Blog

Introduction

Gout is an autoinflammatory metabolic disease characterized by hyperuricemia and monosodium urate crystal deposition. It is the most common inflammatory arthritis, with a significant increase in prevalence in several countries, such as the United States.¹ Epidemiological data show that the global prevalence of gout ranges from 1% to 6.8% in different regions, imposing a substantial burden on patient quality of life and healthcare systems.² Despite deeper understanding of the pathogenesis, diagnosis, and treatment strategies for gout, the identification and management of refractory gout remain major challenges, with refractory gout often characterized by frequent flares, persistent hyperuricemia despite standard therapy, and significant tophi burden.³

Refractory gout is increasingly recognized in clinical practice. These patients often present with frequent polyarticular flare-ups, large tophi formation, joint damage, and musculoskeletal dysfunction, which significantly reduce their quality of life.⁴ Despite receiving standard treatment, these patients still face considerable difficulties in controlling the disease, whether in achieving target uric acid levels or alleviating clinical symptoms. The management of refractory gout not only requires attention to the disease itself but also to its common comorbidities, especially metabolic syndrome.⁵ Metabolic syndrome, a cluster of cardiovascular risk factors, including abdominal obesity, hypertension, dyslipidemia, and glucose metabolism disorders, has become a major cause of morbidity and mortality in both developed and developing countries.⁶ In China, the prevalence of metabolic syndrome has been increasing rapidly (24.9% in urban areas, 19.2% in rural areas, reaching 39.0% in some rural regions),⁷ significantly correlating with gout epidemic trends: over the past 30 years, the age-standardized prevalence of gout has increased by an average of 0.9% annually, with adult prevalence rising to 1.23% in 2019.⁸ Notably, individuals with metabolic syndrome have a two-fold higher risk of developing gout compared to the general population.Epidemiological studies have shown a close link between gout and metabolic syndrome, with the prevalence of metabolic syndrome being significantly higher in gout patients compared to the general population.⁹ The prevalence of metabolic syndrome among gout patients in different regions ranges from 30.1% to 34%.^10,11 Notably, Schlesinger et al¹² reported that gout patients with more severe metabolic syndrome have a significantly increased risk of death.Hyperuricemia is considered a key factor linking gout and metabolic syndrome. Uric acid plays a dual role—functioning as an antioxidant while also promoting inflammation and metabolic disturbances.¹³ At the vascular level, uric acid affects blood vessel function and structure, contributing to hypertension and cardiovascular risk.^14,15 There also exists a bidirectional relationship between uric acid and insulin function. Insulin resistance can decrease uric acid excretion,¹⁶ while elevated uric acid levels may worsen insulin sensitivity, creating a self-reinforcing cycle that impacts both conditions.¹⁷

However, there is limited research on the relationship between refractory gout and metabolic syndrome, particularly in Asian populations. Although studies have explored the association between gout and metabolic syndrome, a comprehensive analysis of the metabolic disorder characteristics in refractory gout patients has not been fully conducted, particularly regarding the nonlinear relationship between metabolic parameters and clinical features in refractory gout patients. The lack of evidence in this area limits clinicians’ ability to perform risk stratification and make individualized treatment decisions for patients with refractory gout.

Therefore, this study aims to explore the association between refractory gout and metabolic syndrome, using propensity score matching to control for confounding factors, and to analyze the nonlinear relationships between metabolic abnormalities and clinical parameters in refractory gout patients. As an important effort to systematically investigate this association in the Chinese population, we hypothesize that the prevalence of metabolic syndrome in refractory gout patients is significantly higher than in non-refractory gout patients, and that the metabolic burden has complex nonlinear relationships with clinical parameters. Clarifying these associations will help identify high-risk subgroups with increased metabolic syndrome risk, provide evidence-based recommendations for the comprehensive management of refractory gout patients, and ultimately improve patient prognosis and quality of life.

Methods

Study Design and Population

This study was a single-center retrospective cohort study that included gout patients who visited the Department of Rheumatology and Immunology at the First Affiliated Hospital of Tianjin University of Traditional Chinese Medicine from January 1, 2014, to December 31, 2024. All patients were diagnosed according to the 2015 American College of Rheumatology/European League Against Rheumatism (ACR/EULAR) gout classification criteria.¹⁸ The inclusion criteria were: (1) age ≥18 and ≤80 years; (2) meeting the ACR/EULAR gout classification criteria; (3) complete clinical data and follow-up records. The exclusion criteria included: (1) secondary gout (eg, gout caused by conditions such as renal failure, hematopoietic disorders like myeloproliferative neoplasms, or the use of specific medications like diuretics); (2) concomitant autoimmune diseases; (3) severe liver or renal dysfunction; (4) malignancy; (5) incomplete data or lost to follow-up. All participants were of Asian ethnicity. This study was approved by the Ethics Committee of the First Affiliated Hospital of Tianjin University of Traditional Chinese Medicine (approval number: TYLL2025[Zi]007). This study was conducted in accordance with the principles of the Declaration of Helsinki. All patient data were anonymized to protect patient privacy and data confidentiality. As this was a retrospective analysis, the ethics committee waived the requirement for informed consent from patients. As a retrospective study, all patients who met the inclusion and exclusion criteria within the specified timeframe were included, and no formal sample size calculation was performed beforehand. All data for this study were sourced from the hospital’s Electronic Medical Record system, which served as the primary data collection instrument. To ensure the accuracy and reliability (validity) of the data, a standardized data abstraction form was utilized. The data were independently extracted by two trained researchers. Any discrepancies identified during this process were resolved by discussion and consensus with a senior researcher. The collected data primarily represent the baseline characteristics at the time of the patient’s comprehensive assessment at our center. For variables dependent on a time span, such as the frequency of gout flares, the calculation was based on the 12-month period preceding the assessment, as documented in the EMR. If essential baseline information was missing from the initial record, subsequent clinical encounter notes within the EMR were reviewed to the supplement data. Comprehensive baseline clinical data were collected, including demographic characteristics (age, sex), clinical features (gout duration, flare frequency, tophi), laboratory tests (uric acid, blood glucose, lipids, inflammatory markers, liver and kidney function), and comorbidities.

Definition of Refractory Gout

According to the International Refractory Gout Working Group’s definition,¹⁹ refractory gout must meet at least two of the following criteria: (1) despite receiving standard urate-lowering treatment (such as allopurinol or febuxostat at maximum tolerated doses), serum uric acid remains ≥6.0 mg/dL (357 μmol/L) or cannot reach target levels; (2) ≥2 gout attacks per year; (3) at least one clinically confirmed tophus; (4) chronic gouty arthritis, manifested as persistent joint swelling or functional limitation; (5) symptoms (such as pain or joint function) remain poorly controlled despite standard treatment (including urate-lowering and anti-inflammatory therapy). This definition follows the 2020 American College of Rheumatology (ACR) gout guidelines to ensure specificity and improve clinical operability.

Definition of Metabolic Syndrome

Metabolic syndrome was defined according to the National Cholesterol Education Program (NCEP) Adult Treatment Panel III (ATP III) criteria,²⁰ requiring at least three of the following five criteria: (1) increased waist circumference (Asian standards: males ≥90 cm, females ≥80 cm, based on the International Diabetes Federation [IDF] standards); (2) triglycerides ≥1.7 mmol/L or on lipid-lowering treatment (such as statins); (3) low HDL cholesterol (males <1.03 mmol/L, females <1.30 mmol/L) or receiving related treatment; (4) blood pressure ≥130/85 mmHg or on antihypertensive medication; (5) fasting blood glucose ≥5.6 mmol/L, on antidiabetic treatment (such as oral hypoglycemic agents or insulin), or diagnosed with type 2 diabetes.

Propensity Score Matching

To control for confounding bias, propensity score matching (PSM) was used in this study. A logistic regression model was applied to calculate the propensity score for each patient, with matching factors including baseline characteristics that might influence the relationship between refractory gout and metabolic syndrome, such as sex, age, uric acid level, diabetes, hypertension, and hyperlipidemia. Variables were selected based on previous literature and clinical relevance. A 1:1 nearest-neighbor matching method was used, matching each refractory gout patient to a non-refractory gout patient with the closest propensity score, allowing a maximum propensity score difference of 0.02 (caliper). The matching effect was assessed using standardized mean differences (SMD), with SMD <0.1 considered a good match. After matching, the data were further analyzed using logistic regression to evaluate the association between refractory gout and metabolic syndrome.

Nonlinear Relationship Analysis

To assess potential nonlinear relationships between clinical parameters and metabolic syndrome or refractory gout, restricted cubic spline models were used. The relationships between age and ESR with metabolic syndrome, as well as ESR, CRP, SBP, DBP, blood glucose, and TG with refractory gout, were modeled separately. The model included four knots located at the 5th, 35th, 65th, and 95th percentiles of the variable distribution. All nonlinear models were adjusted for potential confounding factors, including age, sex, and baseline metabolic risk factors. Results were expressed as adjusted odds ratios (ORs) with 95% confidence intervals (CIs), with a focus on identifying reference points for each clinical parameter and plotting nonlinear relationship curves to show how the OR changes with the parameters.

Statistical Analysis

Statistical analysis was performed using SPSS 25.0 software and R 4.1.0. Normally distributed continuous variables were expressed as mean ± standard deviation, non-normally distributed continuous variables as median (interquartile range), and categorical variables as frequency (percentage). t-tests or Mann–Whitney U-tests were used to compare continuous variable differences between groups, and chi-square tests or Fisher’s exact test were used to compare categorical variable differences. Multivariable logistic regression analysis was used to explore the independent association between refractory gout and metabolic syndrome, with all potential confounders included in the model. Collinearity was assessed using the variance inflation factor (VIF), with VIF >5 indicating collinearity. Subgroup analyses were performed to compare the differences in the association between metabolic syndrome and refractory gout based on flare frequency (<3 times/year vs ≥3 times/year), tophus presence, age (<60 vs ≥60), sex, uric acid level (<480 μmol/L vs ≥480 μmol/L), and disease duration (<3 years vs ≥3 years), with interaction tests used to evaluate statistical significance. All tests were two-sided, and P < 0.05 was considered statistically significant.

