Blog

For adults with metabolic dysfunction–associated steatotic liver disease (MASLD), the Dietary Inflammatory Index (DII) was not significantly linked to key indicators of liver disease severity, including hepatic fibrosis and the Fatty Liver Index (FLI), according to a cross-sectional study published in Frontiers in Nutrition.¹

Researchers used data from the 2017 to 2020 cycles of the National Health and Nutrition Examination Survey to assess whether the inflammatory potential of diet, as captured by the DII, was predictive of MASLD-related risks in this population. They also evaluated whether the Systemic Immune-Inflammation Index (SII) might mediate any such associations. Despite prior studies linking DII to chronic diseases such as metabolic syndrome, the current analysis did not find meaningful correlations with liver-specific pathology.

No Statistically Significant Associations

In the fully adjusted linear regression model, DII was not significantly associated with FLI in adults with MASLD (β, 0.32; 95% CI, –1.393 to 2.034; P = .631). Similarly, logistic regression showed no link between DII and the presence of liver fibrosis (OR, 1.152; 95% CI, 0.885-1.499; P = .210). Subgroup analyses confirmed that these findings were consistent across demographic and metabolic stratifications, with no evidence of effect modification.

Mediation analysis also showed that SII did not significantly mediate the relationship between DII and either FLI or fibrosis. For example, the average causal mediation effect of SII on liver fibrosis was not significant (β, –0.001; 95% CI, –0.002 to 0.001; P = .364), nor was the average direct effect (β, 0.002; 95% CI, –0.013 to 0.017; P = .760).

Systemic vs Organ-Specific Inflammation

The findings underscore potential limitations of using systemic inflammatory indices like DII and SII to assess organ-specific conditions like MASLD. Although DII has been associated with disease onset in prior studies, the researchers said its static, short-term dietary recall format may not adequately reflect the chronic, cumulative dietary exposures that contribute to liver damage over time.

“This temporal mismatch between exposure assessment and disease evolution may further weaken its prognostic utility,” the authors added.

They also suggested MASLD’s pathophysiology may involve a transition from inflammation-driven injury to metabolic toxicity–driven progression, mirroring trends seen in type 2 diabetes. In patients with type 2 diabetes, early insulin resistance can progress to β-cell dysfunction, ultimately culminating into β-cell apoptosis, dedifferentiation, and functional exhaustion.² Early in MASLD progression, gut-derived inflammation may dominate, but over time, hepatocyte lipotoxicity and stellate cell activation become the primary drivers of fibrosis and steatosis.¹ In this context, the findings suggest generalized systemic markers may lose predictive value.

Nutrient-Specific Effects

The researchers noted another limitation of the DII: its failure to capture the distinct biological effects of specific nutrients. For instance, omega-3 fatty acids may reduce hepatic fibrogenesis by inhibiting the NF-κB pathway and reducing oxidative stress. In contrast, fructose and saturated fats can directly promote lipogenesis and hepatocyte damage. By aggregating these effects into a single score, DII may obscure important mechanistic differences.

“This heterogeneity in nutrient-specific effects underscores the need to move beyond generalized inflammatory scores and toward mechanistic dissection of diet–liver interactions at the molecular level,” the authors said. “Such insights are essential for developing targeted and effective nutritional interventions for MASLD.”

References

1. Sang Z, Wang H, Leng Y, et al. Association of dietary inflammatory index with liver fibrosis and fatty liver index in a population with metabolic dysfunction-associated steatotic liver disease: a cross-sectional study. Front Nutr. 2025;12:1594192. doi:10.3389/fnut.2025.1594192
2. Cerf ME. Beta cell dysfunction and insulin resistance. Front Endocrinol (Lausanne). 2013;4:37. doi:10.3389/fendo.2013.00037

A new wearable wristband could significantly improve diabetes management by continuously tracking not only glucose but also other chemical and cardiovascular signals that influence disease progression and overall health. The technology was published in Nature Biomedical Engineering.

The flexible wristband consists of a microneedle array that painlessly samples interstitial fluid under the skin to measure glucose, lactate and alcohol in real time using three different enzymes embedded within the tiny needles. Designed for easy replacement, the microneedle array can be swapped out to tailor wear periods. This reduces the risk of allergic reactions or infection while supporting longer-term use.

Simultaneously, the wristband uses an ultrasonic sensor array to measure blood pressure and arterial stiffness, while ECG sensors measure heart rate directly from wrist pulses. These physiological signals are key indicators of cardiovascular risk, which is often elevated in people with diabetes but is rarely monitored continuously outside of a clinical setting.

Comprehensive and effective management of diabetes requires more than just a single glucose reading.” 

An-Yi Chang, postdoctoral researcher in the Aiiso Yufeng Li Family Department of Chemical and Nano Engineering at UC San Diego

Factors like diet, alcohol intake, exercise and stress influence blood sugar and heart health in ways that traditional monitoring systems cannot fully capture.

“By tracking glucose, lactate, alcohol and cardiovascular signals in real time, this pain-free wristband can help people better understand their health and enable early action to reduce diabetes risk,” added Chang, who is a co-first author on the study with Muyang Lin, Lu Yin and Maria Reynoso, all from the same department.

The development of this wearable system was made possible by the collaboration of the research groups led by Joseph Wang and Sheng Xu, both professors in the Aiiso Yufeng Li Family Department of Chemical and Nano Engineering at the UC San Diego Jacobs School of Engineering. Wang’s group specializes in creating wearables that can monitor multiple chemical biomarkers in the body simultaneously, while Xu’s group specializes in developing wearable ultrasound sensors that can monitor cardiovascular signals deep inside the body. By combining their expertise, the teams designed a device that provides continuous, simultaneous measurement of biomarkers and cardiac signals in a single wearable wristband platform.

A smart device linked to the wristband displays live data streams from the sensors. It shows blood pressure, heart rate and arterial stiffness alongside real-time readings of glucose, alcohol and lactate levels. This enables wearers to see how daily activities-like meals, alcohol intake or exercise-affect their body in real time and in turn, obtain personalized insights into their metabolic and cardiovascular responses.

The wristband demonstrated excellent agreement with commercial devices across a variety of tests. When monitoring glucose, results closely matched those of a blood glucose meter and continuous glucose monitor while simultaneously capturing cardiovascular responses. Similarly, tests tracking alcohol intake aligned with a breathalyzer, and lactate monitoring during exercise paralleled results from a blood lactate meter. At each step, the wristband provided continuous, simultaneous monitoring of additional signals, including real-time quantitative blood pressure, heart rate and arterial stiffness.

This capability could offer wearers a comprehensive physiological snapshot during everyday activities. It could also help patients and clinicians identify dangerous trends before they escalate, potentially alerting users to cardiovascular risks that traditional glucose monitors would miss.

Next steps include expanding the wearable system to include additional chemical and cardiovascular markers, and designing it to be powered by sweat or sunlight. The researchers also envision integrating machine learning algorithms to analyze the vast amounts of personal data the system collects.