Depressive symptoms in asymptomatic stage B heart failure with Type II diabetic mellitus

Abstract Background The presence of concomitant Type II diabetic mellitus (T2DM) and depressive symptoms adversely affects individuals with symptomatic heart failure (HF). Hypothesis In presymptomatic stage B HF, this study hypothesized the presence of greater inflammation and depressive symptoms in T2DM as compared to non‐T2DM Stage B patients. Methods This cross‐sectional study examined clinical parameters, inflammatory biomarkers, and depressive symptoms in 349 T2DM and non‐T2DM men with asymptomatic stage B HF (mean age 66.4 years ±10.1; range 30‐91). Results Fewer diabetic HF patients had left ventricular (LV) systolic dysfunction (P < .05) although more had LV diastolic dysfunction (P < .001). A higher percentage of T2DM HF patients were taking ACE‐inhibitors, beta‐blockers, calcium channel blockers, statins, and diuretics (P values < .05). T2DM HF patients had higher circulating levels of interleukin‐6 (IL‐6) (P < .01), tumor necrosis factor‐alpha (P < .01), and soluble ST2 (sST2) (P < .01) and reported more somatic/affective depressive symptoms (Beck Depression Inventory II) (P < .05) but not cognitive/affective depressive symptoms (P = .20). Among all patients, in a multiple regression analysis predicting presence of somatic/affective depressive symptoms, sST2 (P = .026), IL‐6 (P = .010), B‐type natriuretic peptide (P = .016), and sleep (Pittsburgh Sleep Quality Index [P < .001]) were significant predictors (overall model F = 15.39, P < .001, adjusted R 2 = .207). Conclusions Somatic/affective but not cognitive/affective depressive symptoms are elevated in asymptomatic HF patients with T2DM patients. Linkages with elevated inflammatory and cardiac relevant biomarkers suggest shared pathophysiological mechanisms among T2DM HF patients with somatic depression, and these conditions are responsive to routine interventions, including behavioral. Copyright © 2019 John Wiley & Sons, Ltd.


| INTRODUCTION
The American College of Cardiology/American Heart Association (ACC/AHA) heart failure (HF) staging system denotes four unique stages that emphasize both the evolution and the progression of chronic HF. One of the major goals of the system is to identify individuals at early stages of the disease with the hope that early implementation of therapeutic interventions might ultimately reduce morbidity and mortality. 1 Stage B consists of patients who have developed structural or functional changes but who have never shown signs or symptoms of HF (eg, previous myocardial infarction, asymptomatic valvular disease, left ventricular hypertrophy, reduced ejection fraction). Progression from asymptomatic stage B to symptomatic stage C HF is associated with a 5-fold increase in premature mortality risk. 2 Like HF, Type II diabetic mellitus (T2DM) is a significant health problem, with rates expected to increase an estimated 25% over the next 7 years. 3 T2DM is a significant comorbidity because, along with hypertension and obesity, it represents an independent risk factor for the development of HF. 4,5 Like HF, T2DM is associated with chronic inflammation. 6 Among symptomatic HF patients, presence of T2DM predicts morbidity and mortality. 7 Major depression is common in cardiovascular diseases and associated with poor medical outcomes. 7 In symptomatic HF, studies have substantiated linkages between depression and inflammatory processes. 8,9 In T2DM, a history of depression is associated with a higher rate of microvascular complications and higher HbA1c. 10 Depression in T2DM independently and incrementally predicts incident HF. 11 In addition to major depression, depressive symptoms are also elevated in both HF 8,12 and in T2DM. 13 Screening for depression symptoms can be routinely accomplished using standardized questionnaires such as the Beck Depression Inventory II (BDI-II). 14 Several studies underscore the importance of differentiating the type of depressive symptoms the patient has, whether more somatic/affective or cognitive/affective depressive symptoms. For example, de Miranda Azevedo et al showed that in cardiovascular disease patients it is the somatic/affective depressive symptoms but not the cognitive/affective depressive symptoms that are related to poorer outcomes. 15 Similarly, Delisle et al showed that following myocardial infarction the somatic/affective depressive symptoms contributed more to overall depression than the cognitive/affective depressive symptoms. 16 We previously showed in symptomatic HF patients that a behavioral treatment for depressive symptoms was more successful in reducing somatic/affective rather than cognitive/affective depressive symptoms, which have been linked to worse prognosis in HF. 17 In contrast to the above research among patients with symptomatic HF (stage C), few studies have examined the comorbidity of depressive symptoms and T2DM in presymptomatic stage B HF. This paper sought to examine these relationships, and, in accordance with the studies described above which highlight the importance of differentiating somatic/affective vs cognitive/affective symptoms, [16][17][18] examine potential clinical and inflammatory predictors of these phenomena in asymptomatic HF with and without T2DM. Based on recommendations and cut-points from the American Society of Echocardiography guidelines, the presence of stage B HF was defined as structural or functional heart disease. 19 Measurements were made by sonographers blinded to patient characteristics. 20 Left ventricular ejection fraction (%LVEF) was assessed by echocardiography as part of the patient's routine medical evaluation. Medication usage was obtained from the medical record. T2DM was deemed present when diagnosed by the research participant's primary care physician via medical record abstraction. Upon presentation to the laboratory, a nonfasting blood draw was obtained using a 21-or 23-gauge butterfly needle. Patients performed a 6-minute walk test to assess functional capacity. 21

