamine and elevated norepinephrine in the prefrontal cortex. This situation creates hypodopaminergia (11). Could this vicious cycle of “dopamine-insulin-dopamine” explain the high rates of impaired glucose metabolism in patients given a diagnosis of schizophrenia, even prior to treatment with antipsychotics? This conclusion certainly deserves closer attention and further investigation. Serotonin, on the other hand, mediates insulin secretion in response to elevated glucose levels. Clozapine, perhaps the most effective antipsychotic agent, is a partial serotonin (5-HT2A) antagonist. Some authors offered a serotonergic model for clozapine- and olanzapine-associated diabetes. The review of animal data showed that 5-HT antagonism reduces insulin secretion by blocking pancreatic β-cell responsiveness, thereby increasing serum glucose levels (12). The data on glutamate and its influence on glucose metabolism are controversial. It was well documented that glutamate could induce insulin secretion and improve glucose tolerance in animals. It would have easily explained why patients who were diagnosed with schizophrenia, also known as “hypoglutamatergic” disorder (13), were predisposed to metabolic abnormalities. Today, some scientists are more careful and agree instead that glutamate “modulates” insulin production. However, the mechanism of this regulation is still unclear. Overactivation of the immune system (e.g., from prenatal infection or postnatal stress) may result in overexpression of inflammatory cytokines and subsequent alteration of brain structure and function. Patients diagnosed with schizophrenia have elevated levels of proinflammatory cytokines that activate the kynurenine pathway, producing metabolites that regulate glutamate receptor activity and may also be involved in dopamine regulation. Insulin resistance and metabolic disturbances, which are common in the schizophrenic population, have also been linked to inflammation. Thus, inflammation might be related both to the psychopathology of schizophrenia and to comorbid metabolic disturbances (14).

It is well known that dopamine influences prolactin secretion. Dopamine action is mediated by five dopamine receptors: DR1D, DR2D, DR3D, DR4D, and DR5D. Both increased dopamine function in some brain areas and decreased dopamine action in others may contribute to schizophrenia as well as to T2DM. Prolactin regulates β-cell survival and has a role in glucose metabolism via glucokinase activity and in insulin secretion (15). Sari et al showed that patients with poorly controlled diabetes have reduced serum prolactin levels (16). An altered dopamine pathway in schizophrenia may reduce the prolactin level, impairing carbohydrate metabolism and contributing to the onset of T2DM.

Patients with schizophrenia diagnoses and their first-degree relatives have a two- to five-fold greater risk of T2DM than does the general population, independent of body mass index and antipsychotic medication, suggesting that shared genetic components may contribute to both diseases (17).

Disrupted in schizophrenia 1 (DISC1) is a well-characterized schizophrenia susceptibility gene, also involved with several other psychiatric illnesses including bipolar disorder and major depression. It has also been identified as a major player controlling pancreatic β-cell proliferation and insulin secretion. Loss of DISC1 function results in decreased β-cell proliferation, increased apoptosis, glucose intolerance, reduced insulin secretion, and decreased critical β-cell transcription factors. This suggests that DISC1 dysregulation contributes to T2DM independently of its importance for cognition.

Occasionally, patients develop T2DM within a few weeks or months of beginning antipsychotic treatment—for the most part with clozapine or olanzapine. In 2015 Thomas highlighted the work of Koller and Doraiswamy, whose findings indicated that 80% of patients experienced improvements in glycemic control after discontinuing therapy with clozapine and olanzapine (18). Obesity, sedentary lifestyle, and social health determinants, such as income and housing, are major contributing factors for patients with both diabetes and schizophrenia diagnoses (19).

