As a major global public health challenge, depression contributes substantially to the global disease burden (1)— its core symptoms, defined by the Diagnostic and Statistical Manual of Mental Disorders (DSM-5), directly impair individual functioning and societal productivity. These key symptoms include persistent low mood, anhedonia (loss of interest or pleasure in previously enjoyable activities), generalized fatigue, psychomotor retardation or cognitive slowing, difficulty concentrating, and memory decline; In some cases, individuals may experience hallucinations, delusions, or even suicidal tendencies (2, 3). Together, these symptoms often lead to disrupted daily life, reduced work efficiency, and increased demand for healthcare services.

Fundamentally, depression is a multifactorial disease, and its pathogenesis involves three interrelated dimensions: biology (genetics, neurotransmitter imbalances) (4), psychology (chronic stress, negative cognitive patterns) (5), and society (interpersonal problems, socioeconomic status) (6). Despite extensive research, there remains no unified consensus in the academic community on the exact pathogenesis of depression, as its development typically arises from the complex interplay of the above factors rather than a single cause. In recent years, with advances in pathophysiological research, immune-inflammatory responses and metabolic abnormalities have gradually been confirmed as potential core mechanisms underlying the development of depressive disorders (7). Among these, abnormal serotonin (5-HT) levels and dyslipidemia serve as two key biological indicators. The association and interaction of these indicators with depression have become crucial scientific questions that urgently need clarification in the field.

Serotonin is an important neurotransmitter that plays a critical role in numerous physiological processes, including platelet aggregation, pain, sleep, appetite, muscle contraction, mood, and compulsive behaviors (8). Targeting the serotonin system is a key strategy for developing new potential antidepressants. Serotonin and its receptors are distributed in the central nervous system (CNS), peripheral nervous system (PNS), and multiple non-neuronal tissues, such as those in the gut, cardiovascular system, and blood. Serotonin can bind to cell-surface 5-HT1B receptors to regulate various physiological functions (e.g., pain, sleep, mood, and memory), and its levels have been associated with anxiety, depression, and schizophrenia (9). It has been reported that 5-HT2A receptor density is significantly increased in the cortex of postmortem brain tissue from depressed patients and individuals who died by suicide. This increase may be a compensatory response by the brain to reduced serotonin levels in depressed patients, achieved by upregulating 5-HT2A receptor density (10). Long-term administration of some antidepressant drugs downregulates 5-HT2A receptor density and exerts antidepressant effects by modulating subcortical circuits in the forebrain (11). These findings support a link between 5-HT2A receptor regulation and the therapeutic mechanism of these antidepressants; however, evidence for a direct association with depression pathogenesis remains to be supplemented. Vargas MV et al. (12)demonstrated that primary localization of 5-HT2ARs in cortical neurons is intracellular, cellular import of serotonin leads to structural plasticity and antidepressant-like effects.

The traditional monoamine hypothesis posits a link between brain monoamine neurotransmitter levels and depression onset, which laid the foundation for early antidepressant development but is now considered overly simplified—for instance, it cannot explain the delayed efficacy of monoamine-targeting drugs or why many patients fail to respond. Recent studies have updated this view by situating the monoamine system within a multi-mechanism network, with neuroinflammation as a key regulatory node. In depressed patients, elevated pro-inflammatory cytokines (e.g., IL-6, TNF-α) disrupt monoamine homeostasis: they inhibit tryptophan hydroxylase (reducing 5-HT synthesis) and activate indoleamine 2,3-dioxygenase (shifting tryptophan metabolism to neurotoxic metabolites), exacerbating monoamine imbalance (13, 14). Conversely, monoamines (especially 5-HT) modulate glial activation to regulate neuroinflammation, forming a bidirectional loop (15, 16). Notably, 5-HT2A receptors, which regulate monoamine release, may further mediate this ‘inflammation-monoamine’ crosstalk (17), linking their function to both depression pathogenesis and antidepressant mechanisms more precisely.

