Ye ML, Wang JK. Association of total bilirubin with depression risk in adults with diabetes: A cross-sectional study. World J Clin Cases 2024; 12(18): 3428-3437 [DOI: 10.12998/wjcc.v12.i18.3428]
Corresponding Author of This Article
Man-Li Ye, MD, MSc, Chief Technician, Department of Laboratory Medicine, The Second Affiliated Hospital and Yuying Children's Hospital of Wenzhou Medical University, No. 109 Xueyuanxi Road, Lucheng District, Wenzhou 325000, Zhejiang Province, China. aikoye1207@126.com
Research Domain of This Article
Psychology, Clinical
Article-Type of This Article
Observational Study
Open-Access Policy of This Article
This article is an open-access article which was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/
Man-Li Ye, Department of Laboratory Medicine, The Second Affiliated Hospital and Yuying Children's Hospital of Wenzhou Medical University, Wenzhou 325000, Zhejiang Province, China
Jie-Ke Wang, Department of Hand and Plastic Surgery, The Second Affiliated Hospital and Yuying Children's Hospital of Wenzhou Medical University, Wenzhou 325000, Zhejiang Province, China
Author contributions: Wang JK analyzed and interpreted the data; Ye ML wrote the manuscript; All authors read and approved the final manuscript.
Institutional review board statement: The study was conducted in accordance with the Declaration of Helsinki, and approved by the National Center for Health Statistics Institutional Review Board/Ethics Review Board (NCHS IRB/ERB).
Informed consent statement: Informed consent was obtained from all subjects involved in the study.
Conflict-of-interest statement: The authors declare that they have no competing interests in this section.
Data sharing statement: The datasets analyzed in this study can be found on the NHANES website, available here: https://www.cdc.gov/nchs/nhanes/.
STROBE statement: The authors have read the STROBE Statement—checklist of items, and the manuscript was prepared and revised according to the STROBE Statement—checklist of items.
Open-Access: This article is an open-access article that was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: https://creativecommons.org/Licenses/by-nc/4.0/
Corresponding author: Man-Li Ye, MD, MSc, Chief Technician, Department of Laboratory Medicine, The Second Affiliated Hospital and Yuying Children's Hospital of Wenzhou Medical University, No. 109 Xueyuanxi Road, Lucheng District, Wenzhou 325000, Zhejiang Province, China. aikoye1207@126.com
Received: March 10, 2024 Revised: April 30, 2024 Accepted: May 17, 2024 Published online: June 26, 2024 Processing time: 100 Days and 1.6 Hours
Abstract
BACKGROUND
Individuals with diabetes mellitus are more likely to experience depression, although most patients remain undiagnosed. The relation between total bilirubin and depression has been increasingly discussed, but limited studies have examined the association of total bilirubin with depression risk in adults with diabetes, which warrants attention.
AIM
To investigate the association between total bilirubin levels and the risk of depression in adults with diabetes.
METHODS
The study included adults with diabetes from the National Health and Nutrition Examination Survey 2007-2018. Depression was determined using the Patient Health Questionnaire-9. Multivariable logistic regression, propensity score-matched analysis and restricted cubic spline models were utilized to investigate the association between total bilirubin levels and depression risk in adults with diabetes.
RESULTS
The study included 4758 adults with diabetes, of whom 602 (12.7%) were diagnosed with depression. After adjusting for covariates, we found that diabetic adults with lower total bilirubin levels had a higher risk of depression (OR = 1.230, 95%CI: 1.006-1.503, P = 0.043). This association was further confirmed after propensity score matching (OR = 1.303, 95%CI: 1.034-1.641, P = 0.025). Subgroup analyses showed no significant dependence of age, body mass index, sex, race or hypertension on this association. Restricted cubic spline models displayed an inverted U-shaped association of total bilirubin levels with depression risk within the lower range of total bilirubin levels. The depression risk heightened with the increasing levels of total bilirubin, reaching the highest risk at 6.81 μmol/L and decreasing thereafter.