| Depressive symptom severity
The 21-item BDI-II assesses depressive symptoms, but does not provide a diagnosis of Major Depressive Disorder (MDD). 22  inhibitors. This study did not assess for presence of MDD but rather depressive symptoms only. The BDI provides a total score for depressive symptoms and two sub-scores: one, somatic/affective symptoms, and the other, cognitive/affective symptoms. 17 As noted in the Introduction, prior studies demonstrate the value of differentiating somatic/affective vs cognitive/affective depressive symptoms in cardiac patients. [15][16][17] BDI cognitive symptoms of depression include sadness, guilt, suicidality, and self-criticism. BDI somatic symptoms of depression include tiredness or fatigue, loss of interest, and changes in sleep, sex drive, and weight.

| Sleep quality
The Pittsburgh Sleep Quality Index (PSQI) was used to independently assess sleep. 23 Sleep quality is often disrupted in HF and associated with depressive symptoms, and common symptoms of somatic depression include poor sleep. 24 The PSQI is widely used in sleep research and measures sleep disturbance and usual sleep habits. Its T A B L E 1 Patient characteristics according to group (mean ± SD or percentage value)

| Data analysis
Analytic approaches included chi-square, analysis of variance, and multiple linear regression. All tests were two-tailed. Prior to statistical analyses, data were tested for normality and homogeneity of variance using the Kolmogorov-Smirnov and Shapiro-Wilk normality tests.
Results were considered statistically significant at the P < .05 level.

| RESULTS
A higher percentage of T2DM patients were taking Angiotensin converting enzyme inhibitors (ACE)-inhibitors (P < .02), beta-blockers (P < .02), calcium channel blockers (P < .05), statins (P < .001), aspirin (P < .02), and diuretics (P < .02). T2DM patients had higher circulating levels of IL-6 (P < .01), TNF-R1 (P < .01), and sST2 (P < .01). BNP levels were not statistically significantly different according to group (Table 2).  Table 3. The significant outcome predictors included sST2 (P = .026), IL-6 (P = .010), BNP (P = .016), and sleep (P < .001) (model significance P < .001; adjusted  previously observed in stage A HF patients and is associated with poorer prognosis. 27 T2DM patients also had higher circulating levels of IL-6, TNF-R1, and sST2, of which IL-6 and sST2 were found to be predictors of somatic/affective depression, which was also elevated in the T2DM patients. While we did not have the ability to fully characterize the presence of metabolic syndrome, given the relatively high percentage of diabetes, hypertension, and obesity in this cohort, it is likely that a large percentage of the patients had metabolic syndrome. 28 In T2DM, the comorbidity of depressive symptoms and of major depression is associated with poorer clinical management and poorer outcomes than either alone 29 ; the same pattern is found for coincident depression and HF. 7  Mechanistically, we found that sST2, IL-6, BNP, and poor sleep quality were independent predictors of somatic/affective depression while BMI, type of cardiac dysfunction, and medications were not.
Regarding IL-6, there is a large literature on the role of inflammatory processes in depression, including in HF, which point to neurohormonal and cytokine activation mechanisms linking these comorbidities. 8,10,32 There is also a large literature on inflammation and fatigue that is relevant to these findings of increased somatic depressive symptoms. 33 Limitations of this observational study include the demographics of mostly white older men, thus limiting potential generalizability. In addition, we did not assess for presence of major depression, but rather we focused on depressive symptoms which have been shown in the literature to be highly predictive of poor clinical course. 7 It is important to note that a significant percentage of these patients had depressive symptoms: 39% of the T2DM patients and 29% of the non-T2DM patients. As noted above, depressive symptoms and depressed mood are significant predictors of morbidity and mortality in cardiac and noncardiac populations. 7 Therapeutic lifestyle changes that target diet and exercise are known to positively influence depressed mood. 34,35 We have previously shown that standard cardiac rehabilitation is effective in not only reducing mortality risk but also in reducing depression symptoms in HF, 7 and that other forms of behavioral interventions favorably affect the somatic/affective but not the cognitive/affective depressive symptoms. 17 Given the elevated IL-6, TNF-R1, and sST2 levels in stage B HF patients with both T2DM and depressed mood, additional attention and interventions to forestall potential transition to symptomatic stage C HF may be warranted. 36 The pathophysiology of stage B HF, combined with the presence of T2DM, might synergize to advance the likelihood of poorer outcomes. In addition to depressed mood, stress is a known link between depression and inflammation and metabolic syndrome and could in part account for our observations. 33,37 Studies on approaches to stress reduction have also been shown to be helpful in improving depressed mood and inflammation. 38 Going forward, considering these findings, additional attention to assess depressive symptoms and presence of T2DM in stage B HF is warranted. Assessing depressive symptoms in the clinic using questionnaires such as the BDI or the Hospital Anxiety and Depression scale is relatively easy and low-cost and can provide insight into clinical outcomes. 39,40 ACKNOWLEDGMENT This work was supported by grants HL073355, H-096784, HL136407, UL1TR000100, and UL1TR001442 from the National Institutes of Health.