In the central nervous system, lipids are critical for maintaining the structure, function, and membrane integrity of various cells. Specifically, cholesterol—a key lipid component—modulates cellular signaling pathways and has been implicated in both the mechanism of antidepressant action and the regulation of emotional stability. Since blood cholesterol levels correlate with brain cholesterol levels, a decrease in neuronal membrane lipids can reduce serotonin (5-HT) receptor levels in the membrane, and this reduction is associated with increased risks of depression and suicide (18). This lipid-5-HT axis may further interact with 5-HT-related receptor complexes, which are mechanisms closely linked to major depressive disorder. For instance, Borroto-Escuela DO et al. (19) demonstrated that serotonin heteroreceptor complexes, which mediate signal integration in neurons and astrocytes, are highly relevant to the pathogenesis of major depressive disorder. Moreover, Ambrogini P et al. (20) found that 5-HT1A receptor-FGFR1 heteroreceptor complexes differentially modulate G protein-coupled inwardly rectifying potassium (GIRK) currents in the dorsal hippocampus and dorsal raphe nucleus serotonin neurons between control rats and a genetic depression model. These findings suggest that the function of 5-HT receptor complexes may be altered in the context of depression-related lipid/5-HT dysregulation. Beyond these molecular mechanisms, neuroplasticity changes in key brain regions also contribute to depression pathogenesis, with the hippocampus, prefrontal cortex, and amygdala being core areas involved (21). Notably, these neuroplasticity changes are regulated by multiple systems, including glutamatergic signaling, neurotrophic factors, monoamine neurotransmitters (e.g., 5-HT), and neuroinflammation—hinting at potential crosstalk between lipid/5-HT pathways and neuroplasticity regulation in depression.

In addition to the neurotransmitter system, lipid disorders in metabolic abnormalities have also been shown to be closely associated with depression. Studies have demonstrated that lipid metabolism disorders—specifically, lower levels of total cholesterol and LDL-C—are linked to an increased risk of suicide in patients with depression, making these lipid abnormalities among the high-risk factors for suicide in this population (22, 23). These reduced lipid levels may serve as potential biomarkers for predicting suicide attempts in patients with depression. The study by Nikolić V et al. helps elucidate a potential biological mechanism connecting dietary fiber intake to lipid profiles, thereby providing mechanistic support for the observed association between low HDL-cholesterol levels and depression (24). Recent advances in biomarker research have also provided new insights into the diagnosis and management of depression. These advances not only emphasize electrophysiological markers of depression (25) and genetic and epigenetic factors that underlie neurobiological and inflammatory pathways (26), but also suggest potential crosstalk between lipid metabolism and these neurobiological and inflammatory pathways in the pathogenesis of depression.

In specific populations, this association exhibits significant gender and age differences—for example, Yang et al. (27) found that women are more likely than men to develop dyslipidemia following antidepressant therapy, characterized by elevated triglyceride and LDL-C levels, with these elevations being more pronounced in women. Interestingly, high HDL-cholesterol levels in middle-aged adults have been associated with an increased risk of depression, and a relationship has been observed between depression and levels of total cholesterol and LDL-C (28). Furthermore, Mehdi SMA et al. confirmed, based on multi-generational depression cohorts and national health and nutrition survey data, that depression is significantly associated with lipid disorders (29). This provides empirical evidence at the population level for the involvement of lipid metabolism—particularly abnormal HDL-cholesterol levels—in the pathological process of depression.