CONCLUSION
In adults with diabetes, those with lower levels of total bilirubin were more likely to have depressive symptoms. Serum total bilirubin levels may be used as an additional indicator to assess depression risk in adults with diabetes.
Core Tip: Diabetic adults with lower total bilirubin levels had a higher risk of depression. We adjusted for confounders that might affect the association between total bilirubin and depression risk by analyzing the detailed covariate data. After propensity score matching, this association was further confirmed. Subgroup analyses illustrated that there was no significant dependence of age, body mass index, gender, race and hypertension on this association. Restricted cubic spline models displayed an inverted U-shaped association of total bilirubin with depression risk within the lower range of total bilirubin.
Citation: Ye ML, Wang JK. Association of total bilirubin with depression risk in adults with diabetes: A cross-sectional study. World J Clin Cases 2024; 12(18): 3428-3437
Diabetes is considered one of the major global health crises of the 21st century. In 2021, the global prevalence of diabetes reached one person in every 10 adults[1]. Studies have shown that diabetic patients are more likely than the general population to suffer from depression[2], which adversely affects the quality of life and diabetes outcomes[3,4]. Depression is a prevalent and severe psychiatric disorder. The global prevalence of depression has been increasing, particularly during the current coronavirus disease 2019 pandemic[5-8]. The co-morbidity of diabetes and depression increases the overall mortality rates, which is particularly evident in the older individuals[9,10]. Even though a considerable proportion of diabetic patients struggle with depression, most of them are still not diagnosed or treated[11,12]. Current hypotheses for this co-occurrence include disruptions of the inflammation systems, endocrine disorders and an inactive lifestyle[13-15]. However, much remains unclear. More research is required to reveal the mechanisms and risk factors of depression in patients with diabetes, especially in the early stages of life.
Bilirubin is the final product of the heme cleavage pathway and is an effective antioxidant[16]. Abnormal total bilirubin levels have been observed in patients with psychiatric disorders[17]. Tang et al[18] documented that high bilirubin levels were associated with post-stroke depression. Furthermore, Oren et al[19] reported an association between low nocturnal bilirubin levels and seasonal winter depression. These findings suggest an association between depression and bilirubin levels. Nevertheless, no population-based research has looked into the correlation between total bilirubin levels and depression risk in diabetic patients. Therefore, we executed a cross-sectional study to explore the association of total bilirubin with depression risk in adults with diabetes.
MATERIALS AND METHODS
Data sources and collection
The National Health and Nutrition Examination Survey (NHANES) is a program aimed to evaluate the nutritional status and health of children and adults in the United States. The present study utilized data from NHANES 2007-2018. Detailed statistics are available at http://www.cdc.gov/nchs/nhanes/. The National Center for Health Statistics Institutional Review Board/Ethics Review Board gave its approval to the NHANES procedures and also confirmed that all the participants gave their informed consent. Demographic information and health-history data were collected using questionnaires administered by trained personnel. Blood samples were collected and physical examinations were performed through a mobile examination center (MEC). Before being delivered for testing, blood samples were stored in proper conditions. The Laboratory Procedures Manual of NHANES contains comprehensive information on detailed handling instructions and specimen collection. Participants who were under 20 years old or lacked data on total bilirubin, Patient Health Questionnaire-9 (PHQ-9) and other covariates were excluded. Eventually, 4758 adults with diabetes were included in this study (Figure 1).
Figure 1 Study flowchart.
NHANES: National Health and Nutrition Examination Survey; PHQ-9: Patient Health Questionnaire-9.
Evaluation of depressive symptoms
To evaluate depression status, PHQ–9, a nine-item screening tool in NHANES measuring the frequency of different depressive symptoms over the past two weeks, was utilized[20]. Each one of the nine items provides statements on a four-point scale, with 0 = “not at all,” 1 = “a few days,” 2 = “more than half the days,” and 3 = “almost every day,” resulting in overall points ranging from 0 to 27. Here, PHQ-9 score ≥ 10 was regarded to reflect depression, with a sensitivity and specificity of 88%[21,22].
