Introduction

Despair, depression, anxiety are the common negative emotions in type 2 diabetes. Some studies indicated that the prevalence of anxiety and cognitive impairment were significantly increased in diabetic patients [1,2,3,4]. The proportion of depressive symptoms in diabetic patients is increasing gradually, the rate of depression in diabetes is approximate 20%-30%. More serious consequences occur in diabetes accompanied with emotional disorder patients [5,6,7]. An investigation including 328 T2DM was conducted by Liu's research group and the results showed that the prevalence of anxiety disorders in the patients with complications was significantly higher compared with diabetic patients without complications (48.76% VS. 24.33%) [8]. It showed that emotional disorders were associated with diabetic complications. Bogner et al. have suggested that the higher risk of death was found in diabetic patients combined with depression and anxiety [9]. Endocrinologists suggested that early identification of emotional disorders and multidisciplinary therapy could improve the prognosis of the disease, reduce the incidence of complications and mortality [10, 11]. Recent studies showed that blood glucose fluctuation was closely related to multiple complications, depression and anxiety in T2DM [12, 13]. Standard Deviation of Blood Glucose (SDBG), postprandial glucose excursion (PPGE) and largest amplitude of glycemic excursions (LAGE) were the common indicators of blood sugar fluctuations, which might be associated with depression and anxiety in diabetic patients.

Therefore, to assess the psychological condition of diabetic patients is beneficial for accurate diagnosis, treatment and blood glucose control. Early identification of poor mental state and potential risk factors are helpful to improve the life quality of diabetic patients.

Materials and methods

Study population

A total of 182 T2DM, who were hospitalized in endocrinology department of Jiading District Central Hospital from June 2019 to Apr 2021, became the research subjects of this study. The enrolled patients were distributed into comorbidity group and control group according to the diagnosis of anxiety and depression. After screening, the number of patients was 81 in the comorbidity group and 101 in the control group.

Inclusion and exclusion criteria

Inclusion criteria

  • i) T2DM was diagnosed according to the WHO diagnostic criteria in 2013.

  • ii) Anxiety and depression must be clearly diagnosed.

  • iii) Patients must have sufficient data for the study, including basic information and clinical examination results.

Exclusion criteria

  • i) Patients with mental disorders who cannot complete questionnaires and assessment scales were excluded.

  • ii) Diabetic patients with severe acute complications (serious infection, ketoacidosis, hyperosmolar coma, diabetes foot), hepatic or renal insufficiency, heart deficiency, malignant tumor, malignant anemia, had surgery should be ruled out.

  • iii) Patients receiving hormone therapy or antidepressant drug treatment were excluded.

Research methods and emotional disorders assessment

A questionnaire survey regarding the general information (such as age, gender, body mass index (BMI), smoking status, family history and other medical history materials), behaviors, life style and frequency of food consumption was conducted among the 182 enrolled participants. Meanwhile, the trained physicians assessed the emotional disorders by a variety of scales. The self-rating anxiety scale (SAS) was used to evaluate the state of anxiety. There are 20 items in SAS, the score of each item (range: 1–4) was depending on the severity. The final score was equal to the total score of 20 items multiplied by 1.25 and the standard score was greater than or equal to 50 indicate anxiety. Depressive position was estimated using PHQ-9 scale (0–4: no depression, 5–9: minor depression, 10–14: moderate depression, 15–19: moderately severe major depression, 20–27: severe major depression). Cognitive disorder was recognized by mini mental status examination (MMSE). The total score of the scale was 30 points and the person who score greater than 27 points was thought to be healthy. All patients signed the medical informed consent and agreed to participate in this research. Meanwhile, this study was approved by our local ethics committee.

Biochemical examination

Venipuncture was used to obtain venous blood and the samples were frozen in -70℃ refrigerator. HbA1c, serum cholesterol, hydrocortisone, thyroid function index, blood calcium, blood phosphorus and other indicators were tested in the two groups. Serum biochemicals were measured by automatic biochemical analyzer (Roche D/P/ISE, Switzerland) and HbA1c was measured by high-performance liquid chromatography (HLC-723g7, Japan). C-peptide and cortisol were tested by the method of chemiluminescent immunoassay (Abbott architect i2000 and AutoLumo A2000Plus, USA).