Antipsychotic medications are different in their propensity to cause weight gain and impair glucose tolerance. Clozapine and olanzapine, which are metabolically the most unfavorable antipsychotics, are also the most effective antipsychotic medications. However, medication changes may be possible, and antipsychotics with a low risk of metabolic side effects are preferred for people with schizophrenia with comorbid T2DM. Among antipsychotic agents, ziprasidone is the least likely to cause significant weight gain and should be considered in all patients (Table 1). Aripiprazole and other antipsychotics such as perphenazine, fluphenazine, and haloperidol can be considered as second-line agents for attenuation of weight gain. In primary care settings, such a change should be coordinated with a psychiatrist, because some patients require a certain medication, despite its metabolic side effect, because they did not respond to other medications in the past. In other cases, a medication switch poses a high risk for psychiatric decompensation, so treatment should be directed toward lifestyle changes or other indicated interventions (20). Individuals with schizophrenia may also be taking drugs other than antipsychotics that have metabolic side effects, for example, valproate and some antidepressants, and the possibility of changing these medications should also be assessed (21). It is important to screen for metabolic side effects such as metabolic syndrome. This syndrome is defined as a waist circumference of greater than 40 inches in men or 35 inches in women, high-density lipoprotein concentrations less than 40 mg/dL, and triglyceride concentrations above 150 mg/dL. Among patients with a mental disorder, adherence and follow-up may be an issue, and so glycosylated hemoglobin (HbA1c) testing may be preferable to obtaining a fasting glucose level as a screening test. For those taking antipsychotics, screening should be done at baseline and then at three-month intervals. If the HbA1c levels remain stable, screening can be reduced to six-month or one-year intervals (22). A minimum of 150 minutes of “moderate” intensity physical activity per week is recommended, although this may depend on the patient’s baseline activity levels and comorbidities (23). Metformin is usually indicated as a first-line pharmacotherapy for patients with type 2 diabetes who do not achieve adequate glucose control with diet and lifestyle change alone. Metformin, alone or in combination with lifestyle modification in the treatment and prevention of antipsychotic weight gain, has been studied in several meta-analyses and controlled trials (24). The best result was achieved by adding metformin to lifestyle changes (24). Among patients with chronic schizophrenia on olanzapine, metformin was effective in lessening both weight and HbA1c levels. Orlistat may also be indicated for patients with significant antipsychotic-induced weight gain (25).

Individuals with depression and comorbid diabetes showed decreased cortical gray matter thickness in bilateral prefrontal regions (30). A recent functional magnetic resonance imaging study demonstrated that patients with major depression had pathological activation within the medial prefrontal cortex. This correlates with specific symptoms among patients given a diagnosis of depression, such as feelings of guilt and rumination. Subjects with diabetes demonstrated more diffused functional changes, predominantly in frontotemporal region, hippocampus, and amygdala (31). T2DM is specifically linked to microvascular and neurodegenerative diseases. It is worthy of mentioning that depression among elderly populations is also associated with neurodegeneration, especially at the level of the hippocampus and prefrontal cortex (32). The association between T2DM and major depression is modest and understudied. It certainly requires further investigation.

The hypothalamic-pituitary-adrenal (HPA) axis responds to a stressor by triggering a cascade of reactions. It starts from corticotropin-releasing hormone (CRH) production in the hypothalamus, which stimulates the anterior pituitary gland to release adrenocorticotropin hormone, which then causes cortisol release from the adrenal glands. An excess amount of glucocorticoids can result in increased glucose production via stimulation of gluconeogenesis in the liver and skeletal muscles and reduced insulin sensitivity (33). There are several hypotheses for the mechanisms underlying the association between depression and cortisol-induced diabetes. It is suggested that glucocorticoid receptor (GR) abnormalities, such as their dysfunction or reduced receptor number, could cause hypercortisolemia and oversecretion of CRH. This observation was supported by the demonstration of GR impairment in postmortem studies of patients who had diagnoses of severe mood disorders. Researchers have found that tricyclic antidepressants have the ability to control GR expression and show greater therapeutic effect for patients with depression who have comorbid hypercorticolism (34).

The pancreatic tissue is extremely susceptible to different kinds of stress, such as oxidative stress and cellular stress. The pancreatic β cells respond to that stress with the initiation of an inflammatory process. Numerous studies have shown that tumor necrosis factor α (TNF-α), an inflammatory biomarker, was produced by adipose tissue and could be responsible for increased local and systemic insulin resistance (35). High serum levels of inflammatory cytokines, especially interleukin-6 (IL-6) and C-reactive protein (CRP), have been found in subjects with diabetes. Increased concentration of inflammatory agents might alter the tryptophan-kynurenine pathway and increase serotonin turnover. Thus, low-grade inflammation could be considered a possible link between major depression and T2DM. Serum levels of TNF-α, IL-6, and CRP in patients with diabetes and comorbid depressive symptoms were notably higher than in patients with diabetes mellitus alone (36).