Identification of diabetes
One of the following conditions led to the diagnosis of diabetes: (1) Participants who answered yes to the question “have you ever been told by a doctor that you have diabetes”; (2) hemoglobin A1c (HbA1c) > 6.5%; (3) fasting blood glucose > 7.0 mmol/L; (4) a random or 2-h oral glucose tolerance test blood sugar > 11.1 mmol/L; and (5) using insulin or diabetic medication[23].
Other potential covariates
The study considered sex, race/ethnicity, age, educational level, body mass index (BMI), drinking status, smoking status, ratio of family income to poverty (PIR), hypertension and laboratory parameters including albumin, alkaline phosphatase (ALP), alanine aminotransferase (ALT), aspartate aminotransferase (AST), creatinine, blood urea nitrogen (BUN), lactate dehydrogenase (LDH), gamma-glutamyl transferase (GGT), cholesterol, uric acid and triglycerides as potential covariates. Race was divided into five categories: Mexican American, non-Hispanic black, non-Hispanic white, other Hispanic, and other races. Less than high school, high school equivalent, and college or higher were the different categories for education level. According to standardized protocols, weight in kilograms divided by height in meters squared was used to calculate BMI, which was then categorized into ranges of < 25.0, 25.0 to < 30.0, and ≥ 30.0 kg/m2. Smokers were defined as those who had smoked more than 100 cigarettes in the previous. A respondent was deemed as a drinker if he/she had consumed at least 12 alcoholic drinks annually during his/her life[24]. Hypertension was determined based on self-reported current antihypertensive medication use or a medical diagnosis. PIR was calculated by dividing family income by the poverty threshold in the survey year.
Statistical analysis
All analyses were performed using SPSS 24.0 and the statistical software program R version 4.2.2. To reduce bias and account for potential confounding variables, propensity score matching (PSM) was carried out using the “MatchIt” package of R. Continuous variables are displayed as medians with interquartile ranges or means ± SD, and categorical variables are displayed as frequencies (percentages). Chi-square test, t-test and Kruskal-Wallis test were uesd to analyze the data. The optimal cut-off value for total bilirubin was set as the median. We utilized univariate and multivariate binary logistic regression analyses, with 95% confidence intervals (CI) and odds ratios (OR) generated, to evaluate the link between total bilirubin levels and the risk of depression. We also displayed the unadjusted model 1, model 2 adjusted covariates including age, sex, race, BMI, education level, smoking status, drinking status, hypertension, PIR, ALT, albumin, AST, ALP, BUN, creatinine, GGT, uric acid, LDH, cholesterol and triglycerides, and model 3 after PSM. To investigate whether the relationship between total bilirubin levels and depression risk differed among the subgroups, subgroup analysis was conducted. Stratification factors consisted of age (< 60/≥ 60 years), sex (male/female), hypertension (yes/no), BMI (normal weight/overweight/obese), and race (Mexican American/Non-Hispanic black/ Other Hispanic/Non-Hispanic white/Other races). The heterogeneity of the relationship among the subgroups was assessed using interaction analysis. We used restricted cubic splines with four knots to flexibly model the potential dose-response non-linear association of total bilirubin with depression risk. P value < 0.05 was considered to demonstrate statistically significant differences.
RESULTS
Baseline characteristics
This study included 59842 potential participants from the NHANES (2007–2018), of which 34770 adults (≥ 20 years) completed the MEC screening and interviews. Then those with missing data on PHQ-9 scores and total bilirubin levels (n = 6482) were excluded. We identified 5396 adults with diabetes who met our inclusion criteria and excluded those with missing covariate data (n = 638). The remaining 4758 adults were included in this study. Figure 1 shows the study flowchart of the exclusion criteria. The baseline characteristics of the participants were presented in Table 1, showing that depression in patients with diabetes was correlated with higher levels of BMI, ALP, GGT, cholesterol and triglycerides. Moreover, they were more likely to be females, less educated, smokers, hypertensives, in the lower age group, and with lower levels of PIR, BUN, creatinine, total bilirubin and albumin. There were no differences in drinking status, AST, ALT, LDH and uric acid levels between diabetic patients with and without depression.
