Galectin-3 Binding Protein, Depression, and Younger Age Were Independently Associated With Alexithymia in Adult Patients With Type 1 Diabetes

Aims: Alexithymia has been linked to cardiovascular disease. The aim was to explore whether the immuno-inflammatory variables galectin-3 binding protein (Gal3BP), soluble (s)CD163 and galectin-3 were independently associated with alexithymia, while controlling for known risk factors for cardiovascular disease, such as depression, anxiety, impaired glycemic control, obesity, smoking, and physical inactivity in patients with type 1 diabetes (T1D). Methods: Cross-sectional design. The participants were consecutively recruited from one diabetes out-patient clinic. Alexithymia, depression and anxiety were assessed by self-report instruments. Blood samples, anthropometrics, and blood pressure were collected, supplemented with data from electronic health records. High Gal3BP was defined as ≥3.3 μg/ml, high sCD163 as ≥0.6 μg/ml, high galectin-3 as ≥2.6 ng/ml, impaired glycemic control as HbA1c >70 mmol/mol (>8.6%) and abdominal obesity as waist circumference ≥ 1.02 m for men and ≥ 0.88 m for women. Results: Two hundred and ninety two patients participated (men 56%, aged 18–59 years, alexithymia prevalence 15%). Patients with alexithymia had higher prevalence of depression (34 vs. 6%, p < 0.001), anxiety (61 vs. 30%, p < 0.001), high Gal3BP levels (39 vs. 17%, p = 0.004), high HbA1c levels (46 vs. 24%, p = 0.006), and abdominal obesity (29 vs. 15%, p = 0.045). Depression [adjusted odds ratio (AOR) 6.5, p < 0.001], high Gal3BP levels (AOR 2.4, p = 0.035), and age (AOR 0.96, p = 0.027) were independently associated with alexithymia. Abdominal obesity (AOR 4.0, p < 0.001), high Gal3BP levels (AOR 2.8, p = 0.002), and depression (AOR 2.9, p = 0.014) were associated with high HbA1c. Abdominal obesity and anxiety were associated [Crude odds ratio (COR) 2.4, p = 0.006]. Conclusions: T1D patients with alexithymia had higher prevalence of high Gal3BP levels, depression, impaired glycemic control, anxiety, and abdominal obesity, which are known risk factors for cardiovascular disease. Only high Gal3BP levels, depression, and younger age were independently associated with alexithymia in adult patients with T1D.


INTRODUCTION
Alexithymia literally denotes "no words for feelings." Alexithymia is a state or personality trait characterized by limited ability to identify and describe feelings, and to distinguish between sensations in the body caused by emotional arousal and sensations of other origins (1)(2)(3). An externally oriented cognitive style and constricted imaginal processes are other features of alexithymia (1)(2)(3). Alexithymia is also characterized by inflexible emotion regulation with inadequate emotional reactivity, prolonged emotional states and failed habituation (4). Emotional neglect, physical or sexual abuse during childhood, as well as neuropsychiatric and genetic disturbances may cause alexithymia (5)(6)(7).
As alexithymia previously has been linked to cardiovascular disease and mortality (8,9), we wanted to explore whether selected known risk factors for cardiovascular disease and mortality were directly or indirectly linked with alexithymia in patients with T1D. The main aim was to explore whether the immuno-inflammatory variables Gal3BP, sCD163, and galectin-3 were independently associated with alexithymia controlling for depression, anxiety, metabolic and life style factors, medication, and cardiovascular complications.

Participants and Study Design
The study has a cross sectional design and included 292 (69%) patients out of 424 eligible patients with T1D (15). Inclusion criteria were T1D of ≥1-year duration, in patients 18-59 years of age. Exclusion criteria were pregnancy, severe somatic and psychiatric disorders such as cancer, hepatic failure, end-stage renal disease, Cushing's disease, severe autoimmune disorders, psychotic disorders such as schizophrenia, bipolar disorder, severe personality disorders, severe substance abuse, cognitive deficiency (due to stroke, dementia, or intellectual disability), or inadequate knowledge of the Swedish language. No patients used antipsychotic drugs (15). The patients were consecutively recruited during a 9-month period, from 25 March 2009 to 28 December 2009, from the largest out of two hospital diabetes outpatient clinics in Region Kronoberg, Sweden. The catchment population was 125,000. Self-report questionnaires were used to assess alexithymia, depression and anxiety. Blood samples, anthropometrics, and blood pressure were collected, supplemented with data from electronic health records (12,15,31,(40)(41)(42)(47)(48)(49). The study was performed in accordance with the Declaration of Helsinki, and was approved by the Regional Ethical Review Board of Linköping University, Linköping (Registration no. M120-07, T89-08). All participants provided written informed consent.