Results

Propensity Score Distribution Before and After Matching

Before propensity score matching, there was significant overlap between the refractory gout group and the non-refractory gout group in the 0.25–0.6 range. After matching, the distributions of the two groups nearly completely overlapped, indicating that matching effectively reduced baseline differences. A 1:1 nearest-neighbor matching with a caliper of 0.2 resulted in 1,353 matched pairs. After matching, the two groups achieved good balance in terms of gender (SMD = 0.011) and diabetes (SMD = 0.011), while slight imbalance remained in age (SMD = 0.045), uric acid levels (SMD = 0.151), and hyperlipidemia (SMD = 0.044), which were adjusted as covariates in subsequent analyses (Figure 1).

Figure 1 Distribution of propensity scores before and after matching. The left panel shows the density plot of propensity scores before matching, indicating a baseline imbalance between the refractory gout group (blue-green) and the non-refractory gout group (pink). The right panel shows the distribution after 1:1 nearest-neighbor matching, demonstrating a significant improvement in balance as the distributions of the two groups nearly overlap.

Baseline Characteristics

A total of 4111 gout patients were included in this study, comprising 1972 with refractory gout and 2139 with non-refractory gout. Prior to propensity score matching, the two groups exhibited significant differences across multiple baseline characteristics (Table 1). Specifically, patients in the refractory gout group were younger, had a higher proportion of males, and a significantly higher prevalence of metabolic syndrome. Furthermore, this group demonstrated poorer metabolic profiles, including blood lipids and glucose, elevated levels of inflammatory markers such as C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR), a greater burden of tophi, and a higher frequency of acute flares (all P < 0.05, detailed in Table 1).Following propensity score matching, balance was achieved between the groups in terms of sex (P=0.765), hypertension (P=0.906), diabetes (P=0.767), and hyperlipidemia (P=0.253). However, the refractory gout group continued to show a significantly higher prevalence of metabolic syndrome (52.5% vs 15.8%, P<0.001) and associated metabolic abnormalities. Detailed comparisons are presented in Table 2.

Table 1 Comparison of Baseline Characteristics Before Propensity Score Matching Between Refractory Gout and Non-Refractory Gout Patients

Table 2 Comparison of Baseline Characteristics After Propensity Score Matching Between Refractory Gout and Non-Refractory Gout Patients

Univariate and Multivariate Analysis of Refractory Gout

Univariable logistic regression analysis showed that metabolic syndrome (OR=5.877, P<0.001), uric acid level (OR=1.001, P<0.001), ESR (OR=1.042, P<0.001), and the presence of tophi (OR=9.025, P<0.001), among other factors, were significantly associated with refractory gout.In the multivariable logistic regression analysis (Table 3), after adjusting for potential confounders, metabolic syndrome remained the strongest independent factor associated with refractory gout (adjusted OR=9.689, 95% CI: 5.727–16.392, P<0.001). Other significant factors included elevated uric acid levels (adjusted OR=1.005, P<0.001), elevated ESR (adjusted OR=1.039, P<0.001), the presence of tophi (adjusted OR=7.164, P<0.001), and decreased HDL levels (adjusted OR=0.272, P=0.009). Age was negatively associated with refractory gout (adjusted OR=0.982, P=0.013), suggesting that younger patients may face a higher risk of developing the condition.

Table 3 Univariate and Multivariate Logistic Regression Analysis of Factors Associated with Refractory Gout

Subgroup Analysis of the Association Between Metabolic Syndrome and Its Components with Refractory Gout Based on Different Factors

To explore how the effects of various factors differed across populations, we conducted a subgroup analysis (Table 4). The results showed that when stratified by age, presence of tophi, flare frequency, uric acid level, and disease duration, the strength of the association between metabolic syndrome and refractory gout did not exhibit a significant interaction (all P-interaction > 0.05). This suggests that metabolic syndrome is likely a common and independent risk factor across these subgroups.However, a significant interaction was observed in the association between blood glucose levels and refractory gout (P-interaction < 0.001). Specifically, this association was more pronounced in patients with tophi (OR=2.52 vs 1.87) and in those with a disease duration of ≥3 years (OR=2.42 vs 2.00), suggesting that a state of long-term hyperglycemia may play a more significant role in the progression to refractory gout.

Table 4 Subgroup Analysis of the Association Between Metabolic Syndrome and Its Components with Refractory Gout Based on Different Factors

Nonlinear Relationships Between Clinical Parameters and Metabolic Syndrome and Refractory Gout

To investigate potential complex patterns between various clinical indicators and disease risk, we employed restricted cubic spline models (Figure 2). The analysis revealed several significant nonlinear relationships.Regarding the risk of metabolic syndrome, age exhibited an inverted U-shaped trend, with the risk peaking at approximately 58 years. Regarding the risk of refractory gout, multiple indicators demonstrated nonlinear characteristics: ESR showed a bell-shaped relationship, with the highest risk around 25 mm/h; CRP had a positive nonlinear association (reference point at 8 mg/L); and systolic blood pressure (SBP) displayed a U-shaped relationship with risk, with a nadir around 138 mmHg, suggesting that both high and low blood pressure may increase risk. These findings suggest that the clinical interpretation of these markers should extend beyond assessing whether they fall within a “normal range” to considering their specific risk-associated intervals.

Figure 2 Nonlinear relationships between clinical parameters and the risk of metabolic syndrome or refractory gout. (A and B) show the relationship of age (A) and ESR (B) with metabolic syndrome risk. (C-H) display the relationship of ESR (C), CRP (D), SBP (E), DBP (F), blood glucose (G), and TG (H) with refractory gout risk. Solid red lines represent the adjusted odds ratios (ORs) and the shaded areas represent the 95% confidence intervals (CIs). The horizontal dashed line at OR=1.0 indicates the reference for no increased risk. The vertical dashed line indicates the reference point for the corresponding clinical parameter. All models were adjusted for potential confounders including age, sex, and baseline metabolic risk factors. A P for non-linearity < 0.05 was considered statistically significant. Abbreviations: ESR, Erythrocyte Sedimentation Rate; CRP, C-Reactive Protein; SBP, Systolic Blood Pressure; DBP, Diastolic Blood Pressure; TG, Triglycerides.

Discussion

This study systematically assessed the association between refractory gout and metabolic syndrome, along with their clinical features, using propensity score matching analysis. Our results indicate that the prevalence of metabolic syndrome in patients with refractory gout was significantly higher than in those with non-refractory gout (52.5% vs 15.8%, P < 0.001). Multivariate analysis showed that metabolic syndrome was the strongest independent factor associated with refractory gout (adjusted OR = 9.689, 95% CI: 5.727–16.392, P < 0.001). Furthermore, this study is the first to systematically explore the nonlinear relationships between clinical parameters and metabolic syndrome as well as refractory gout. We found that age showed a rising and then falling trend in relation to metabolic syndrome, with the risk peaking around 58 years; ESR showed a bell-shaped relationship with the risk of metabolic syndrome. Subgroup analysis revealed that the association between blood glucose and refractory gout was more significant in patients with tophi (OR = 2.52 vs 1.87) and those with a disease duration ≥3 years (OR = 2.42 vs 2.00) (P-interaction < 0.001).

Unlike general gout, refractory gout has more unique pathophysiological mechanisms, which lead to a stronger association with metabolic syndrome. Refractory gout patients often have persistent hyperuricemia, widespread urate crystal deposition, and a chronic inflammatory state.¹⁹ This chronic inflammation leads to the sustained release of pro-inflammatory cytokines, such as IL-1β, TNF-α, and IL-6, which not only participate in the inflammatory cascade of gout attacks but also directly contribute to insulin resistance, endothelial dysfunction, and lipid metabolism disturbances.²¹ Additionally, refractory gout patients typically experience more frequent arthritis flare-ups and greater tophi formation, reflecting higher uric acid burden and more severe crystal deposition, which further exacerbates metabolic disturbances. Vazquez-Mellado et al¹¹ reported a prevalence of metabolic syndrome of 82% in Mexican male gout patients. Yoo et al¹⁰ found a prevalence of 44% in Korean male gout patients, compared to 5% in the general Korean population, with an association to the severity of refractory gout. Notably, our study is the first to systematically assess the association between refractory gout and metabolic syndrome in a Chinese population, using a more rigorous study design. Compared to the 54.6% prevalence reported by Doualla-Bija et al²² in sub-Saharan African populations, the prevalence of metabolic syndrome in our study (52.5%) in Chinese refractory gout patients was similar, which may reflect the universality of this association, despite racial and regional differences.

The molecular mechanisms underlying the association between refractory gout and metabolic syndrome may involve multiple complex pathways. First, persistent hyperuricemia itself may be an important driver of metabolic syndrome. Research shows that uric acid is not only an antioxidant but also an active molecule that promotes inflammation and metabolic disturbances.²³ In vascular endothelial cells, uric acid can inhibit the production of nitric oxide, leading to vascular remodeling.¹⁴ Uric acid also promotes the proliferation and migration of vascular smooth muscle cells, which are important pathological bases for hypertension and cardiovascular diseases.¹⁵ Second, there is a bidirectional relationship between refractory gout and insulin resistance. On the one hand, insulin resistance enhances renal tubular sodium reabsorption, reducing uric acid excretion and leading to hyperuricemia;¹⁶ on the other hand, persistently high levels of uric acid can suppress β-cell proliferation and promote insulin resistance, creating a vicious cycle. Our study found that refractory gout patients had higher fasting blood glucose levels and lower HDL-C levels, supporting this mechanism. Takir et al¹⁷ demonstrated in clinical trials that urate-lowering therapy can significantly improve insulin sensitivity and systemic inflammatory markers, even in asymptomatic hyperuricemia patients.