Table 1 Baseline characteristics of participants, n (%).
Covariates
Diabetes mellitus (n = 4156)
Diabetic depression (n = 602)
P value
Age (yr)
61.12 ± 13.44
58.74 ± 12.63
< 0.001
Sex
< 0.001
Male
2268 (54.57)
222 (36.88)
Female
1888 (45.43)
380 (63.12)
Race/Ethnicity
< 0.001
Mexican American
713 (17.16)
110 (18.27)
Other Hispanic
436 (10.49)
86 (14.29)
Non-Hispanic white
1549 (37.27)
239 (39.70)
Non-Hispanic black
1015 (24.42)
127 (21.10)
Other races
443 (10.66)
40 (6.64)
BMI (kg/m2), mean ± SD
32.13 ± 7.23
35.07 ± 9.19
< 0.001
Ratio of family income to poverty
2.41 ± 1.55
1.57 ± 1.23
< 0.001
Education level
< 0.001
Did not graduate from high school
1261 (30.34)
264 (43.85)
Graduated from high school
995 (23.94)
132 (21.93)
College education or above
1900 (45.72)
206 (34.22)
Smoking status, ≥ 100 cigarettes during their lifetime
2026 (48.75)
355 (58.97)
< 0.001
Drinking status, ≥ 12 alcohol drinks a year
2874 (69.15)
418 (69.44)
0.889
Hypertension
2665 (64.12)
436 (72.43)
< 0.001
Laboratory metrics
Bilirubin (μmol/L)
10.93 ± 5.35
10.17 ± 5.10
0.001
Albumin (g/L)
41.35 ± 3.46
40.42 ± 3.67
< 0.001
ALT (U/L)
26.39 ± 20.96
30.28 ± 59.21
0.577
AST (U/L)
26.07 ± 15.14
29.71 ± 43.88
0.444
ALP (U/L)
75.59 ± 29.63
81.22 ± 29.41
< 0.001
BUN (mmol/L)
5.82 ± 2.94
5.69 ± 3.10
0.035
Creatinine (μmol/L)
89.69 ± 63.29
89.20 ± 70.31
0.027
GGT (U/L)
36.44 ± 44.20
45.22 ± 62.07
< 0.001
LDH (U/L)
138.58 ± 34.68
141.18 ± 37.72
0.090
Uric acid (μmol/L)
344.87 ± 93.23
340.47 ± 95.02
0.280
Cholesterol (mmol/L)
4.77 ± 1.21
4.90 ± 1.30
0.018
Triglycerides (mmol/L)
2.11 ± 1.99
2.33 ± 2.00
0.009
Association of total bilirubin with depression risk
The median total bilirubin level in adults with diabetes was 10.26 μmol/L. Participants were divided into two groups according to their serum total bilirubin levels (low, ≤ 10.26 μmol/L; and high, > 10.26 μmol/L). In univariate logistic regression analysis, a low level of total bilirubin was associated with an increased risk of depression in patients with diabetes (OR = 1.531, 95%CI: 1.276–1.836, P < 0.001, Table 2). After adjusting for sex, age, BMI, race, education level, smoking status, drinking status, hypertension, PIR, albumin, AST, ALP, ALT, BUN, creatinine, GGT, Uric acid, LDH, cholesterol and triglycerides levels, the association remained significant (OR = 1.230, 95%CI: 1.006-1.503, P=0.043, Table 2).
Table 2 Association between total bilirubin and diabetic depression in different models.