Definitions of Abdominal and General Obesity, Severe Hypoglycaemia Episodes, and Cardiovascular Complications
Abdominal obesity was defined as waist circumference (WC) ≥1.02 m for men and as WC ≥0.88 m for women, and general obesity was defined as BMI ≥30 kg/m 2 for both sexes. A severe episode of hypoglycaemia was defined as hypoglycaemia to such a degree that the patient needed help from another person. Severe hypoglycaemia episodes occurring during a period of 6 months prior to inclusion were included. Cardiovascular complications were defined as cardiac failure, ischemic heart disease, stroke, or transient ischemic attack (TIA) (15).

Smoking and Physical Activity
Smokers were defined as having smoked any amount of tobacco during the last year (15). There were four categories of physical activity including at least 30 min of moderate activity: less than once a week, 1-2 times a week, 3-5 times a week, and daily.

Blood Pressure, Antihypertensive Drugs, and Indications for Treatment of Hypertension
Blood pressure was measured in the sitting position by a nurse. The antihypertensive drugs which were used by the patients were previously described (41). The use of antihypertensive drugs was dichotomized into users and non-users (41). Indications for antihypertensive drugs were systolic blood pressure >130 mm Hg and/or diastolic blood pressure >80 mm Hg (41).

Serum-Lipids, Lipid-Lowering Drugs, and Indications for Treatment of Hyperlipidemia
Serum-lipids were collected after an overnight fast. Lipid lowering drugs were HMG CoA-reductase inhibitors, and the use of lipid lowering drugs was dichotomized into users and non-users (41). Indications for lipid lowering drugs were total cholesterol (TC) >4.5 mmol/l (>1.74 mg/dl) and/or LDLcholesterol >2.5 mmol/l (>97 mg/dl) (41).

Insulin
Patients used either multiple daily insulin injections (MDII) or continuous subcutaneous insulin infusion (CSII).

Antidepressants
The antidepressants which were used by the patients were previously described (41). The use of antidepressants was dichotomized into users and non-users.

Statistical Analysis
Analysis of data distribution using histograms revealed that age, diabetes duration, Gal3BP, sCD163, galectin-3, triglycerides, systolic, and diastolic BP, were not normally distributed. Data were presented as median [quartile (q) 1 , q 3 ; minmax], and analyses were performed with Mann-Whitney U test. Fisher's Exact Test (two-tailed) and Linear-by-linear association were used to analyse categorical data, and data were presented as N (%).
Medians and prevalence rates for the variables included in the study were compared between patients with and without alexithymia. Variables where these comparisons showed P < 0.10, and sex, age, and cardiovascular complications independent of P-values, were included in the further analyses. Crude odds ratios (CORs) for these variables were calculated with alexithymia as dependent variable. Variables with P < 0.10 for the CORs, and age and sex independent of P-values, were entered into multiple logistic regression analysis (Backward: Wald) with alexithymia as dependent variable in three separate models, and with HbA1c >70 mmol/mol as dependent variable in one model, and with cardiovascular complications as dependent variable in one model (54), The Hosmer and Lemeshow test for goodness-of-fit and Nagelkerke R 2 were used to evaluate each multiple logistic regression analysis model. ROC analyses were performed. After log transformation (Lg10) of Gal3BP, multiple linear regression analysis (Backward) was performed with Gal3BP (Lg10) as dependent variable and with alexithymia, depression, and anxiety as independent variables. Bivariate and partial correlation analyses were performed. Pearson correlation coefficients, collinearity tolerance coefficients and variance inflation factor (VIF) coefficients were calculated. Confidence intervals (CIs) of 95% were used. P < 0.05 was considered statistically significant. SPSS R version 25 (IBM, Chicago, Il, USA) was used.
In bivariate correlations analyses, the Pearson Correlation coefficients were for Gal3BP and alexithymia 0.330 (p < 0.001), and for Gal3BP and depression 0.125 (p = 0.035). When partial correlation analysis was performed, controlling for alexithymia, the correlation coefficient for Gal3BP and depression was 0.044, (p = 0.46).
Collinearity analyses were calculated with alexithymia as dependent variable, and the included independent variables were age, diabetes duration, sex, depression, anxiety, abdominal obesity, Gal3BP, sCD163, HbA1c, and HDL. The results of the collinearity analyses showed that the collinearity tolerance coefficients were all ≥ 0.67 and the VIF coefficients were all ≤ 1.49.
Associations with high HbA1c (>70 mmol/mol) are presented in Table 4

