Blood sugar, stress hormone and inflammation combine to dull thinking in depression

A new study has mapped, in unusually fine detail, how three biological forces—metabolic dysfunction, a stress-related signaling molecule called neuropeptide Y, and chronic low-grade inflammation—conspire to erode thinking skills in people with major depressive disorder. The work, published in Biology of Sex Differences, also shows that this biochemical conspiracy plays out very differently in men, premenopausal women, and postmenopausal women, a finding that could reshape how clinicians identify and treat patients whose depression comes bundled with cognitive decline.

Major depressive disorder is far more than a mood condition. Beyond low mood and lost interest, many patients struggle with memory, attention, and executive function—deficits that often persist between depressive episodes and interfere with work, relationships, and daily life. At the same time, researchers have long noted that depression travels with metabolic disturbances such as insulin resistance, and with elevated levels of inflammatory molecules circulating in the blood. What has remained murky is how these threads connect to one another, and why some patients develop cognitive problems while others do not.

The research team, led by investigators at Xiamen Xianyue Hospital affiliated with Xiamen Medical College, recruited 300 people with major depressive disorder—100 men, 100 premenopausal women, and 100 postmenopausal women—along with 150 age- and body mass index-matched healthy controls. Participants were assessed between February 2021 and September 2024 using a battery of measures designed to capture the full biological and clinical picture: the triglyceride-glucose index, a simple calculated marker of insulin resistance; serum neuropeptide Y measured by enzyme-linked immunosorbent assay; inflammatory markers including interleukin-6, tumor necrosis factor-alpha, and C-reactive protein; appetite ratings on a visual analog scale; depression severity on the 17-item Hamilton Depression Rating Scale; and cognition using the Montreal Cognitive Assessment.

The results were striking from the first comparison. Patients with depression showed significantly higher triglyceride-glucose index values, higher neuropeptide Y levels, and greater inflammation than controls, alongside markedly lower cognitive scores. Neuropeptide Y, a peptide released during stress that also regulates appetite and energy balance, was elevated most prominently in premenopausal women—a detail that immediately signaled the importance of sex and reproductive status in the underlying biology.

Correlation analyses deepened the picture. Neuropeptide Y tracked positively with both the triglyceride-glucose index and appetite ratings, with correlation coefficients ranging from 0.43 to 0.52, suggesting that as metabolic dysfunction worsened, the peptide rose in tandem with increased appetite. But the same molecule told a darker story about the brain: it correlated negatively with cognitive performance, with coefficients between −0.35 and −0.46. The stronger a patient’s metabolic derangement and appetite disturbance, the worse their performance on tests of memory and thinking—and once again, these relationships were strongest in premenopausal women.

To test whether neuropeptide Y actually serves as a conduit between metabolism and cognition, the team used a statistical technique called moderated mediation analysis. The findings revealed a layered pathway. Inflammation partially mediated the link between neuropeptide Y and cognitive scores, accounting for roughly 39 percent of the total effect. In other words, high neuropeptide Y appears to fuel inflammatory processes, and those inflammatory signals in turn chip away at cognitive function.

The full chain ran even further back. Neuropeptide Y and inflammation jointly mediated the relationship between the triglyceride-glucose index and cognition, with the joint indirect effect explaining 36.88 percent of the total association. Critically, this mediation was moderated by sex and reproductive status: the pathway was most powerful in premenopausal women, where the indirect effect accounted for 42.37 percent of the total—meaning that in younger women, nearly half of the connection between poor metabolic health and cognitive impairment flows through elevated neuropeptide Y and inflammation.

The authors suggest several mechanisms that could underlie these sex differences. Neuropeptide Y levels are known to vary with estrogen status, and estrogen interacts with both metabolic regulation and immune signaling. Premenopausal women, with higher circulating estrogen, may mount a distinct metabolic and inflammatory response to depression—one in which appetite changes driven by neuropeptide Y are more pronounced, and in which the downstream inflammatory consequences for the brain are amplified. After menopause, this coupling appears to loosen, producing a different risk architecture.

Beyond clarifying mechanism, the study carries immediate clinical promise in the form of a diagnostic tool. The researchers combined four blood measures—the triglyceride-glucose index, neuropeptide Y, interleukin-6, and tumor necrosis factor-alpha—into an integrated biomarker panel and tested its ability to distinguish patients with cognitive impairment using receiver operating characteristic analysis. The panel achieved an area under the curve of 0.869, substantially outperforming the triglyceride-glucose index alone, which managed 0.748. The difference was statistically robust. A simple blood draw, in other words, could one day flag which patients with depression are most at risk of the cognitive dimension of the illness.

The authors caution that the cross-sectional design captures only a snapshot, so cause and effect cannot be definitively established, and longitudinal studies are warranted to confirm whether correcting metabolic dysfunction early can prevent cognitive decline. Still, the implications are considerable. If the pathway holds, interventions targeting insulin sensitivity, neuropeptide Y signaling, or inflammation—tailored to a patient’s sex and reproductive stage—could offer a biological handle on the cognitive symptoms that make depression so disabling. The study is also a reminder that psychiatric illness is embodied: mood, metabolism, immunity, and hormones are not separate stories but a single, sex-specific web, and untangling it may finally explain why the brain falters when the body’s chemistry goes awry.