As shown in Figure 2, restricted cubic splines visualized a non-linear association between total bilirubin levels and depression risk in adults with diabetes (P for non-linearity < 0.05). Within the lower range of total bilirubin levels, an inverted U-shaped association between total bilirubin levels and depression risk was observed. The risk of depression heightened with the increasing of total bilirubin levels, reaching the highest point at 6.81 μmol/L and decreasing thereafter. The plot also depicted a clear reduction in depression risk with higher levels of total bilirubin, reaching the lowest risk at 17.40 μmol/L before plateauing.
Figure 2 Restricted cubic spline regression curve of the odds ratios (OR) of total bilirubin with depression in adults with diabetes.
PSM analysis
The covariates (sex, age, race, education level, BMI, drinking status, smoking status, hypertension, PIR, albumin, AST, ALP, ALT, creatinine, BUN, LDH, GGT, uric acid, cholesterol and triglycerides) were balanced using 1:1 PSM, resulting in 1583 pairs of patients (Table 3). After PSM, the association of total bilirubin levels with the risk of depression remained consistent (OR = 1.303, 95%CI: 1.034-1.641, P = 0.025).
Table 3 Characteristics of participants before and after propensity score matching analysis, n (%).
As multiple variables were associated with depression, we conducted a stratified subgroup analysis with distinct characteristics. The results of the subgroup analysis are exhibited in Figure 3. The interaction test revealed that the association between total bilirubin and depression risk was not significantly different across stratification, illustrating that there was no significant dependence of age, BMI, sex, race or hypertension on this association (all P for interaction > 0.05).
Figure 3 Subgroup analysis for the association between total bilirubin and depression risk in adults with diabetes.
DISCUSSION
As far as we are aware, this study is the first to reveal the close correlation between total bilirubin levels and depression risk in patients with diabetes. Our findings demonstrated that diabetic patients with depression had lower total bilirubin levels than those without depression. In addition, low levels of total bilirubin were associated with an increased risk of depression in adults with diabetes.
Depression is defined by continuous sadness and a loss of pleasure or interest in formerly rewarding or enjoyable activities[25]. Meta-analyses have indicated that diabetes is a risk factor for depression[26], and people with diabetes are twice as likely to experience depression as people without the condition[27]. Depression in combination with other chronic diseases has a significant economic impact. According to a study using data from the Medical Expenditure Panel Survey 2004–2011, a nationally representative assessment of healthcare expenditures, individuals with diabetes and no depression incurred an average medical cost of $10,016 (95%CI: 9589–10442), whereas those with symptomatic depression had an average medical cost of $20105 (95%CI: 18103–22106)[28]. This co-occurrence imposed a strain on more than just the healthcare system. There was evidence that depression interacted with physical sickness to increase the severity of other disorders, and vice versa[29]. As a result, early evaluation of depression in high-risk populations, like those with diabetes, could help patients avoid experiencing both physical and psychological harm. For all these reasons, it appears crucial for the delivery of healthcare services to manage the co-occurrence of depression and diabetes. Despite the increasing awareness of the incidence and significance of depression, there is a dearth of research on factors that can identify individuals with diabetes who are at high risk of developing depression, and how to use this understanding to treat and prevent populations at risk.
The association of bilirubin levels with depression in individuals without diabetes has been investigated. For example, low nocturnal bilirubin levels were found to be related to seasonal winter depression[19]. Peng et al[30] found that lower serum bilirubin concentrations were correlated with major depressive disorder in the Chinese Han population. Interestingly, patients with psychiatric problems were observed to have higher concentrations of bilirubin oxidative metabolites (biopyrrins) in their urine, and the Hamilton Depression Rating Scale score was found to be correlated with the biopyrrin levels[31]. According to our study, diabetic patients with depression had lower levels of total bilirubin than those without depression. Our research also revealed that in adults with diabetes, low total bilirubin levels were associated with an increased risk of depression. After matching for possible confounders, our results further confirmed the above association of total bilirubin with depression risk in diabetic adults. However, the precise mechanism underlying the correlation between total bilirubin levels and diabetes-related depression remains unclear. Oxidative stress, derived from the overproduction of reactive oxygen species, can cause cell damage and various disorders, such as neurodegeneration, major depressive disorder and cancer[32-35]. The brain is more susceptible to oxidative stress owing to its high oxygen consumption and weak antioxidant capacity[32]. Michel et al[36] observed oxidative stress process in the frontal cortex of patients with recurrent depressive disorders. These results highlight the significant possibility that oxidative stress plays a significant role in the development of depression. Bilirubin, an efficient antioxidant, is associated with oxidation resistance, immunosuppression, and anti-inflammation in many disorders[30]. In patients with depression, bilirubin as an endogenous antioxidant can be destroyed by excessive oxidative stress, leading to excessive consumption of bilirubin, which may explain why diabetic adults with depression have decreased bilirubin levels.