Main Findings
The main findings of this study of 292 adult patients with T1D were that high Gal3BP levels, depression and younger age were independently associated with alexithymia. Abdominal obesity, anxiety and high HbA1c levels (> 70 mmol/mol) were associated with alexithymia in univariate analyses. Abdominal obesity was only in combination with anxiety associated with alexithymia in multivariate analyses. Furthermore, high HbA1c levels were not independently associated with alexithymia though high HbA1c levels were associated with three different variables, abdominal obesity, high Gal3BP levels and depression, see Figure 1. The  Frontiers in Psychiatry | www.frontiersin.org  main alexithymia subfactor was difficulty identifying feelings (DIF), which was correlated with Gal3BP.

Alexithymia and Immunity
Gal3BP, which previously was linked to metabolic disturbances, cardiovascular disease and mortality (21,22,27), was the only immuno-inflammatory variable independently associated with alexithymia. Gal3BP is important as it has the potential to induce pro-inflammatory cytokines (21,26). Higher levels of Gal3B has been demonstrated in people with diabetes mellitus compared to people without diabetes (21). In these T1D patients, there was an association between high Gal3BP levels and impaired glycemic control expressed as high HbA1c levels. In patients with previous myocardial infarction, high Gal3BP levels and diabetes mellitus were independently associated with increased risk for the development of angina pectoris or reinfarction (22). Gal3BP has been suggested as a novel important biomarker for cardiovascular risk assessment (22). To our knowledge, an association between alexithymia and Gal3BP has not previously been demonstrated.

Alexithymia and Depression
Depression is a clear risk factor for cardiovascular disease and all-cause mortality (38). The patients with alexithymia in this study had a very high prevalence of depression, which is in accordance with previous research (9,(14)(15)(16). Alexithymia has been suggested as a risk factor for depression even though alexithymic features seem to increase during depressive episodes (55,56). Several immuno-inflammatory disturbances have been demonstrated in depressive states such as increased levels of galectin-3 and decreased soluble TWEAK (31,32,41,43). Galectin-3 not only serves as a biomarker, but is also a contributor to cardiovascular disease and mortality (22,24,25).

Alexithymia and Metabolic Disturbances
Alexithymia has previously been linked to obesity and binge eating disorders (11). Obesity is a major risk factor for cardiovascular disease (13,21,34) and there is some evidence that anxiety contributes to cardiovascular disease (39). Anxiety and abdominal obesity were closely linked in this study, and these two factors combined were associated with alexithymia. One possible mechanism is that patients with alexithymia cannot discriminate between the unease caused by anxiety and the unease caused by hunger. Another possibility is that people with alexithymia cannot distinguish between symptoms of hypoglycaemia and anxiety. Difficulty in distinguishing between bodily sensations due to emotional arousal and due to other causes is a main characteristic of alexithymia (2). In both cases, the consequences of this particular alexithymia feature may be increased food intake with subsequent weight gain. Obesity might also cause anxiety, as patients with diabetes are regularly informed about the deleterious impact obesity may have on their health (34,49). The patients with and without alexithymia did neither differ by levels of blood pressure or s-lipids, nor by the prevalence of antihypertensive medication or lipid-lowering drugs usage, which differ from previous research (13).

Alexithymia and Life Style
Smoking habits and levels of physical activity didn't differ between patients with and without alexithymia.