To determine whether there is a relationship between T2DM and major depression, scientists performed a genetic analysis on 160,000 twins from Sweden and Denmark. When evaluating a sample of Swedish twins, there was a 31% correlation between diabetes and depression in males and a 75% correlation in females. When analyzing data from Danish twins, an overlap between diabetes and depression was found in 87% of men and 74% of women (37). However, there is little knowledge about the genetic interaction between T2DM and major depression. Therefore, more research is necessary on this topic. Risk factors for patients with diabetes developing depression include female gender, poor social support, stress, recurrent hypoglycemia, longer duration of diabetes, and presence of long-term complications (38).

While treating patients with depression and comorbid diabetes, primary care providers face tough challenges. Some patients reported that the stigma associated with depression prevented them from seeking help (39). Undiagnosed subclinical depression in diabetic patients requires more extensive measures, such as careful medical history taking and screening. Commonly used screening measures are Becks Depression Inventory, Well-Being Questionnaire, and Patient Health Questionnaire (PHQ). These are simple tools for general practitioners to diagnose depression, monitor patients over time, and guide them to the next level of disease management (40). Psychiatric referral is indicated for individuals with a history of severe emotional distress/ suicidal ideations or suicidal attempts in the past. In other cases, the primary care provider can request psychiatric consultation to clarify a proper treatment plan or if the patient fails to respond to antidepressants prescribed by the primary care physician. The goal of collaborative care of patients with comorbid depression and diabetes is not just to improve the courses of both diseases but also to prevent a number of complications and decrease the level of disability in such patients (41).

Cognitive-behavioral therapy can be effective at treating depressive symptoms and may also reduce HbA1c levels (42). Fluoxetine and sertraline demonstrated results consistent with lower glucose levels in T2DM patients with comorbid depression. Nortriptyline has shown a worsening of indices of glucose control. However, the improvement of depression with nortriptyline showed independent beneficial changes on HbA1c (43). Patients should be assessed for complications or long-term effects of diabetes. Bupropion, venlafaxine, or duloxetine should be considered before serotonin selective receptor inhibitors (SSRIs), when appropriate. They have been shown to relieve different kinds of chronic pain, associated with comorbid T2DM.

When triggered by a stressor, the HPA axis and sympathetic nervous system create a response. Corticosterone and catecholamines could also influence glucose metabolism. It is a well-known fact that epinephrine and norepinephrine can stimulate glucose production via gluconeogenesis and glycogenolysis (47). It could aggravate the course of T2DM in the setting of comorbid anxiety disorders. A recent animal study showed that mice with insulin resistance in the brain would exhibit a high rate of dopamine turnover, which clinically manifests in anxiety-like behavior (48).

Primary care providers might have difficulties in differentiating anxiety disorder with comorbid diabetes from primary anxiety illness. Some anxiety disorders can mask or mimic symptoms of hypoglycemia occurring among patients with diabetes. In other cases, patients with preexisting anxiety and T2DM can experience increased severity of their anxiety symptoms (49). All individuals with diabetes in the collaborating general practices should be screened for signs of anxiety, using the Patient Health Questionnaire-9 (PHQ-9) and General Anxiety Disorder-7 scale (GAD-7). If a patient scores below 7 for PHQ-9 and below 8 for GAD-7, only monitoring is needed. If patient scores above these cut-offs, education on “coping with anxiety” for at least 10 weeks should be offered. If this education is not helpful or if symptoms of anxiety worsen, treatment with medication is indicated (50). However, medications used for the treatment of anxiety disorders, such as SSRIs, benzodiazepines, and beta-adrenergic blockers, could possibly interfere with glycemic control and mask normal physiological warning signs of an upcoming hypoglycemic episode (51).