Lipid and Inflammatory Markers Show Weak, Inconsistent Links to Macular Degeneration

An exploratory cross-sectional analysis from the Beichen Eye Study has found that several serum lipid and inflammatory biomarkers show only nominal or model-dependent associations with age-related macular degeneration (AMD), and none of the composite markers the researchers prioritized survived correction for multiple comparisons. The study, published in BMC Ophthalmology, analyzed data from 4,748 participants aged 50 years and older, of whom 1,245 had AMD, corresponding to a prevalence of 26.22% (95% CI 24.98%–27.49%).

The researchers assessed eight composite biomarkers derived from routine blood measurements: the triglyceride-to-HDL-C ratio, the HDL-C-to-LDL-C ratio, the neutrophil-to-HDL-C ratio (NHR), the lymphocyte-to-HDL-C ratio (LHR), the monocyte-to-HDL-C ratio (MHR), the platelet-to-HDL-C ratio (PHR), the platelet-to-lymphocyte ratio (PLR), and the neutrophil-to-lymphocyte ratio (NLR).

In the primary tertile-based analysis, the highest tertile of the TG-to-HDL-C ratio showed a nominal inverse association with AMD (odds ratio 0.80, 95% CI 0.69–0.95; P for trend 0.010), and the highest NHR tertile also showed a nominal inverse association (odds ratio 0.83, 95% CI 0.69–0.98; P for trend 0.032). However, the study had prespecified a stringent Bonferroni-corrected significance threshold of P < 0.00625 for these eight composite biomarkers, and no biomarker met that threshold.

The four standard lipid parameters—total cholesterol, triglycerides, HDL-C, and LDL-C—were examined in a prespecified secondary analysis with its own corrected threshold of P < 0.0125. Here, the highest triglyceride tertile showed an inverse association with AMD (odds ratio 0.79, 95% CI 0.67–0.94; P for trend 0.006), which did meet the secondary threshold. HDL-C, by contrast, showed no association in the tertile analysis (P for trend 0.165).

The picture shifted again in continuous-variable sensitivity analyses, where HDL-C was positively associated with AMD (P = 0.010 and 0.017), while the triglyceride association attenuated (P = 0.020 and 0.052). The authors emphasize that this positive HDL-C signal appeared only in the continuous-variable models and not in the tertile-based analysis, underscoring how sensitive the findings were to the way the biomarkers were modeled.

All analyses used multivariable logistic regression adjusted for a broad set of potential confounders, including age, sex, body mass index, fasting glucose, diabetes, hypertension, smoking, alcohol consumption, coronary heart disease, and statin use.

The authors are explicit about the limits of their work. As a cross-sectional study, it measured biomarkers and AMD status at the same time, so it cannot establish whether the observed lipid differences preceded the disease or resulted from it. The associations that did emerge were described as nominal or model-dependent, and the paper concludes that the findings identify candidate biomarkers for validation in prospective cohorts but do not support their use in clinical risk assessment.

The study’s design also carries statistical implications. Testing eight composite biomarkers plus four standard lipid parameters creates a substantial risk of false-positive findings, which is precisely why the researchers applied Bonferroni correction and reported which results survived. By that standard, only the inverse triglyceride association in the secondary analysis cleared its prespecified bar, and even that signal weakened in the continuous-variable sensitivity analyses.

For readers hoping for a simple blood test to gauge AMD risk, the message is one of caution rather than promise. The most robust observation in this large, population-based dataset was a modest inverse association between higher triglyceride levels and prevalent AMD, alongside a fragile positive HDL-C association that appeared under only one modeling approach. Whether these patterns hold up over time, and whether they reflect causal biology or residual confounding, will require the prospective cohort studies the authors call for.

In the meantime, the Beichen Eye Study results serve as a reminder of how easily biomarker associations can shift with analytical choices. Ratios that combine lipid and immune measurements, though convenient and increasingly popular in the literature, did not outperform the standard lipid fractions here once multiple-comparison correction was applied. The findings add a carefully quantified, if deliberately restrained, data point to the ongoing debate about the metabolic underpinnings of age-related macular degeneration.

Subject of Research: Cross-sectional associations between serum lipid and inflammatory biomarkers and age-related macular degeneration

Article Title: Serum lipid and inflammatory markers in age-related macular degeneration: an exploratory cross-sectional analysis from the Beichen Eye Study

Article References: Zou, G., Gao, F., Zhang, L., & Li, X. (2026). Serum lipid and inflammatory markers in age-related macular degeneration: an exploratory cross-sectional analysis from the Beichen Eye Study. BMC Ophthalmology. https://doi.org/10.1186/s12886-026-05413-4

Image Credits: AI Generated

DOI: 10.1186/s12886-026-05413-4

Keywords: age-related macular degeneration, serum lipids, inflammatory biomarkers, triglycerides, HDL cholesterol, cross-sectional study, Beichen Eye Study, Bonferroni correction, logistic regression, BMC Ophthalmology, Serum, lipid