Our study is the first to reveal the nonlinear relationships between multiple clinical parameters and refractory gout as well as metabolic syndrome. Age showed a rising and then falling trend in relation to metabolic syndrome, with the risk peaking around 58 years, which may reflect the dynamic changes of risk factors over the disease’s natural history. Younger and middle-aged patients face lifestyle-related factors (such as dietary habits, work stress, and reduced physical activity), while older patients may show a decrease in risk due to “survivor effect” or pharmacological interventions. ESR demonstrated a bell-shaped relationship with both metabolic syndrome and refractory gout, with a reference point at 25, suggesting that a moderate inflammatory state might have the greatest impact on metabolic disturbances, whereas excessively high ESR may reflect the presence of other serious diseases, masking the direct association with metabolic disorders. CRP showed a positive nonlinear relationship with refractory gout (reference point = 8), which is consistent with the findings of this study, highlighting the key role of chronic low-grade inflammation in metabolic disturbances.²⁴

Notably, we observed a U-shaped relationship between SBP and refractory gout (reference point = 138), which differs from previous studies and may reflect unique blood pressure regulation mechanisms in gout patients. Both low and high SBP may increase the risk of refractory gout through different pathways—hypotension may be associated with reduced renal perfusion and impaired uric acid clearance, while hypertension exacerbates urate crystal deposition through vascular injury and inflammatory responses.²⁵ Blood glucose levels were significantly positively associated with refractory gout (reference point = 5.5), particularly in patients with tophi and those with a disease duration ≥3 years, suggesting that long-term hyperglycemia may promote the refractory progression of gout through multiple pathways, including exacerbating oxidative stress, promoting crystal deposition, and affecting uric acid excretion.²⁶ In contrast, we found that TG showed a negative nonlinear relationship with refractory gout (reference point = 1.7), with a stronger association in patients with uric acid <480 μmol/L. This unexpected finding may indicate the complex regulatory role of TG in gout depending on different uric acid levels, possibly influenced by the use of TG-modulating drugs, which warrants further investigation.

In subgroup analysis, we observed a significant interaction in the association between blood glucose and refractory gout, with stronger associations in patients with tophi and those with a disease duration ≥3 years. This finding suggests that tophi formation may reflect a greater disease burden and longer disease exposure time, thereby enhancing the association between metabolic disturbances and refractory gout.²⁷ Additionally, we did not observe significant differences in the strength of the association between metabolic syndrome and refractory gout in different age, sex, and disease characteristic subgroups, indicating that metabolic syndrome may be a relatively independent risk factor whose impact on refractory gout is not significantly modulated by these basic characteristics.

Our study has several important implications for clinical practice. First, refractory gout patients should routinely be screened for metabolic syndrome, especially early in the disease course (around 58 years). Reference points for clinical assessment, such as ESR around 25, CRP around 8, SBP around 138, blood glucose around 5.5, and TG around 1.7, may serve as important indicators. Second, particular attention should be paid to blood glucose control in patients with tophi and a disease duration ≥3 years, as these patients may be more sensitive to the adverse effects of hyperglycemia. Third, treatment strategies should take into account the nonlinear relationships between clinical parameters and disease risk, avoiding the oversimplified approach of “the lower the better” or “within standard range” for treatment goals, and instead determining optimal treatment targets based on nonlinear risk curves.²⁸ Lastly, considering the increased cardiovascular risk in these patients, comprehensive management strategies should be part of routine care.

The strengths of this study lie in its large sample size and rigorous study design. By using propensity score matching, we effectively controlled for potential confounders. The application of restricted cubic spline models allowed us to explore the nonlinear relationships between clinical parameters, refractory gout, and metabolic syndrome, providing more detailed risk assessment. Additionally, systematic subgroup analysis helped identify high-risk populations, providing evidence for clinical decision-making. This study has several limitations that should be acknowledged. First, as a retrospective cross-sectional study, it cannot establish causality between metabolic syndrome and refractory gout. Second, the single-center design may limit the external validity of our findings, particularly given the diversity of lifestyle and dietary habits across different regions of China. Third, some potential confounders were not fully assessed, including detailed dietary patterns, physical activity levels, medication adherence, and genetic factors. Furthermore, reliance on self-reported disease history may have introduced recall bias. Finally, as this study spanned a decade, we could not account for potential minor variations in laboratory assay standards or equipment over this period, which is a common limitation in long-term retrospective studies.Despite these limitations, our findings open several avenues for future research. Prospective cohort studies are needed to explore the temporal relationship and causality between refractory gout and metabolic syndrome. Randomized controlled trials should evaluate the impact of different urate-lowering therapy intensities and types on the components of metabolic syndrome. Furthermore, exploring novel biomarkers, such as inflammatory cytokines, adiponectin, and microRNAs, is crucial for the early identification of high-risk patients. Investigating the potential benefits of anti-inflammatory treatments, like IL-1 inhibitors, on metabolic parameters in this population is also a promising direction. Ultimately, these efforts should aim to develop personalized treatment strategies that adjust clinical management based on the specific features and nonlinear risk patterns identified in our study.

Conclusion

This study shows that the prevalence of metabolic syndrome is significantly higher in patients with refractory gout, and the metabolic burden has complex nonlinear relationships with clinical parameters. The nonlinear associations between blood glucose levels, ESR, CRP, SBP, and TG with refractory gout provide new perspectives for clinical risk assessment and management. As an autoinflammatory metabolic disease, refractory gout should not be simply viewed as arthritis but recognized as a systemic disease. Multidisciplinary collaborative management is crucial to reduce the risk of metabolic complications in these patients. Future research needs to further clarify the causal relationship between hyperuricemia and metabolic abnormalities, as well as the long-term impact of urate-lowering therapy on improving metabolic markers, providing stronger evidence for the comprehensive management of refractory gout patients.

Ethics Statement

This study was approved by the Ethics Committee of the First Affiliated Hospital of Tianjin University of Traditional Chinese Medicine (Ethical approval number: TYLL2025[Zi]007). This study was conducted in accordance with the principles of the Declaration of Helsinki. All patient data were anonymized to protect patient privacy and data confidentiality. As this study was a retrospective analysis, the ethics committee waived the requirement for informed consent from the patients.

Funding

This study was supported by the following projects: Major Difficult Diseases Integrated Traditional and Western Medicine Clinical Collaboration Project-01 Direct Funding-2024 Traditional Chinese Medicine Career Inheritance and Development (Batch 2), Project Number: 20240905; Study on the Syndromes of Rheumatic Diseases (RA, SS, Gout) and Expert Consensus Development, Project Number: 20240204011; Research on Integrated Traditional and Western Medicine Diagnosis and Treatment Plans for Major Difficult Diseases (Guidelines for the Integrated Diagnosis and Treatment of Sjogren’s Syndrome), Project Number: 2023382; Major Scientific and Technological Achievements and Popularization of the “Toxicity-Based Treatment” in Gout and Its Important Complications, Project Number: 22KPXMRC00180; National Famous Old TCM Experts’ Inheritance Studio, Project Number: 975022-2024; Efficacy and Mechanism of Yin-Nourishing and Detoxifying Chinese Medicine in the Treatment of Sjogren’s Syndrome, Project Number: GZY-KJS-2024-05; Inheritance and Innovation of Traditional Chinese Medicine “Hundred, Thousand, and Ten Thousand Talents” Project (Qihuang Project) (TCM People’s Education Letter [2018] No. 12); National TCM Management Bureau Key Subject Capacity Building Project in TCM Bianbing Studies (2018ZDXK001); Chinese Ethnic Medicine Association Research Project (2020MZ319-350601).

Disclosure

The authors report no conflicts of interest in this work.

References

1. Chen-Xu M, Yokose C, Rai SK, et al. Contemporary prevalence of gout and hyperuricemia in the United States and decadal trends: the National Health and Nutrition Examination Survey, 2007–2016. Arthritis Rheumatol. 2019;71(6):991–999. doi:10.1002/art.40807

2. Dehlin M, Jacobsson L, Roddy E. Global epidemiology of gout: prevalence, incidence, treatment patterns and risk factors. Nat Rev Rheumatol. 2020;16(7):380–390. doi:10.1038/s41584-020-0441-1

3. Fels E, Sundy JS. Refractory gout: what is it and what to do about it? Curr Opin Rheumatol. 2008;20(2):198–202. doi:10.1097/BOR.0b013e3282f4eff5

4. Becker MA, Schumacher HR, Benjamin KL, et al. Quality of life and disability in patients with treatment-failure gout. J Rheumatol. 2009;36(5):1041–1048. doi:10.3899/jrheum.071229

5. Pascual E, Andrés M, Vázquez-Mellado J, et al. Severe gout: strategies and innovations for effective management. Joint Bone Spine. 2017;84(5):541–546. doi:10.1016/j.jbspin.2016.10.004

6. Saklayen MG. The global epidemic of the metabolic syndrome. Curr Hypertens Rep. 2018;20(2):12. doi:10.1007/s11906-018-0812-z

7. Li R, Li W, Lun Z, et al. Prevalence of metabolic syndrome in Mainland China: a meta-analysis of published studies. BMC Public Health. 2016;16:296. doi:10.1186/s12889-016-2870-y

