Simple Blood Ratio Combining Uric Acid and HDL Cholesterol Flags Hypertension Risk in Two Large Cohorts

Hypertension remains one of the most widespread cardiovascular conditions in the world, quietly elevating the risk of heart attack, stroke, kidney failure, and vascular dementia while often producing no symptoms until serious damage has already occurred. Because the condition is so tightly interwoven with metabolic and vascular risk factors, researchers have long searched for blood markers that could identify people at elevated risk before a formal diagnosis is made. A new cross-sectional study from Aerospace Center Hospital in Beijing, published in BMC Endocrine Disorders, now points to a remarkably simple candidate: the ratio of serum uric acid to high-density lipoprotein cholesterol, known in the literature as UHR. Drawing on more than 26,000 adults from two of the largest population health databases in the United States and China, the team found that this single calculated ratio was independently and positively associated with self-reported physician-diagnosed hypertension, and they converted the finding into a nomogram, a graphical risk calculator, that performed acceptably in stratifying hypertension risk across both populations.

The study leveraged the National Health and Nutrition Examination Survey, NHANES, covering cycles from 1999 through 2018 with 16,056 participants, and the China Health and Retirement Longitudinal Study, CHARLS, from 2011 with 10,686 participants. NHANES is a nationally representative survey of the civilian noninstitutionalized population of the United States that combines questionnaires, physical examinations, and laboratory measurements, while CHARLS follows middle-aged and older adults in China with comparable biochemical and lifestyle data. By extracting serum uric acid and high-density lipoprotein cholesterol levels from each cohort and dividing the former by the latter, the researchers computed UHR for every participant. The appeal of this ratio lies in its biology: uric acid is the end product of purine metabolism and has been implicated in endothelial dysfunction, oxidative stress, and vascular inflammation, whereas high-density lipoprotein cholesterol, HDL-C, is the antiatherogenic lipoprotein fraction that participates in reverse cholesterol transport and protects the vascular wall. Dividing a pro-vascular risk molecule by a protective one yields a marker that captures the net metabolic balance between the two opposing forces.

Before running any traditional statistical models, the team conducted an exploratory machine learning analysis using Extreme Gradient Boosting, XGBoost, on the full dataset in each cohort. XGBoost is a decision-tree ensemble algorithm that iteratively fits new trees to the residuals of previous predictions and can rank the relative importance of every input variable in association with an outcome. In both NHANES and CHARLS, UHR emerged with a stable and notable relative importance for self-reported physician-diagnosed hypertension status, suggesting that the ratio carried genuine predictive signal rather than statistical noise. Baseline comparisons reinforced this impression, revealing that UHR differed significantly between participants with and without hypertension in both cohorts.

To test whether the association was independent of confounding factors, the researchers constructed three hierarchically adjusted logistic regression models. Model 1 was a crude model containing UHR alone. Model 2 additionally adjusted for age, gender, and education level. Model 3 was fully adjusted for a comprehensive panel of demographic characteristics, lifestyle factors, comorbidities, and hematological and biochemical indicators. Across progressive adjustment, elevated UHR remained independently and positively associated with increased hypertension risk in a graded manner, meaning that higher ratios corresponded to higher odds of hypertension even after accounting for the full set of relevant covariates. The graded pattern is particularly important from a clinical standpoint, because it implies a continuous risk relationship rather than a threshold effect, allowing risk to be stratified across the entire population distribution of the ratio.

Stratified analyses were then conducted to assess the stability of the UHR-hypertension association across subgroups defined by age, sex, smoking exposure, and other characteristics. The positive association proved consistent across most subgroups in both databases, indicating that the relationship was not confined to a particular demographic slice but held broadly across the two very different populations sampled. This cross-population consistency is a notable strength of the study, because NHANES and CHARLS sample populations with distinct dietary patterns, genetic backgrounds, healthcare systems, and metabolic risk profiles, and a marker that behaves similarly in both settings is more likely to reflect a genuine biological relationship than a population-specific artifact.

The final component of the study was the construction of a nomogram, a graphical scoring tool that translates a statistical prediction model into a point-based chart that clinicians can use at the bedside or in screening programs. Each predictor, including UHR, contributes points based on its magnitude, and the summed points map to an estimated probability of hypertension. The team evaluated the nomogram using standard performance metrics, including the area under the receiver operating characteristic curve, AUC, for discrimination, calibration plots for agreement between predicted and observed probabilities, and decision curve analysis, DCA, for clinical net benefit. The constructed nomogram showed favorable calibration, clinical net benefit, and discriminative performance in both databases, meaning that the tool not only ranked individuals correctly by risk but also produced probability estimates that tracked observed hypertension prevalence and offered net benefit across clinically relevant threshold probabilities.

The authors conclude that UHR is an independent and stable factor associated with self-reported physician-diagnosed hypertension risk, and that the UHR-integrated nomogram shows acceptable performance in hypertension risk stratification, with potential utility as a candidate tool for early screening and individualized clinical assessment pending further prospective validation. The emphasis on prospective validation is critical. Because the study is cross-sectional, it captures a snapshot in time and cannot establish that elevated UHR precedes the onset of hypertension. Self-reported physician-diagnosed hypertension also depends on access to healthcare and diagnostic practices, which may differ between the United States and China and could introduce measurement variability. The authors themselves frame the nomogram as a candidate tool requiring further validation before it can be recommended for routine clinical use.

Nevertheless, the practical appeal of UHR is hard to overstate. Both serum uric acid and HDL cholesterol are measured routinely in standard metabolic panels, so the ratio can be calculated from blood tests that millions of people already undergo each year, without additional cost, specialized equipment, or novel assays. If prospective studies confirm the relationship, UHR could serve as a low-cost screening marker that flags individuals who might benefit from closer blood pressure monitoring, lifestyle counseling, or earlier clinical evaluation. The integration of machine learning feature importance with traditional multivariate adjustment in this study also illustrates a methodological trend in modern epidemiology, in which exploratory algorithms help identify promising variables that are then subjected to rigorous confounder-adjusted testing, combining the pattern-detection strength of machine learning with the causal-inference discipline of classical statistics.

For the scientific community, the study adds to a growing body of evidence linking uric acid metabolism, lipoprotein biology, and cardiovascular risk into a unified metabolic picture. The balance between a pro-inflammatory, pro-oxidative metabolite and the antiatherogenic cholesterol fraction appears to carry information about vascular health that neither component conveys alone. As hypertension affects more than a billion people worldwide and remains a leading contributor to global disease burden, tools that enable earlier identification of elevated risk, even simple ones built from routine blood tests, could have substantial public health impact. The UHR nomogram from this study is not yet ready for the clinic, but it represents a concrete step toward individualized, data-driven hypertension screening that could eventually reach populations in both high-income and middle-income settings.

Subject of Research: hypertension, uric acid, HDL cholesterol, cardiovascular risk, nomogram, NHANES, CHARLS

Article Title: A cross-sectional study on the positive association between serum uric acid-to-high-density lipoprotein cholesterol ratio and self-reported physician-diagnosed hypertension with nomogram construction using NHANES and CHARLS data

Article References: A cross-sectional study on the positive association between serum uric acid-to-high-density lipoprotein cholesterol ratio and self-reported physician-diagnosed hypertension with nomogram construction using NHANES and CHARLS data. (n.d.). https://doi.org/10.1186/s12902-026-02654-8

Image Credits: AI Generated

DOI: 10.1186/s12902-026-02654-8

Keywords: cross-sectional, positive, association, between, serum, uric, acid-to-high-density, lipoprotein, cholesterol, ratio, self-reported, physician-diagnosed