Calculation method of blood glucose fluctuation index

The blood glucose levels of pre-prandial and postprandial 2 h blood glucose of three meals and blood glucose before bedtime were measured, then these glucose values were marked as a, b, c, d, e, f, g, respectively. The blood glucose fluctuation indicators including PPGE, LAGE and SDBG were calculated according to the above glucose levels and the calculation formulas are shown below:

$$\overline{\mathbf{x} }=\frac{a+b+c+d+e+f+g}{7}$$
$$\mathbf{P}\mathbf{P}\mathbf{G}\mathbf{E}=\boldsymbol{ }\frac{\left(b-a\right)+\left(d-c\right)+\left(f-e\right)}{3};$$
$$\mathbf{S}\mathbf{D}\mathbf{B}\mathbf{G}= \sqrt{\frac{{(a-\overline{x })}^{2}+{(b-\overline{x })}^{2}+{(c-\overline{x })}^{2}+{(d-\overline{x })}^{2}+{(e-\overline{x })}^{2}+{(f-\overline{x })}^{2}+{(g-\overline{x })}^{2}}{7-1}}$$

Statistical method

The software STATA version 12.0 (STATA Corp., College Station, Tex) was used to evaluate the collected data. Data consistent with the normal distribution were presented as mean ± standard deviation (Mean ± SD). The numeration data and categorical variables were compared by chi-square analysis or Fisher's exact test. Differences of continuous variables between the two groups were tested by Student’s t-test. Logistic regression analysis was used to identify the associated factors for depression and anxiety disorders in T2DM. The p value less than 0.05 was considered to be statistically significant.

Result

General information and clinical parameters

A total of 182 participants were enrolled into this study. The 81 diabetic patients were distributed into comorbidity group according to the diagnosis of anxiety and depression, the rest of the 101 T2DM were into control group. There were no significant differences in age, BMI, waistline, family history and the ratio of other chronic disease between the two groups. The basic information of the study population and analysis results were shown in Table 1. The results showed that the proportion of female in comorbidity group was significantly higher than the ratio of control group (p = 0.002). The clinical characteristics of all the participants were shown in Table 2. Compared with control group, the biochemical markers including thyroid function parameters and indicators of liver and kidney function were balanced and had no statistical differences. However, SDBG and PPGE in comorbidity group were higher than those in control group, and the differences were statistically significant (p = 0.032 and p = 0.037).

Table 1 General information of the study population
Full size table
Table 2 Clinical characteristics of the two groups
Full size table

Comparison of sleep status and life behavior

In the comorbidity group, people who had the habit of nap** account for 66.67%, which was higher than 55.45% in control group, but the difference was not statistically significant (p = 0.124). Significant differences of sleep quality and PSQI were existed in the two groups (p < 0.001 and p < 0.001, Fig. 1). The results of behavioral questionnaires survey (Table 3) showed the habits including alcohol, tea, and smoking habit of the two group had no statistically significant difference (p = 0.083, p = 0.65 and p = 0.095). The difference of dietary structure between comorbidity and control groups was statistically significant (p < 0.001).

Fig. 1
figure 1

Contrastive analysis of Pittsburgh sleep quality index and sleep quality

Full size image
Table 3 The comparison of sleep quality and behavioral style
Full size table

Stratification analysis

In this study, there were statistical differences in the gender ratio between the two groups. So the status of anxiety, depression and cognitive disorder in both male and female group were analyzed in this study. The analysis results showed that ratios of cognitive disorder, anxiety and depression in female patients were all significantly higher than that in male diabetes patients (p = 0.001, p < 0.001 and p < 0.001). The differences of MMSE score, SAS score and PHQ9 score were also statistically significant between female T2DM group and male patient group (p = 0.001, p < 0.001 and p < 0.001). The specific analysis results were shown in Table 4. In this research, there was a significant association between gender and depression, thus it was a confounding factor needing to be controlled.