8. Zhu B, Wang Y, Zhou W, et al. Trend dynamics of gout prevalence among the Chinese population, 1990–2019: a joinpoint and age-period-cohort analysis. Front Public Health. 2022;10:1008598. doi:10.3389/fpubh.2022.1008598

9. Choi HK, Ford ES, Li C, et al. Prevalence of the metabolic syndrome in patients with gout: the Third National Health and Nutrition Examination Survey. Arthritis Rheum. 2007;57(1):109–115. doi:10.1002/art.22466

10. Yoo HG, Lee SI, Chae HJ, et al. Prevalence of insulin resistance and metabolic syndrome in patients with gouty arthritis. Rheumatol Int. 2011;31(4):485–491. doi:10.1007/s00296-009-1304-x

11. Vázquez-Mellado J, García CG, Vázquez SG, et al. Metabolic syndrome and ischemic heart disease in gout. J Clin Rheumatol. 2004;10(3):105–109. doi:10.1097/01.rhu.0000129082.42094.fc

12. Schlesinger N, Elsaid MI, Rustgi VK. The relationship between metabolic syndrome severity and the risk of mortality in gout patients: a population-based study. Clin Exp Rheumatol. 2022;40(3):631–633. doi:10.55563/clinexprheumatol/2rn9fv

13. Yuan H, Yu C, Li X, et al. Serum uric acid levels and risk of metabolic syndrome: a dose-response meta-analysis of prospective studies. J Clin Endocrinol Metab. 2015;100(11):4198–4207. doi:10.1210/jc.2015-2527

14. Kang DH, Park SK, Lee IK, et al. Uric acid-induced C-reactive protein expression: implication on cell proliferation and nitric oxide production of human vascular cells. J Am Soc Nephrol. 2005;16(12):3553–3562. doi:10.1681/ASN.2005050572

15. Kang DH, Han L, Ouyang X, et al. Uric acid causes vascular smooth muscle cell proliferation by entering cells via a functional urate transporter. Am J Nephrol. 2005;25(5):425–433. doi:10.1159/000087713

16. Perez-Ruiz F, Aniel-Quiroga MA, Herrero-Beites AM, et al. Renal clearance of uric acid is linked to insulin resistance and lower excretion of sodium in gout patients. Rheumatol Int. 2015;35(9):1519–1524. doi:10.1007/s00296-015-3242-0

17. Takir M, Kostek O, Ozkok A, et al. Lowering uric acid with allopurinol improves insulin resistance and systemic inflammation in asymptomatic hyperuricemia. J Investig Med. 2015;63(8):924–929. doi:10.1097/JIM.0000000000000242

18. Neogi T, Jansen TL, Dalbeth N, et al. 2015 gout classification criteria: an American College of Rheumatology/European League Against Rheumatism collaborative initiative. Ann Rheum Dis. 2015;74(10):1789–1798. doi:10.1136/annrheumdis-2015-208237

19. Baraf HS, Becker MA, Gutierrez-Urena SR, et al. Tophus burden reduction with pegloticase: results from Phase 3 randomized trials and open-label extension in patients with chronic gout refractory to conventional therapy. Arthritis Res Ther. 2013;15(5):R137. doi:10.1186/ar4318

20. National Cholesterol Education Program (NCEP). Expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (Adult Treatment Panel III). Third Report of the National Cholesterol Education Program (NCEP) expert panel on detection, evaluation, and treatment of high blood cholesterol in adults (Adult Treatment Panel III) final report. Circulation. 2002;106(25):3143–3421. doi:10.1161/circ.106.25.3143

21. Thottam GE, Krasnokutsky S, Pillinger MH. Gout and metabolic syndrome: a tangled web. Curr Rheumatol Rep. 2017;19(10):60. doi:10.1007/s11926-017-0688-y

22. Doualla-Bija M, Lobe Batchama Y, Moutchia-Suh J, et al. Prevalence and characteristics of metabolic syndrome in gout patients in a hospital setting in sub-Saharan Africa. Diabetes Metab Syndr. 2018;12(6):1007–1011. doi:10.1016/j.dsx.2018.06.015

23. Puig JG, Martínez MA. Hyperuricemia, gout and the metabolic syndrome. Curr Opin Rheumatol. 2008;20(2):187–191. doi:10.1097/BOR.0b013e3282f4b1ed

24. Ridker PM, Buring JE, Cook NR, et al. C-reactive protein, the metabolic syndrome, and risk of incident cardiovascular events: an 8-year follow-up of 14 719 initially healthy American women. Circulation. 2003;107(3):391–397. doi:10.1161/01.cir.0000055014.62083.05

25. Mersal Ezat A, Morsi Ahmed A, Hassanein Alaa M, et al. Does hypertension contribute to eliciting gout symptoms in hyperuricemic patients? Egypt J Hosp Med. 2023;92(1):6824–6831. doi:10.21608/ejhm.2023.318798

26. Pan A, Teng GG, Yuan JM, et al. Bidirectional association between diabetes and gout: the Singapore Chinese Health Study. Sci Rep. 2016;6(9):25766. doi:10.1038/srep25766

27. Dalbeth N, Pool B, Gamble GD, et al. Cellular characterization of the gouty tophus: a quantitative analysis. Arthritis Rheum. 2010;62(5):1549–1556. doi:10.1002/art.27356

28. FitzGerald JD, Dalbeth N, Mikuls T, et al. 2020 American College of Rheumatology guideline for the management of gout. Arthritis Care Res. 2020;72(6):744–760. doi:10.1002/acr.24180

Metformin, a drug primarily used for the treatment of diabetes, has been thought to provide clinical benefits extending to neurodegenerative conditions, including Alzheimer’s disease (AD). Results published in Cureus showed that metformin may not be the optimal pharmacologic agent for patients with diabetes who are trying to target neurodegeneration as their primary concern.¹

“Metformin’s use for DM treatment is generally considered to be a first-line option due to affordability, accessibility, and effectiveness in lowering A1C, as evidenced here with patients having taken metformin having reduced values when compared to sulfonylureas and short-acting insulins with differences of 0.1% and 0.56%, respectively,” the study authors wrote.¹

In Frontiers in Neuroscience, investigators conducted a systematic review and meta-analysis to determine if metformin can prevent cognitive dysfunction, with results showing that the occurrence of cognitive decline decreased for patients with diabetes, but only for the prevention of dementia and not AD. Further, from the authors of a study in Neurobiology of Aging, metformin use did correlate to slower decline on a score of global cognition compared with patients who did not use metformin. However, when examining any diabetes medication, there was no association observed for diabetes medication and cognitive function compared with no diabetes medication.^2,3

In the current study, investigators compared metformin, glucagon-like peptide-1s (GLP-1s), insulin, and sulfonylureas for hemoglobin A_1c, AD development, and mortality. Then, investigators compared the protective value of metformin for patients with diabetes compared to those without diabetes. Data was included from TriNetX, with investigators including patients who were documented to take one of the drug classes of interest. Patients were older than 50 years.¹

Analysis group A included metformin compared with GLP-1s, group B included metformin and sulfonylureas, group C included metformin and short-acting insulin, and group D included metformin users with diabetes and metformin users without diabetes. The results showed a significantly higher risk of AD for analysis A, metformin compared with GLP1s, and there was no statistically significant difference for analysis B or analysis C. As for patients with and without diabetes, the risk of patients with diabetes developing AD who were taking metformin was higher than those without, according to the study authors.¹

As for mortality, analysis A indicated statistically insignificant risk differences, and groups B and C demonstrated lower mortality risk. For patients with and without diabetes, the mortality risk was insignificant. When looking at hemoglobin A_1c, the investigators found that patients taking metformin had a value of 6.53% compared with GLP-1s having 6.28%. Further, for metformin and sulfonylureas, the hemoglobin A_1c values were 6.58% and 6.68%, respectively, and for metformin and short-acting insulins, the values were 6.60% and 7.16%, respectively.¹

“Metformin’s neuroprotective effects against AD, compared to other treatments, were inconsistent,” the authors concluded. “There was no comparison between analyses A-C, in which metformin demonstrated a protective effect against neurologic decline based on the incidence of an AD diagnosis.”

READ MORE: Diabetes Resource Center

Ready to impress your pharmacy colleagues with the latest drug information, industry trends, and patient care tips? Sign up today for our free Drug Topics newsletter.

REFERENCES

1. DiGiovanni A, Shehaj A, Millar D, Tse C, Rizk E. Utility of Pharmacological Agents for Diabetes Mellitus in the Prevention of Alzheimer’s Disease: Comparison of Metformin, Glucagon-Like Peptide-1 (GLP-1) Agonists, Insulin, and Sulfonylureas. Cureus. 2025;17(7):e87350. Published 2025 Jul 5. doi:10.7759/cureus.87350

2. Zhang JH, Zhang XY, Sun YQ, Lv RH, Chen M, Li M. Metformin use is associated with a reduced risk of cognitive impairment in adults with diabetes mellitus: A systematic review and meta-analysis. Front Neurosci. 2022;16:984559. Published 2022 Aug 25. doi:10.3389/fnins.2022.984559

3. Sood A, Capuano AW, Wilson RS, et al. Metformin, age-related cognitive decline, and brain pathology. Neurobiol Aging. 2024;133:99-106. doi:10.1016/j.neurobiolaging.2023.10.005

Women with multiple sclerosis (MS) are significantly less likely than men to be treated with disease-modifying therapies (DMTs) and highly effective DMTs (HE-DMTs), a new study showed.