At the mechanistic level, short-term population-based follow-up studies have shown that metabolic and hunger-related hormones (e.g., growth hormone, cortisol, leptin, ghrelin) directly affect serum lipid levels (30),or interact synergistically with immune inflammatory responses to jointly promote the occurrence and aggravation of depressive symptoms. Additionally, the use of antidepressant drugs has also been reported to alter lipid metabolism (31), which further intensifies the interaction between metabolic and hunger-related hormones and immune inflammatory responses. It is worth noting that metabolic syndrome is a pathological state characterized by multiple risk factors, including obesity, hypertension, dyslipidemia, and abnormal glucose metabolism. It promotes the development of cardiovascular and cerebrovascular diseases, as well as diabetes mellitus, and importantly, its association with depression has also been linked to lipid metabolism abnormalities (32–34). Studies have indicated that the association between depression and dyslipidemia is similar in strength and direction across different ages (35). Hence, lipid levels have the potential to be biomarkers for the diagnosis, classification, and prognosis of depression. However, there is still no definitive conclusion about the relationship between depression and dyslipidemia at different ages, and more studies are needed to explore this relationship.

Collectively, a large number of studies have confirmed correlations between lipid levels, 5-HT levels, and depression; however, the association between specific lipid parameters (e.g., HDL, LDL, triglycerides) and 5-HT-related indices and depression risk has rarely been investigated, and few studies have analyzed the bidirectional and causal relationships among these factors. Therefore, this study explored the bidirectional and causal relationships between depression and both serum 5-HT levels and dyslipidemia via a meta-analysis, thereby providing reliable evidence for the prevention and treatment of depression and a theoretical basis for further investigating its pathogenesis.

The main objective of this meta-analysis was to systematically analyze the bidirectional and causal relationships between lipid parameters and depression: specifically, to investigate whether dyslipidemia exacerbates depression severity, whether depression further affects lipid levels, and whether these clinical indicators can serve as biomarkers for assessing depression risk. By analyzing seven included studies, we found that changes in serum HDL and triglyceride levels were significantly associated with an increased risk of depression (HDL: OR = 1.28, 95% CI: 1.07–1.53; triglycerides: OR = 1.41, 95% CI: 1.20–1.66; both P < 0.05). In contrast, compared with non-depressed individuals, depressive disorders were not significantly associated with changes in serum HDL, LDL, triglyceride, or total cholesterol levels (HDL: OR = 0.88, 95% CI: 0.58–1.35; LDL: OR = 1.05, 95% CI: 0.88–1.24; triglycerides: OR = 1.05, 95% CI: 0.91–1.21; total cholesterol: OR = 1.11, 95% CI: 0.90–1.32; all P > 0.05).

In summary, the risk of depression is associated with changes in serum HDL and triglyceride levels, and these lipid indicators can serve as biomarkers for assessing depression risk. The pathogenesis of depression is complex, involving biological, psychological, and social environmental factors. This study focuses on the bidirectional relationship between blood lipids and depression; investigating this association is of great significance for understanding the pathogenesis of depression, formulating prevention strategies, and developing individualized treatment plans for patients.

At the same time, this study can help clinicians evaluate the lipid status of patients with depression, develop targeted and timely treatment plans, and provide guidance for family members on patient care—such as instructing patients to adhere to a diet rich in dietary fiber and low in saturated fat, and engage in moderate-intensity physical activity to manage lipid levels. Ultimately, this may help minimize the exacerbation or recurrence of depression in affected patients through lipid control.

For clinical practice, the following specific recommendations are provided: Clinicians are recommended to perform a four-component lipid profile (total cholesterol, triglycerides, LDL-C, HDL-C) and apolipoprotein A1/B testing in patients with depression during their first visit. This is particularly important for patients with obesity, metabolic syndrome, or those receiving long-term antidepressant therapy—some of which may affect lipid metabolism. For these high-risk patients, lipid testing should be repeated every 3–6 months to dynamically monitor changes in lipid metabolism.

If a depressed patient presents with abnormal lipid parameters and moderate-to-severe depressive symptoms, it is recommended to consult with an endocrinologist first. Based on the consultation advice, psychological intervention, antidepressant treatment, and lipid-lowering intervention can be combined: start with lifestyle modifications; if lipid levels remain abnormal after 3 months, short-term lipid-lowering medications may be initiated—with close monitoring of potential adverse effects and drug-drug interactions with antidepressants.