Potential Explanatory Mechanisms
As this study is a cross-sectional study, we cannot determine any causality. We will, however, discuss potential explanatory mechanisms. There is evidence from previous research that parts of the innate immune system which during the evolution were developed to fight pathogens, might be activated by environmental psychosocial stress or by the anticipation of perceived danger (43,44). Persons with alexithymia may be inclined to anticipate danger, particularly if they during childhood were exposed to physical abuse, emotional neglect or other types of trauma, which are risk factors for alexithymia (5). According to previous research alexithymia is characterized by inflexible emotional dysregulation with dysfunctional emotional reactivity and prolonged emotional states (4), which potentially could trigger prolonged immune reactions in case of perceived or anticipated danger. The immune system acts in a coordinated way with the sympathetic nervous system and the HPA-axis (2,3,43). The increased baseline sympathetic activity or the disturbances of the HPA axis observed in alexithymia (2, 3), may trigger the immune system with increased levels of Gal3BP. The increased Gal3BP levels are deleterious for patients with T1D as Gal3BP induces a number of pro-inflammatory cytokines (21). Increased levels of Gal3BP contribute to metabolic disturbances (27), and increased Gal3BP levels are risk factors for cardiovascular disease and all-cause mortality (21,22).
Blood glucose levels also increase by activation of the stress system (57), which may contribute to the demonstrated increased HbA1c levels.

Future Research
In future research it would be of interest to explore the impact of alexithymia, Gal3BP and depression on cardiovascular complications in a longitudinal study. Another subject for research would be to explore whether therapies aiming at increased emotional awareness could ameliorate alexithymia features (51). At a molecular level it would be of interest to further explore stress, alexithymia, depression, sterile inflammation and danger-associated molecular patterns (DAMP) (44).

Strengths and Limitations of the Study
Strengths of our study are that patients with severe comorbidities such as cancer, severe autoimmune disorders, hepatic failure, end-stage renal disease, psychotic and bipolar disorders were excluded as these disorders, or medication for these disorders, may have impact on the immune system (26,27). We controlled for relevant variables previously linked to cardiovascular disease and mortality. The logistic regression models were elaborated for the associations, and calibrated and validated for goodness of fit with the data variables. The included variables were checked for multicollinearity, and no significant evidence of multicollinearity was demonstrated. ROC analyses were performed and the area under the curve was only significant for Gal3BP and alexithymia, not for depression. Multiple linear regression analysis was performed and alexithymia was associated with log transformed Gal3BP, which was not the case for depression. Altogether, these analyses support the result of the multiple logistic regression analyses, Gal3BP and depression were independently associated with alexithymia. Finally, precise ELISA techniques were used and showed low intra-assay coefficients of variation for Gal3BP, sCD163 and galectin-3 (31,47,48).
Cross-sectional studies have limitations as causality cannot be confirmed. Only a limited number of patients with cardiovascular disease participated in the study, so an association between Gal3BP and cardiovascular complications could neither be confirmed nor excluded. The P-value was quite low, which could indicate a type-2 error. Depression, anxiety and alexithymia were only assessed by self-report instruments, not by a structured interview. TAS-20 is, however, extensively used in research, and evidence collected for 26 years supports that the scale adequately measures the alexithymia construct (3,8,9,11,14,17,18,52). HADS has also shown high validity for assessing symptoms of depression and anxiety both at an individual and a collective level (53).

CONCLUSIONS
Gal3BP, depression and younger age were independently associated with alexithymia. The patients with alexithymia also had higher levels of HbA1c, and higher prevalence of abdominal obesity and anxiety, but these factors were not independently linked to alexithymia. The findings of this study contribute to the understanding of the complex processes which may contribute to the development of cardiovascular disease in patients with T1D and alexithymia.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by Regional Ethical Review Board of Linköping University, Linköping, Sweden (Registration no. M120-07, T89-08). The patients/participants provided their written informed consent to participate in this study.

AUTHOR CONTRIBUTIONS
RS, JD, MH, and EM participated as investigators and reviewed, edited, and approved the final version of the manuscript. RS contributed with his knowledge of psychology and initiated the research group's interest in the health implications of alexithymia. JD and MH contributed with their knowledge of immunology, performed the analyses of the immunoinflammatory variables, and they take the full responsibility for these analyses. EM integrated the knowledge of psychology, clinical medicine and immunology, performed the statistical analysis, is the guarantor of this work and, as such, had full access to all the data in the study, and takes responsibility for the integrity of the data and the accuracy of the data analysis.

FUNDING
This research was supported by the Southern Healthcare Region, Lund, Sweden; by the Research Council of South Eastern Sweden (FORSS), Linköping, Sweden (708401); and by the Research and Development Fund of Health Care, Region Kronoberg, Växjö, Sweden (936138). The funding sources were not involved in the collection, analysis and interpretation of data, in the writing of the report, or in the decision to submit the article for publication.