At comparable levels of disease severity, women had 8% lower odds of receiving any DMT and 20% lower odds of receiving HE-DMT. While pregnancy may partially explain undertreatment in women, this doesn’t fully account for the disparity between the sexes.

“When used early, MS drugs can delay the burden of the disease, so women who are not treated could have worse outcomes in the long term and an increased risk of long-term disability. This loss of chance is not acceptable anymore as there are drugs that are compatible with pregnancy or can continue to fight the disease long after people stop them when they are trying to conceive,” study investigator Sandra Vukusic, MD, PhD, University of Lyon in Lyon, France, said in a news release.

The study was published online on July 30 in Neurology.

Most Patients With MS Are Women

The researchers hypothesized that DMTs may be underprescribed in women, particularly during childbearing years, due to potential clinician bias and previously unclear data on the safety of DMTs before, during, and after pregnancy.

Although women make up roughly 75% of patients with MS, a study published last year showed women are more likely to be taken off DMTs and less likely to have treatment escalated than men.

For the study, the researchers reviewed data of 22,657 patients with relapsing-remitting MS between 1997 and 2022 from the Observatoire Français de la Sclerose en Plaques (74.4% women; mean age at onset, 29 years).

The primary outcome was the annual likelihood of each sex prescribed a DMT while accounting for disease severity and periods of pregnancy and postpartum.

The second outcome was the annual probability of patients receiving an HE-DMT or a specific DMT, and how this was influenced by sex, accounting for covariates such as time, age, and disease duration.

To evaluate the potential impact of pregnancy on prescribing patterns, the researchers analyzed treatment rates in relation to time to childbirth; 36.3% of patients had at least one pregnancy.

Data were analyzed using longitudinal logistic modeling with generalized estimating equations and an inverse probability of censoring weighting.

Disturbing Disparities

Before adjusting for variables, investigators found no significant difference in DMT use between men and women.

After adjusting for periods of pregnancy, postpartum, and disease severity, women were significantly less likely than men to receive a DMT (odds ratio [OR], 0.92; 95% CI, 0.87-0.97) and were even less likely to receive an HE-DMT (OR, 0.80; 95% CI, 0.74-0.86).

“Disturbingly, these disparities appeared as early as 1-2 years after disease onset and persisted at least over the first decade of follow-up,” Gabriel Bsteh, PhD, professor and neurologist at the Medical University of Vienna, and Harald Hegen, MD, PhD, researcher at Innsbruck Medical University, both in Austria, wrote in an accompanying editorial.

Women had a higher mean annual relapse rate and were less likely to be prescribed teriflunomide, S1PR modulators, and anti-CD20s.

The number of treated women decreased 18 months before childbirth, from 42.6% to 27.9% around the time of conception and 11.1% at childbirth.

‘A Timely Wake-Up Call’

A limitation of the study was that the data used were from an expert center on MS — the treatment disparity between sexes could be higher in nonexpert settings. Other limitations included a lack of data on potential pregnancy complications related to discontinuing DMT use and feasibility of replicating results in more diverse populations.

“Anticipation of pregnancy was probably an important factor in this difference between women and men with MS, but there could also be a reluctance to use these treatments when they may actually be the best way to manage the disease and delay disability,” Vukusic noted.

Both Bsteh and Hegen called for clinicians to familiarize themselves with the latest guidelines for safe DMT use in pregnancy and increase education for patients. They described the findings as a “timely wake-up call: Even as our therapeutic arsenal expands, sex-based inertia threatens to blunt its benefit for women with MS.”

See article for full list of author disclosures.

Introduction

Bipolar disorder (BD) is a persistent mental health condition characterized by recurrent episodes of depression interspersed with periods of mania or hypomania.¹ According to findings from the Global Burden of Disease (GBD) study, bipolar disorder is ranked as the sixth leading cause of illness burden attributable to mental and substance use disorders, as measured by disability-adjusted life-years (DALYs).² The implications of bipolar disorder extend beyond the individual, significantly affecting both the physical and mental well-being of patients, while also imposing substantial economic burdens on families and society at large.^3–5 In the clinical management of bipolar disorder, mood stabilizers and antipsychotic medications, including lithium (Li), valproate (VPA), and quetiapine, are predominantly utilized. Research has indicated that patients with bipolar disorder are at an increased risk of developing metabolic disorders following pharmacological treatment, with hypertriglyceridemia being particularly pronounced among these conditions.^6,7

Patients diagnosed with bipolar disorder who also present with comorbid metabolic diseases exhibit more complex clinical manifestations, face greater challenges in treatment, experience poorer prognoses, and have an increased risk of depressive episodes.⁸ Triglycerides serve as a critical lipid marker in the context of metabolic diseases,⁹ and elevated triglyceride levels can not only precipitate metabolic disturbances but also heighten the risk of atherosclerosis, diabetes, and pancreatitis.^10–15 Recent research indicates a correlation between elevated triglyceride levels in individuals with bipolar disorder and increased left ventricular wall thickness, which subsequently raises the risk of heart failure.¹⁶ Furthermore, it is noteworthy that the mortality rate associated with vascular diseases stemming from hypertriglyceridemia in this patient population is approximately double that of the general population,¹⁷ marking it as a significant contributor to mortality among individuals with bipolar disorder.

Research conducted across various countries and regions has consistently demonstrated that the prevalence of hypertriglyceridemia among individuals diagnosed with bipolar disorder is significantly higher than that observed in the general population, with notable variations in prevalence rates. For instance, a study in the Netherlands reported a comorbidity rate of hypertriglyceridemia of 35.3% among patients with bipolar disorder, in stark contrast to a rate of 20.1% in the control group.¹⁸ Similarly, in Spain, the prevalence was found to be 36.1%,¹⁹ while in Taiwan, a study of outpatients with bipolar disorder indicated a prevalence rate as high as 36.8%.²⁰ Furthermore, in Pennsylvania, United States, 41% of patients with bipolar disorder met the diagnostic criteria for hypertriglyceridemia.²¹ These discrepancies in prevalence rates may be attributed to a multitude of factors, including geographical location, ethnicity, lifestyle choices, and healthcare standards. Consequently, there is a pressing need for further research to explore these variations across different regions.

In addition to pharmacological influences, researchers have identified several potential factors that may account for the elevated prevalence of comorbid hypertriglyceridemia among individuals diagnosed with bipolar disorder. From a lifestyle perspective, individuals with bipolar disorder frequently exhibit poor dietary and exercise habits, which can predispose them to hypertriglyceridemia.²² Regarding physiological mechanisms, studies indicate that the levels of peroxisome proliferator-activated receptor gamma (PPARγ) in patients with bipolar disorder are significantly lower than those observed in control groups. PPARγ functions as a critical regulator of immune and metabolic processes, and its diminished levels may facilitate an increase in triglyceride concentrations.²³ Furthermore, external factors such as substance use, including alcohol and nicotine, are more prevalent among patients with bipolar disorder, which can substantially elevate the risk of developing hypertriglyceridemia.²⁴

Moreover, the sample sizes reported in the existing literature are typically constrained. This comprehensive cross-sectional study aimed to investigate the prevalence of hypertriglyceridemia and its related clinical factors in patients diagnosed with bipolar disorder in Anhui Province, China. The goal was to provide substantial scientific evidence that could enhance the prevention and management strategies for both mental health disorders and cardiovascular diseases by conducting a thorough analysis of the clinical characteristics associated with these two conditions.

Material and Methods

Participants

This extensive cross-sectional study involved patients diagnosed with bipolar disorder (male/female ratio = 552/520, depressive episode/manic episode ratio = 375/697) who were admitted to the Affiliated Psychological Hospital of Anhui Medical University. The demographic information and test results of the patients were collected anonymously from the electronic health record system. The inclusion criteria for the study were as follows: 1) patients diagnosed with bipolar disorder according to the ICD-10 criteria, confirmed by two or more attending psychiatrists; 2) patients aged between 18 and 60 years; 3) no history of antidepressant, antipsychotic, or other medication use within three months prior to enrollment; 4) absence of alcohol, tobacco, or other substance dependence; and 5) no history of convulsive electroconvulsive therapy in the preceding three months. The exclusion criteria included: 1) pregnant or lactating women; 2) individuals with neurodegenerative diseases, such as congenital neurodevelopmental delay or Alzheimer’s disease; and 3) patients with organic brain diseases or severe physical illnesses. To mitigate confounding factors, the study adhered strictly to these criteria, resulting in the inclusion of 1,072 patients with bipolar disorder. This study was performed in line with the principles of the Declaration of Helsinki. The study was approved by the Medical Ethics Committee (AMHC) of Anhui Mental Health Center. Because of the retrospective nature of the study and the fact that all data (including basic personal information and detailed medical records) were collected anonymously and encrypted, the Ethics Committee waived the requirement to obtain informed consent.

Demographic Variables

Comprehensive demographic data were gathered from patients diagnosed with bipolar disorder who satisfied the specified inclusion criteria. The variables assessed included sex, age, educational attainment, age at onset of the disorder, duration of the illness, marital status, height, and weight. Body Mass Index (BMI) was computed using the formula weight in kilograms divided by height in meters squared. In accordance with the “Guidelines for the Prevention and Control of Overweight and Obesity in Chinese Adults”, body weight was categorized into four classifications based on BMI, with corresponding numerical values assigned for analytical purposes: a BMI of less than 18.5 was classified as underweight (0), a BMI ranging from 18.5 to less than 24 was categorized as normal weight (1), a BMI from 24 to less than 28 was designated as overweight (2), and a BMI of 28 or greater was classified as obesity (3).