Egger’s test results indicated no significant publication bias among the included studies, suggesting that the results of this meta-analysis are relatively reliable—at least in terms of publication bias control. However, several limitations should be acknowledged: First, the included studies were not strictly categorized by key variables, and most lacked explicit subgrouping. This inconsistency may have introduced some selection bias into the results. Second, the scales used to assess depression in the included studies were not standardized, which may have compromised the accuracy of outcome assessment. Third, the number of included studies investigating the association between lipid parameters and depression was relatively small. Additionally, many relevant studies only assessed whether overall lipid levels were abnormal and did not report detailed lipid parameters, so these studies could not be included in this meta-analysis. Furthermore, the search was limited to English-language publications during the literature retrieval process, which may exclude region-specific evidence published in other languages. Therefore, future studies should include non-English literature to deepen our understanding of this association.

The articles included in this study cover regions of France (n = 4 studies), China (n = 1 study), Canada (n = 1 study), and Tel Aviv, Israel (n = 1 study). This uneven distribution might affect the generalizability of result interpretation. First, interference from regional and dietary cultural differences: The included studies were from France (with a Mediterranean diet pattern, characterized by high intake of unsaturated fats and dietary fiber), China (with an East Asian diet pattern, characterized by high carbohydrate intake and low saturated fat intake), Canada (with a Western diet pattern, high in processed foods and saturated fat), and Tel Aviv, Israel (with a Middle Eastern diet pattern, high in olive oil and spice intake). Varied dietary patterns have regulatory effects on lipid metabolism, and these regional and dietary cultural differences may lead to variations in the magnitude of the association between lipid levels, 5-HT, and depression across studies—for example, the high unsaturated fat intake in the Mediterranean diet may mitigate the negative association between low HDL and depression, resulting in a weaker effect size in French studies (50, 51). Second, the potential influence of genetic background: The included studies covered the Caucasian population (from France, Canada, and Tel Aviv, Israel) and the East Asian population (from China). Different populations vary in the allele frequencies of genes related to lipid metabolism and key rate-limiting enzymes for 5-HT synthesis. These genetic differences may lead to population-specific sensitivity thresholds for the effects of lipids and 5-HT on depression, thereby increasing heterogeneity among studies. Therefore, future research directions should consider the following improvements: use subgroup analysis (stratified by region, population, and dietary pattern) or meta-regression analysis (with “diet type” and “population” as covariates) to quantify the impact of heterogeneity sources on the association effect and improve the accuracy of result interpretation.

Finally, although serum 5-HT was included in our search strategy, we did not identify any studies that met the inclusion criteria. The missing 5-HT data may lead to two limitations: First, although this study revealed the bidirectional association between lipid indicators and depression, the lack of 5-HT data prevented us from conducting mediation analysis to verify whether lipids regulate serum 5-HT levels to indirectly affect the onset of depression—thus failing to clarify the key mediating pathway. Second, there is a potential impact on the applicability of the study results. Some patients with depression may have abnormal serum 5-HT levels independently of lipids (e.g., primary 5-HT synthesis disorder). Due to the lack of this indicator data, this study cannot distinguish subgroup differences between “pure lipid-related depression” (depression with normal serum 5-HT levels) and “depression related to both lipids and 5-HT” (depression with concurrent lipid abnormalities and 5-HT dysregulation). This may reduce the applicability of the study conclusions to patients with comorbid abnormal 5-HT levels and also fails to provide a reference for whether it is necessary to monitor 5-HT in conjunction with lipid indicators to optimize depression risk stratification in clinical practice.

In summary, our meta-analysis shows that among serum lipid parameters, lower HDL levels and higher triglyceride levels exhibit a stronger association with depression onset. This relationship suggests that integrating these lipid indices could aid early intervention, potentially improving prognosis and quality of life in depressed patients. Despite this potential, current evidence remains insufficient. Further large-scale, prospective studies—particularly those focusing on clinical utility, determination of cutoff values, and validation in specific populations—are needed before these biomarkers can be widely implemented in clinical practice.