Table 4 Anxiety, depression and cognitive disorder status in male and female group
Full size table

Analysis of the influence factors associated with anxiety and depression in T2DM

Logistic regression analysis was used to identify the influence factors for anxiety and depression disorders in T2DM. According to the comparative analysis results between the two groups, it was found that sex ratio, the blood glucose fluctuation index, slee** status and dietary habit were statistically significant. Therefore, sex ratio and other traditional risk factors such as smoking and alcohol consumption were considered as potential confounders. After controlling the confounders, the results showed that SDBG and sleep quality were associated with depression and anxiety disorders in T2DM (p = 0.040 and p < 0.001) and the OR values of these factors were 7.588 (1.097–52.069) and 4.428 (2.649–7.401), respectively. The male–female ratio, age, BMI, smoke, alcohol and dietary habit were not associated with depression and anxiety in T2DM (p = 0.801, p = 0.393, p = 0.337, p = 0.652, p = 0.489 and p = 0.828, separately). PPGE, LAGE and MBG had no effects on depression and anxiety in T2DM (p = 0.437, p = 0.180 and p = 0.836). The analysis results were shown in Table 5.

Table 5 Analysis of factors related to anxiety and depression in T2DM
Full size table

Discussion

In the current study we found that SDBG was significantly associated with depression and anxiety in T2DM (p = 0.032) and blood sugar that fluctuated widely was associated with depression and anxiety (OR = 7.558, p = 0.040). The T2DM patients having poor self-regulating ability might lead to a wide range of blood glucose fluctuations, multiple complications were emerging including common complications, psychological and emotional diseases [14,15,16]. The reason of SDBG fluctuation associated with the prevalence of depression and anxiety might be that long-term blood glucose disorder and increased complications could cause the proportion of anxiety and depression increasing.

Our results comported with several prior studies that good sleep quality, health of dietary patterns and regular behaviors were considered as the advantage factors, which could improve the depressive symptoms [17,18,19]. Logistic analysis results showed that sleep disorder was the risk factor for depression and anxiety in T2DM patients (OR = 4.428, 95%CI: 2.649–7.401, p < 0.001). The traditional risk factors such as smoking and alcohol, were not statistically associated with depression and anxiety in this study (p = 0.652 and p = 0.489). The cause might be that gender imbalance between the two groups and insufficient sample size.

According to Hussain's systematic review analysis, the prevalence of depression was 26.67% ~ 29% in diabetes [20]. However, the prevalence of emotional distress in our study was 45.05%, the increased prevalence of anxiety and depression might be probably associated with diagnostic mistakes, delayed therapy. Peyrot team's findings confirmed that gender was the independent risk factors for emotional disorders in diabetic patients [21]. Our results were consistent with those research findings, the female patients were more likely to suffer from anxiety, depression and cognitive disorder (p = 0.001, p < 0.001 and p < 0.001). As we all know that menopause can lead to endocrine disorders and emotional fluctuation in women, this might account for the difference in gender.

This research was a rigorous retrospective study and focused on the correlation of blood glucose fluctuation, sleep quality and the prevalence of depression and anxiety in T2DM. However, it should be noted that there were some limitations in this research. The research data were collected from the electronic medical record and questionnaires, it was impossible to eliminate information bias, selection bias and confounding bias completely. We can only minimize the effects of these biases by collecting data objectively and reasonable statistical analysis. Large-sample and multicenter studies were needed to clarify the causal relationship between blood glucose fluctuation and emotional disorders in T2DM.

Conclusion

To conclude, this retrospective analysis indicated that blood glucose fluctuation and sleep quality were associated with the increased prevalence of depression and anxiety in T2DM. It is known that early identification of poor mental state and potential risk factors are helpful to improve the life quality of diabetic patients.