Neuropeptide Y itself has a long research history that helps explain why it sits at the center of this pathway. It is one of the most abundant neuropeptides in the mammalian nervous system, co-released with norepinephrine during stress, where it classically acts to buffer the cardiovascular and emotional impact of acute stressors. Yet the same peptide is also a potent orexigenic signal, stimulating food intake and promoting fat storage when released in hypothalamic circuits. This dual identity—stress resilience on one hand, metabolic promotion on the other—may account for the seemingly paradoxical findings in the new study, in which higher neuropeptide Y accompanied greater appetite but poorer cognition. Chronic elevation of a peptide designed for short-term stress responses may carry costs that only become apparent over time, particularly in metabolic and immune systems.

The choice of the triglyceride-glucose index as the study’s metabolic anchor reflects broader trends in cardiometabolic research. Unlike direct measures of insulin resistance, which require fasting insulin assays or dynamic testing, the index is computed from routine fasting triglyceride and glucose values, making it inexpensive and easy to deploy in large cohorts and clinical settings. It has been validated across numerous populations as a surrogate for insulin resistance and has repeatedly been associated with adverse outcomes ranging from cardiovascular disease to non-alcoholic fatty liver disease. Its appearance here as a predictor of cognitive impairment in depression extends that literature into psychiatry, and its practicality matters: a marker that requires only a standard metabolic panel could be incorporated into routine psychiatric care far more readily than specialized testing.

The inflammatory markers used in the study likewise represent well-characterized players in the biology of depression. Interleukin-6 and tumor necrosis factor-alpha are pro-inflammatory cytokines that can signal to the brain through both humoral and neural routes, influencing neurotransmitter metabolism, hypothalamic-pituitary-adrenal axis activity, and neuroplasticity. Elevated peripheral inflammation has been reported in subsets of depressed patients for decades, and previous work has linked inflammatory activity to specific symptom dimensions, including fatigue, psychomotor slowing, and cognitive difficulties. The present findings sit comfortably within that tradition while adding a mechanistic twist: inflammation appears not simply as a correlate of depression but as a downstream conduit through which metabolic and neuropeptide signals reach the brain.

The statistical architecture of the study also deserves note. Moderated mediation analysis allows researchers to test both an indirect pathway—whether one variable transmits the effect of another—and whether that transmission differs across subgroups. By applying this framework separately to men, premenopausal women, and postmenopausal women, the investigators could quantify how the same biological chain varies in strength depending on hormonal context. The confidence intervals reported for the indirect effects excluded zero across the full sample and in the premenopausal subgroup, lending statistical weight to conclusions that might otherwise rest on visual inspection of subgroup differences alone.

The diagnostic analysis likewise illustrates methodological principles worth understanding. The area under the receiver operating characteristic curve expresses, on a scale from 0.5 to 1.0, how well a marker separates affected from unaffected individuals, with values above 0.8 generally considered useful discrimination. The jump from 0.748 for the triglyceride-glucose index alone to 0.869 for the four-marker panel, confirmed by a formal comparison test, demonstrates the additive value of measuring multiple biological dimensions rather than any single one. This multibiomarker approach mirrors strategies already standard in cardiovascular risk assessment, where combinations of lipid, inflammatory, and metabolic measures outperform any lone indicator.

Finally, the study’s framing around reproductive stage rather than sex alone points toward a more nuanced future for psychiatric biomarker research. Menopause represents a natural experiment in estrogen withdrawal, and the loosening of the neuropeptide Y–inflammation–cognition coupling observed after menopause suggests that ovarian hormones actively shape how metabolic stress is translated into neural injury. Disentangling these hormonal contributions may ultimately identify which patients benefit most from metabolic or anti-inflammatory interventions, moving psychiatry closer to biologically stratified treatment.

Subject of Research: How neuropeptide Y connects metabolic dysfunction and inflammation to cognitive impairment in major depressive disorder across sex and menopausal status.

Article Title: Interplay of neuropeptide Y, metabolic dysfunction, and inflammation in cognitive impairment of major depressive disorder: a sex-stratified study

Article References: Yuan, Q., Elhassan, M. A. M., Zhang, H., Wang, L., Zhu, X., Wu, Z., Lin, D., & Huang, Z. (2026). Interplay of neuropeptide Y, metabolic dysfunction, and inflammation in cognitive impairment of major depressive disorder: a sex-stratified study. Biology of Sex Differences. https://doi.org/10.1186/s13293-026-00981-y

Image Credits: AI Generated

DOI: 10.1186/s13293-026-00981-y

Keywords: major depressive disorder, neuropeptide Y, cognitive impairment, triglyceride-glucose index, inflammation, insulin resistance, sex differences, menopause, biomarkers, interleukin-6, IL-6, MoCA