The evaluation of depression is primarily based on the physician’s clinical experience, the patient’s self-assessment, and symptoms in clinical practice. When primary care providers and other clinicians lack specialized training or sufficient time for a thorough clinical assessment, the diagnosis of depression can be quite challenging. A combination of common biochemical indicators may provide objective and useful information for evaluating and predicting depression.
Our study came with several advantages. First, the sample size was relatively large to reveal a correlation between total bilirubin levels and depression risk in patients with diabetes. In addition, it emphasized the potential value of using total bilirubin level to identify individuals with diabetes who are at a high risk of developing depression. Moreover, we adjusted for confounders that might affect the association between total bilirubin levels and depression risk by analyzing the detailed covariate data. Notwithstanding, this study had certain limitations. First, causality could not be proven because of the cross-sectional study design. Prospective investigations are necessary to confirm this hypothesis. Second, serum total bilirubin levels may be affected by the health status of the patient at the time of blood collection. Hence, it can not be completely ruled out that the total bilirubin level might be influenced by certain disease-related information that was not obtained during the entire investigation process. Finally, some research covariates were dependent on self-reporting, which might lead to recall problems or incorrect interpretations of questions.
CONCLUSION
To sum up, we found that in adults with diabetes, those with lower levels of total bilirubin were more likely to exhibit depressive symptoms. Hence, serum total bilirubin levels may be used as an additional indicator to assess the risk of depression in patients with diabetes. Further research is necessary to fully understand the role of total bilirubin in diabetic depression.
ACKNOWLEDGEMENTS
We gratefully thank all the study participants for their contribution.
Footnotes
Provenance and peer review: Unsolicited article; Externally peer reviewed.
Peer-review model: Single blind
Specialty type: Psychology, clinical
Country of origin: China
Peer-review report’s classification
Scientific Quality: Grade D
Novelty: Grade B
Creativity or Innovation: Grade B
Scientific Significance: Grade C
P-Reviewer: Liang Y, China S-Editor: Liu JH L-Editor: A P-Editor: Zhang XD
Frankenthal D, Keinan-Boker L, Bromberg M. Rise in depression and anxiety during the COVID-19 pandemic: results from a cross-sectional national survey.Psychol Health Med. 2023;28:139-147.
[PubMed] [DOI][Cited in This Article: ][Reference Citation Analysis (0)]
Bădescu SV, Tătaru C, Kobylinska L, Georgescu EL, Zahiu DM, Zăgrean AM, Zăgrean L. The association between Diabetes mellitus and Depression.J Med Life. 2016;9:120-125.
[PubMed] [DOI][Cited in This Article: ]
Chand SP, Arif H.
Depression. 2023 Jul 17. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2024 Jan-.
[PubMed] [DOI][Cited in This Article: ]
Mellon SH, Wolkowitz OM, Schonemann MD, Epel ES, Rosser R, Burke HB, Mahan L, Reus VI, Stamatiou D, Liew CC, Cole SW. Alterations in leukocyte transcriptional control pathway activity associated with major depressive disorder and antidepressant treatment.Transl Psychiatry. 2016;6:e821.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 24][Cited by in F6Publishing: 24][Article Influence: 3.0][Reference Citation Analysis (0)]