Clinical Assessment

Blood samples were obtained from each patient between 6 a.m. and 8 a.m. following an overnight fast of 8 to 12 hours. The samples were promptly transported to the hospital’s laboratory department for analysis within one hour of collection. Plasma biochemical parameters were assessed using a commercial automatic biochemical analyzer. The parameters measured included blood glucose, total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), triglycerides (TG), alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase (ALP), gamma-glutamyl transpeptidase (GGT), apolipoprotein A1 (ApoA1), apolipoprotein B (ApoB), and uric acid levels. According to the Chinese Blood Lipid Management Guidelines 2023, hypertriglyceridemia is defined as a fasting blood triglyceride concentration of 150 mg/dL (1.7 mmol/L) or higher.²⁵

Statistical Analysis

The analysis was performed using the R programming language. To evaluate the normality of continuous independent variables, the Shapiro–Wilk test and QQ plot were utilized. For continuous variables that conformed to a normal distribution, a t-test was conducted, while the Wilcoxon rank-sum test was applied to continuous variables that exhibited a skewed distribution. Categorical variables were assessed using the Chi-square test. Following this, a multivariate analysis was carried out on independent variables that demonstrated statistical significance (p<0.05) in the univariate analysis. The variance inflation factor (VIF) was calculated to detect multicollinearity among the independent variables, with a VIF threshold exceeding 5 indicating the presence of multicollinearity. After excluding independent variables that contributed to multicollinearity, logistic regression was employed to investigate their association with hypertriglyceridemia. The Akaike information criterion (AIC) was used to determine the combination of independent variables that minimized the AIC value. In this study, a significance level of p<0.05 was established.

Results

Sociodemographic Data

This study encompassed a total of 1,072 eligible patients diagnosed with bipolar disorder, comprising 829 individuals in the non-hypertriglyceridemia cohort and 243 individuals in the hypertriglyceridemia cohort. The observed prevalence of hypertriglyceridemia among patients with bipolar disorder was determined to be 22.6%. The median age of participants in the hypertriglyceridemia group was 33.4 years, which was significantly older than the median age of 30 years in the non-hypertriglyceridemia group. Furthermore, the prevalence of overweight (38.3%) and obesity (30.0%) was markedly higher in the hypertriglyceridemia group compared to the non-hypertriglyceridemia group. Conversely, the incidence of underweight (1.6%) and normal weight (30.0%) individuals was considerably lower in the hypertriglyceridemia group than in the non-hypertriglyceridemia group. Statistical analysis revealed no significant differences in the type of bipolar episode, gender, height, systolic blood pressure, or educational attainment between the two groups (p>0.05). However, both age and weight were significantly greater in the hypertriglyceridemia group compared to the non-hypertriglyceridemia group (p<0.001). (Refer to Table 1 for additional details).

Table 1 Demographic and Clinical Characteristics of Patients with Bipolar Disorder with or Without Hypertriglyceridemia

Clinical Symptoms and Metabolic Parameters

In comparison to the non-hypertriglyceridemia cohort, the hypertriglyceridemia cohort exhibited a significantly greater prevalence of diabetes (p=0.002). Furthermore, the hypertriglyceridemia group demonstrated markedly elevated levels of blood glucose, ALT, AST, ALP, GGT, TC, uric acid, and apolipoprotein B, all of which were significantly higher than those observed in the non-hypertriglyceridemia group (p<0.001). Conversely, the high-density lipoprotein (HDL) levels in the hypertriglyceridemia group were significantly lower than those in the non-hypertriglyceridemia group (p<0.001). No significant difference was found in the levels of apolipoprotein A1 between the two groups (p>0.05) (see Table 1).

Risk Factors

In order to investigate the risk factors associated with comorbid hypertriglyceridemia in individuals diagnosed with bipolar disorder, the variance inflation factor was employed to assess multicollinearity among the independent variables. Subsequently, multivariate logistic regression analysis was conducted to identify the risk factors for comorbid hypertriglyceridemia in this patient population (refer to Table 2). The results of the logistic regression analysis indicated that BMI (OR = 1.51, p < 0.001, 95% CI = 1.23–1.84), blood glucose levels (OR = 1.51, p < 0.001, 95% CI = 1.23–1.84), and TC (OR = 2.88, p < 0.001, 95% CI = 2.34–3.55) were identified as significant risk factors for comorbid hypertriglyceridemia in patients with bipolar disorder. Conversely, HDL-C was found to be a protective factor against comorbid hypertriglyceridemia, with an odds ratio of 0.07 (p < 0.001, 95% CI = 0.03–0.15).

Table 2 Analysis of Risk Factors Associated with Hypertriglyceridemia in Patients with Bipolar Disorder

In light of the aforementioned findings, we have successfully developed a clinically applicable nomogram designed to accurately assess the individualized risk of hypertriglyceridemia among patients with bipolar disorder in Anhui Province, China (refer to Figure 1). This model integrates four independent risk factors: BMI, fasting blood glucose (Glu), TC, and HDL, thereby establishing a multi-dimensional risk assessment framework. The classification of BMI is as follows: a BMI of less than 18.5 is categorized as underweight (0), a BMI between 18.5 and 24 is classified as normal weight (1), a BMI between 24 and 28 is considered overweight (2), and a BMI of 28 or greater is classified as obesity (3). Each predictor variable is assigned a corresponding scale within the nomogram’s coordinate system, allowing clinicians to swiftly ascertain the risk score associated with each indicator through a vertical mapping method (for instance, a BMI of 28 kg/m² yields a maximum score of 13 points, while a Glu level of 18 mmol/L corresponds to 35 points). By summing the scores of all variables (with a total score range of 0–180 points), clinicians can directly interpret the risk of hypertriglyceridemia for patients with bipolar disorder in Anhui Province on the risk probability axis. The nomogram demonstrates robust predictive performance and clinical utility, effectively differentiating between high-risk and low-risk patients. It serves as an intuitive visualization tool for clinicians, facilitating the development of personalized prevention and intervention strategies, thereby enhancing treatment outcomes and the quality of life for patients.

Figure 1 Nomogram to evaluate hypertriglyceridemia in patients with bipolar disorder in Anhui, China. Abbreviations: BMI, body mass index; Glu, fasting blood glucose; TC, total cholesterol; HDL, high-density lipoprotein.

The model’s capacity to differentiate between patients diagnosed with bipolar disorder and those exhibiting hypertriglyceridemia in Anhui Province, China, was assessed utilizing Receiver Operating Characteristic (ROC) plots (refer to Figure 2). The area under the curve (AUC) for models employing individual independent variables—namely BMI, blood glucose, total cholesterol, and high-density lipoprotein—was recorded at 0.659 (95% CI: 0.623–0.696), 0.632 (95% CI: 0.623–0.696), 0.716 (95% CI: 0.623–0.696), and 0.616 (95% CI: 0.623–0.696), respectively. The AUC serves as a critical metric for evaluating the model’s discriminative ability, with values approaching 1 indicating enhanced discrimination. Importantly, the AUC for the composite model incorporating all four variables reached 0.803 (95% CI: 0.772–0.834), significantly surpassing the AUC values of any individual variable. This finding underscores the efficacy of our multivariate integration in augmenting the model’s ability to discern whether patients with bipolar disorder also present with hypertriglyceridemia. Furthermore, the optimal cutoff point for the combined model demonstrated a sensitivity of 0.770 and a specificity of 0.727, indicating that this threshold effectively balances high sensitivity with adequate specificity for the identification of patients at elevated risk for hypertriglyceridemia.

Figure 2 Ability of related factors to discriminate between bipolar disorder patients with and without hypertriglyceridemia. The area under the curve (AUC) of BMI, Glu, TC, HDL and their combination were 0.659 (95% CI: 0.623–0.696), 0.632 (95% CI: 0.623–0.696), 0.716 (95% CI: 0.623–0.696), 0.616 (95% CI: 0.623–0.696) and 0.803 (95% CI: 0.772–0.834), respectively. Best cutoff means a specific value on the abscisce and ordinate, which represents the optimal combination of sensitivity and specificity of the test: specificity: 0.727, sensitivity:0.770. Abbreviations: BMI, body mass index; Glu, fasting blood glucose; TC, total cholesterol; HDL, high-density lipoprotein.

Discussion

This study, for the first time, investigated the prevalence of hypertriglyceridemia among individuals diagnosed with bipolar disorder in Anhui, China, revealing a rate of 22.6%. In contrast, a prevalence rate of 12.2% was recorded within the general population of Anhui, China.²⁶ Additionally, the study identified BMI, blood glucose levels, and TC as significant risk factors for hypertriglyceridemia in this patient population, while HDL was identified as a protective factor. The logistic regression model developed in this study, incorporating these four variables, effectively estimated the risk of comorbid hypertriglyceridemia in Chinese patients with bipolar disorder residing in Anhui province.

The incidence of hypertriglyceridemia among Chinese patients diagnosed with bipolar disorder in Anhui Province is reported to be 22.6%. In comparison, the reported international statistics indicate that the prevalence is 41% in Pennsylvania,²¹ 35.3% in the Netherlands,¹⁸ 36.1% in Spain,¹⁹ and 36.8% in Taiwan.²⁰ The current study benefits from a substantial inpatient sample size, in contrast to other investigations that either employed smaller sample sizes or included outpatient populations, which may account for the observed discrepancies in prevalence rates. Furthermore, variations in lifestyle factors, food safety regulations, dietary customs, exercise patterns, and other regional characteristics may also contribute to these differences. For instance, dietary habits characterized by high saturated fat, carbohydrate intake, or elevated glycemic index, alongside the prevalence of physical activities such as walking or cycling, as well as patterns of excessive alcohol consumption, could significantly influence lipid profiles. Additionally, regional disparities in the selection and utilization of pharmacological treatments for bipolar disorder may be influenced by local healthcare systems, the rigor of monitoring adverse drug reactions, and practices regarding the timely adjustment of medication dosages or substitutions.

Further investigation indicated that the BMI of patients diagnosed with bipolar disorder in the hypertriglyceridemia cohort was significantly elevated compared to those in the non-hypertriglyceridemia cohort, a finding that aligns with the outcomes of the Spanish study.¹⁹ This phenomenon may be attributed to various underlying mechanisms. In individuals with bipolar disorder, elevated triglyceride levels can adversely affect cognitive functioning, resulting not only in diminished cognitive flexibility²⁷ but also heightening the risk of executive function impairment.³ Executive function encompasses several critical domains, including action planning, inhibition, and impulse control, all of which are vital for the long-term objective of sustaining a healthy weight.²⁸ When executive function is compromised, adherence to a regular dietary regimen becomes challenging, thereby predisposing the individual to an increase in BMI. Furthermore, an elevated BMI contributes to the accumulation of body fat, particularly in the abdominal region, which subsequently enhances the liver’s synthesis of very low-density lipoprotein (VLDL), leading to a direct rise in plasma triglyceride levels and perpetuating a detrimental cycle. Consequently, it is imperative for clinicians to implement weight management strategies for patients with bipolar disorder who present with a BMI above the normal range.

The present study identified a significant positive correlation between triglyceride levels and total cholesterol in individuals diagnosed with bipolar disorder. Research conducted by Patel et al further substantiates this finding, indicating that patients exhibiting hypertriglyceridemia are at an increased risk for elevated total cholesterol levels.²⁹ This relationship may be attributed to the unhealthy dietary patterns frequently observed in individuals with bipolar disorder, which often include high sugar and high fat consumption. Elevated sugar intake is a primary contributor to increased triglyceride levels, while a diet rich in fats is associated with heightened total cholesterol levels.³⁰ Both triglycerides and cholesterol are critical components of blood lipids, and their abnormal concentrations are not only closely linked but can also lead to mixed hyperlipidemia when both are elevated, posing significant health risks.

Furthermore, individuals diagnosed with bipolar disorder who also present with elevated triglyceride levels exhibit a markedly heightened risk of abnormal blood glucose levels, a finding that aligns with the research conducted by Calkin CV et al^31,32 This association may be attributable to the administration of atypical antipsychotic medications.³³ Additionally, functional genetic variants in the MTNR1B gene have been shown to impede insulin secretion in patients with bipolar disorder, resulting in increased fasting blood glucose levels.³⁴ Moreover, the characteristic sleep disturbances associated with bipolar disorder not only elevate inflammatory markers but also contribute to heightened levels of stress hormones, such as cortisol, which can further exacerbate blood glucose levels. Conversely, certain studies have indicated that elevated triglyceride levels may independently induce insulin resistance and compromise β-cell functionality, thereby leading to impaired fasting glucose.³⁵ Given the cumulative impact of these various risk factors, patients with bipolar disorder who also experience hypertriglyceridemia are at an increased risk for abnormal blood glucose levels. Consequently, it is imperative to closely monitor blood glucose abnormalities in this patient population.

The aforementioned variables, including BMI, TC, and blood glucose levels, exhibit a positive correlation with triglyceride levels. Notably, patients with bipolar disorder who present with hypertriglyceridemia demonstrate significantly lower levels of HDL compared to those without hypertriglyceridemia, a finding that aligns with the research conducted by Langsted et al^36,37 Firstly, cognitive impairments in individuals with bipolar disorder may lead to dietary preferences that are high in sugars and fats, which can subsequently lower serum HDL levels. This reduction in HDL levels diminishes its anti-inflammatory and antioxidant functions, thereby exacerbating the cognitive deficits experienced by these patients and creating a detrimental feedback loop.^38,39 Secondly, from a lipoprotein metabolism perspective, cholesteryl ester transfer protein (CETP) facilitates the transfer of cholesteryl esters (CE) from HDL to apolipoprotein B-containing lipoproteins (VLDL and low-density lipoprotein [LDL]) in exchange for TG when triglyceride levels are elevated. This exchange results in the production of smaller, denser HDL particles that are enriched with triglycerides and metabolized more rapidly, leading to decreased HDL levels.⁴⁰ Furthermore, research by Chatterjee et al indicates that HDL engages in component exchange with triglyceride-rich lipoproteins to lower TG levels, suggesting that higher concentrations of HDL facilitate the normalization of TG levels, thereby highlighting the interaction between TG and HDL.⁴¹ Previous investigations have demonstrated that managing sugar and carbohydrate intake,²⁹ as well as increasing the consumption of Omega-3 fatty acids,⁴² can effectively reduce TG levels and enhance HDL levels. Consequently, it is imperative for clinicians to consider the dietary and lifestyle habits of patients with bipolar disorder and to implement proactive interventions aimed at improving lipid management and preventing cardiovascular diseases.

Several limitations of this study merit consideration. Firstly, the cross-sectional design utilized does not allow for the clarification of causal relationships between the variables and comorbid hypertriglyceridemia, nor does it capture the dynamic fluctuations of these variables among individuals with bipolar disorder. Secondly, the sample was confined to hospitalized patients with severe mental disorders, thereby excluding outpatients with less severe conditions, which may introduce a potential bias in the statistical analysis. Thirdly, the study did not address the lifestyle habits of patients with bipolar disorder, such as smoking, dietary choices, and physical activity, which could be critical factors in the assessment of comorbid hypertriglyceridemia. Fourthly, although the study population consisted of individuals who had not received medication in the three months prior to the study, the possible impact of previous medication on the findings cannot be overlooked. Lastly, despite the relatively large sample size, the data were collected from a single-center study, which may restrict the generalizability of the results to the wider population in Anhui, China. Given these limitations, future research should consider employing time series analysis to investigate the prevalence and clinical significance of hypertriglyceridemia in patients with bipolar disorder, including both outpatients and healthy controls, thereby expanding the study’s scope and accounting for lifestyle factors and psychiatric medication history. This approach would provide more comprehensive and accurate evidence to guide clinical diagnosis and treatment.

Conclusion

This study presents, for the first time, the prevalence of hypertriglyceridemia among patients diagnosed with bipolar disorder in Anhui Province, China, and elucidates its significant correlation with various clinical indicators. This study contributes novel epidemiological data and clinical relevance to the existing literature in this domain, although the results may not be applicable to the broader population of patients with bipolar disorder in China, given that the sample was obtained from a single inpatient facility. Our findings indicate that the prevalence of hypertriglyceridemia in this population is 22.6%. Multivariate analysis identified BMI, blood glucose levels, and TC as critical risk factors for hypertriglyceridemia, while HDL emerged as a significant protective factor. Consequently, it is imperative for clinical practitioners to routinely monitor serum HDL and TC levels in patients with bipolar disorder and to closely observe trends in these parameters to facilitate the timely identification of dyslipidemia. Additionally, given the elevated rates of overweight and obesity among this patient population, it is essential to implement screening for these conditions. Encouraging patients who are overweight or obese to manage their body weight through appropriate dietary modifications, increased physical activity, and other lifestyle changes may not only enhance their lipid profiles and mitigate the risk of hypertriglyceridemia but could also positively influence the overall prognosis of bipolar disorder and decrease the likelihood of cardiovascular diseases and other related complications. Future research should consider employing multi-center longitudinal cohort studies in conjunction with genomics, metabolomics, and other omics technologies to further investigate the molecular mechanisms underlying comorbid hypertriglyceridemia in patients with bipolar disorder, as well as the impact of various treatment strategies on lipid levels and disease outcomes.

Acknowledgments

We would like to acknowledge the support provided by the Affiliated Psychological Hospital of Anhui Medical University for this study. This study would not have been possible without its assistance.

Funding

The funding for this study was provided by the Hospital Project of Hefei Fourth People’s Hospital under grant numbers HFSY2023ZD01 and HFSY2023YB05. This study was also sponsored by the 2024 Anhui Province Traditional Chinese Medicine Inheritance and Innovation Research Project (Sponsorship Number: 2024CCCX094).

Disclosure

The authors report no conflicts of interest in this work.

References

1. Martinez‐Aran A, Vieta E, Torrent C. et al. Functional outcome in bipolar disorder: the role of clinical and cognitive factors. Bipolar Disord. 2007;9(1–2):103–113. doi:10.1111/j.1399-5618.2007.00327.x

2. Whiteford HA, Degenhardt L, Rehm J, et al. Global burden of disease attributable to mental and substance use disorders: findings from the Global Burden of Disease Study 2010. Lancet. 2013;382(9904):1575–1586. doi:10.1016/S0140-6736(13)61611-6

3. Sánchez‐Ortí JV, Balanzá-Martínez V, Correa-Ghisays P, et al. Specific metabolic syndrome components predict cognition and social functioning in people with type 2 diabetes mellitus and severe mental disorders. Acta Psychiatrica Scandinavica. 2022;146(3):215–226. doi:10.1111/acps.13433

4. Cloutier M, Greene M, Guerin A, Touya M, Wu E. The economic burden of bipolar I disorder in the United States in 2015. J Affective Disorders. 2018;226:45–51. doi:10.1016/j.jad.2017.09.011

5. Wu Q, Zhang X, Liu Y, Wang Y. Prevalence and Risk Factors of Comorbid Obesity in Chinese Patients with Bipolar Disorder. Diabetes Metab Syndr Obes. 2023;16:1459–1469. doi:10.2147/dmso.S404127

6. Kohler-Forsberg O, Sylvia LG, Thase M, et al. Lithium plus antipsychotics or anticonvulsants for bipolar disorder: comparing clinical response and metabolic changes. Aust N Z J Psychiatry. 2023;57(1):93–103. doi:10.1177/00048674221077619

7. Croatto G, Vancampfort D, Miola A, et al. The impact of pharmacological and non-pharmacological interventions on physical health outcomes in people with mood disorders across the lifespan: an umbrella review of the evidence from randomised controlled trials. Mol Psychiatry. 2023;28(1):369–390. doi:10.1038/s41380-022-01770-w

8. Fagiolini A, Kupfer DJ, Houck PR, Novick DM, Frank E. Obesity as a correlate of outcome in patients with bipolar I disorder. Am J Psychiatry. 2003;160(1):112–117. doi:10.1176/appi.ajp.160.1.112

9. Wan H, Cao H, Ning P. Superiority of the triglyceride glucose index over the homeostasis model in predicting metabolic syndrome based on NHANES data analysis. Sci Rep. 2024;14(1):15499. doi:10.1038/s41598-024-66692-9

10. Packard CJ, Boren J, Taskinen M-R. Causes and Consequences of Hypertriglyceridemia. Front Endocrinol. 2020;11:252. doi:10.3389/fendo.2020.00252

11. Zhang BH, Yin F, Qiao YN, Guo SD. Triglyceride and Triglyceride-Rich Lipoproteins in Atherosclerosis. Front Mol Biosci. 2022;9:909151. doi:10.3389/fmolb.2022.909151

12. Grundy SM, Cleeman JI, Daniels SR, et al. Diagnosis and management of the metabolic syndrome: an American Heart Association/National Heart, Lung, and Blood Institute Scientific Statement. Circulation. 2005;112(17):2735–2752. doi:10.1161/CIRCULATIONAHA.105.169404

13. Hossain P, Kawar B, El Nahas M. Obesity and diabetes in the developing world–a growing challenge. N Engl J Med. 2007;356(3):213–215. doi:10.1056/NEJMp068177

14. Guo YY, Li HX, Zhang Y, He WH. Hypertriglyceridemia-induced acute pancreatitis: progress on disease mechanisms and treatment modalities. Discov Med. 2019;27(147):101–109.

15. Niknam R, Moradi J, Jahanshahi KA, Mahmoudi L, Ejtehadi F. Association Between Metabolic Syndrome and Its Components with Severity of Acute Pancreatitis. Diabetes Metabolic Syndrome Obesity. 2020;13:1289–1296. doi:10.2147/dmso.S249128

16. Chen PH, Hsiao CY, Chiang SJ, Chung KH, Tsai SY. Association of lipids and inflammatory markers with left ventricular wall thickness in patients with bipolar disorder. J Affect Disord. 2024;358:12–18. doi:10.1016/j.jad.2024.05.020

17. Weiner M, Warren L, Fiedorowicz JG. Cardiovascular morbidity and mortality in bipolar disorder. Ann Clin Psychiatry. 2011;23(1):40–47.

18. Schuster MP, Borkent J, Chrispijn M, et al. Increased prevalence of metabolic syndrome in patients with bipolar disorder compared to a selected control group-a Northern Netherlands LifeLines population cohort study. J Affect Disord. 2021;295:1161–1168. doi:10.1016/j.jad.2021.08.139

19. Garcia-Portilla MP, Saiz PA, Benabarre A, et al. The prevalence of metabolic syndrome in patients with bipolar disorder. J Affect Disord. 2008;106(1–2):197–201. doi:10.1016/j.jad.2007.06.002

20. Chang HH, Chou CH, Chen PS, et al. High prevalence of metabolic disturbances in patients with bipolar disorder in Taiwan. J Affect Disord. 2009;117(1–2):124–129. doi:10.1016/j.jad.2008.12.018

21. Fagiolini A, Frank E, Scott JA, Turkin S, Kupfer DJ. Metabolic syndrome in bipolar disorder: findings from the Bipolar Disorder Center for Pennsylvanians. Bipolar Disorders. 2005;7(5):424–430. doi:10.1111/j.1399-5618.2005.00234.x

22. Kilbourne AM, Rofey DL, McCarthy JF, Post EP, Welsh D, Blow FC. Nutrition and exercise behavior among patients with bipolar disorder. Bipolar Disord. 2007;9(5):443–452. doi:10.1111/j.1399-5618.2007.00386.x

23. Kasak M, Ceylan MF, Hesapcioglu ST, Senat A, Ö E. Peroxisome Proliferator-Activated Receptor Gamma (PPARγ) Levels in Adolescent with Bipolar Disorder and Their Relationship with Metabolic Parameters. J Mol Neurosci. 2022;72(6):1313–1321. doi:10.1007/s12031-022-02000-2

24. Kumar A, Narayanaswamy JC, Venkatasubramanian G, Raguram R, Grover S, Aswath M. Prevalence of metabolic syndrome and its clinical correlates among patients with bipolar disorder. Asian J Psychiatry. 2017;26:109–114. doi:10.1016/j.ajp.2017.01.020

25. Joint Committee on the Chinese Guidelines for Lipid M. Chinese guidelines for lipid management (2023). Zhonghua Xin Xue Guan Bing Za Zhi. 2023;51(3):221–255. doi:10.3760/cma.j.cn112148-20230119-00038

26. Xu W, Xing XY, He Q, et al. A cross-sectional study on the prevalence and related factors of dyslipidemia among adults in Anhui province, in 2015. Zhonghua Liu Xing Bing Xue Za Zhi. 2020;41(2):195–200. doi:10.3760/cma.j.issn.0254-6450.2020.02.011

27. Van Rheenen TE, McIntyre RS, Balanzá-Martínez V, Berk M, Rossell SL. Cumulative Cardiovascular Disease Risk and Triglycerides Differentially Relate to Subdomains of Executive Function in Bipolar Disorder; preliminary findings. J Affect Disord. 2021;278:556–562. doi:10.1016/j.jad.2020.09.104

28. Dalkner N, Bengesser SA, Birner A, et al. Metabolic Syndrome Impairs Executive Function in Bipolar Disorder. Front Neurosci. 2021;15:717824. doi:10.3389/fnins.2021.717824

29. Patel J, Sharma T, Allan C, Curnew G. Use of Lifestyle Modifications for Management of a Patient with Severely High Total Cholesterol (> 14 mmol/L) and Triglycerides (> 40 mmol/L). J Lifestyle Med. 2021;11(1):43–46. doi:10.15280/jlm.2021.11.1.43

30. Hernandez-Cordero S, Barquera S, Rodriguez-Ramirez S, et al. Substituting water for sugar-sweetened beverages reduces circulating triglycerides and the prevalence of metabolic syndrome in obese but not in overweight Mexican women in a randomized controlled trial. J Nutr. 2014;144(11):1742–1752. doi:10.3945/jn.114.193490

31. Elliott SS, Keim NL, Stern JS, Teff K, Havel PJ. Fructose, weight gain, and the insulin resistance syndrome. Am J Clin Nutr. 2002;76(5):911–922. doi:10.1093/ajcn/76.5.911

32. Calkin CV, Ruzickova M, Uher R, et al. Insulin resistance and outcome in bipolar disorder. Br J Psychiatry. 2015;206(1):52–57. doi:10.1192/bjp.bp.114.152850

33. McElroy SL, Keck Jr PE. Metabolic syndrome in bipolar disorder: a review with a focus on bipolar depression. J Clin Psychiatry. 2014;75(1):46–61. doi:10.4088/JCP.13r08634

34. Hukic DS, Lavebratt C, Frisen L, et al. Melatonin receptor 1B gene associated with hyperglycemia in bipolar disorder. Psychiatr Genet. 2016;26(3):136–139. doi:10.1097/YPG.0000000000000131

35. Ma M, Liu H, Yu J, et al. Triglyceride is independently correlated with insulin resistance and islet beta cell function: a study in population with different glucose and lipid metabolism states. Lipids Health Dis. 2020;19(1):121. doi:10.1186/s12944-020-01303-w

36. Coello K, Vinberg M, Knop FK, et al. Metabolic profile in patients with newly diagnosed bipolar disorder and their unaffected first-degree relatives. Int J of bipolar Disord. 2019;7(1). doi:10.1186/s40345-019-0142-3

37. Shapiro LR, Kennedy KG, Dimick MK, Goldstein BI. Elevated atherogenic lipid profile in youth with bipolar disorder during euthymia and hypomanic/mixed but not depressive states. J Psychosom Res. 2022;156:110763. doi:10.1016/j.jpsychores.2022.110763

38. Hui L, Yin XL, Chen J, et al. Association between decreased HDL levels and cognitive deficits in patients with bipolar disorder: a pilot study. Int J Bipolar Disord. 2019;7(1):25. doi:10.1186/s40345-019-0159-7

39. Huang YJ, Tsai SY, Chung KH, Chen PH, Huang SH, Kuo CJ. State-dependent alterations of lipid profiles in patients with bipolar disorder. Int J Psychiatry Med. 2018;53(4):273–281. doi:10.1177/0091217417749786

40. Girona J, Amigó N, Ibarretxe D, et al. HDL Triglycerides: a New Marker of Metabolic and Cardiovascular Risk. Int J Mol Sci. 2019;20(13):3151. doi:10.3390/ijms20133151

41. Chatterjee C, Sparks DL. Hepatic lipase, high density lipoproteins, and hypertriglyceridemia. Am J Pathol. 2011;178(4):1429–1433. doi:10.1016/j.ajpath.2010.12.050

42. Wang T, Zhang X, Zhou N, et al. Association Between Omega-3 Fatty Acid Intake and Dyslipidemia: a Continuous Dose-Response Meta-Analysis of Randomized Controlled Trials. J Am Heart Assoc. 2023;12(11):e029512. doi:10.1161/JAHA.123.029512