Association between urinary albumin‐to‐creatinine ratio within normal range and hypertension among adults in the United States: Data from the NHANES 2009–2018

Abstract Background Recently it was suggested that urine albumin‐to‐creatinine ratio (uACR), even within the normal range, can be associated with hypertension, but only a few studies have examined. Therefore, this study aimed to determine the association between normal range uACR and the prevalence of hypertension. Methods The research used data from the 2009 to 2018 National Health and Nutrition Examination Survey, which included 14,919 participants. We defined the uACR as the amount of albumin (mg/dL) divided by creatinine (g/dL) in randomly voided urine. Hypertension was defined as mean systolic blood pressure ≥130 mmHg, or diastolic ≥80 mmHg, or were taking hypertension medication or were informed of a hypertension diagnosis by a physician/health professional. Results In multivariable‐adjusted models, per 5 mg/g uACR increment, the hypertension prevalence increased 1.31‐fold (OR, 1.31; 95% CI 1.23–1.40), the odds [95% confidence interval (CI)] for hypertension prevalence were 2.25 (1.86–2.72) for those in the highest quartile compared to those in the lowest quartile. The nonlinear relationship between hypertension prevalence and uACR was found by visually assessing images (p for nonlinearity<.001). In addition, in the subgroup analysis stratified by body mass index, the lower the BMI, the stronger the association between uACR and hypertension prevalence. Conclusions Even within the normal range, subtly elevated uACR was associated with an increased prevalence of hypertension in the USA general population, and this association may be enhanced in individuals with low BMI. Further research is needed to assess the clinical applicability of these findings.


| BACKGROUND
Chronic kidney disease (CKD) is a significant public health issue, 1 About 37 million adult Americans have CKD, according to the Centers for Disease Control and Prevention and prevalence projections indicate that they will rise as the population ages and the obesity and diabetes epidemics spread, 2,3 although end-stage renal disease (ESKD) development is a serious outcome of CKD, cardiovascular problems are mostly responsible for the high morbidity and mortality in these people. 4,5 Individualized hypertension control targets are an important intervention for CKD management in primary care. 6,7 In addition to controlling the risk factors currently known to be associated with blood pressure, we also need to focus on the role of other biomarkers in blood pressure. Albuminuria was closely associated with the risks of cardiovascular diseases (CVD) and several chronic diseases, [8][9][10] and changes in albuminuria observed in trials of cardiorenal preventive therapies strongly correlate with clinical endpoints. 11 The urinary albumin-to-creatinine ratio (uACR), which is a wellknown indicator of glomerular injury and a crucial diagnostic indicator of chronic kidney disease (CKD). 4 UACR is a reliable method for monitoring the excretion of urine protein and has become a clinical qualitative and quantitative diagnostic index for proteinuria that can replace the traditional 24-h urinary protein quantification. 12 Given the negative effects of albuminuria on CVD and the ambiguity of intervention when individual urinary protein levels rise slightly in the normal range (<30 mg/g), it is important to investigate the relationship between uACR and hypertension. Furthermore, fewer studies have examined the associations between uACR and hypertension, and no relevant research has been conducted in the NHANES population.
Therefore, we investigated the association between uACR and hypertension in US adults using data from the 2009 to 2018 National Health and Nutrition Examination Survey (NHANES). We also looked at whether the association varies by participant characteristics like age, gender, BMI, behavioral risk factor, and comorbidities, which could help with future clinical management of uACR.

| Study design and population
The Centers for Disease Control and Prevention (CDC) of America conducts the NHANES on a 2-year cycle, participants are noninstitutionalized individuals from the United States who are chosen through a complex stratified, multistage sampling design. In the current study, we used data from five NHANES cycles (2009-2010, 2011-2012, 2013-2014, 2015-2016, and 2017-2018) to investigate the association between uACR and blood pressure among NHANES participants. All information was obtained from the Public Data General Release file documents. The National Center for Health Statistics Ethics Review Board approved the NHANES, and all participants provided written informed consent before completing the NHANES. 13 In the present study, we analyzed NHANES participants (age ≥20 years) from five cycles of NHANES (2009)(2010)(2011)(2012)(2013)(2014)(2015)(2016)(2017)(2018), data on demographics, exams, lab tests, and questionnaires were gathered.
Participants who have incomplete data on laboratory urine data, blood pressure, and pregnant women were excluded, participants had missing demographic data and other covariates data also excluded.
After exclusions, this study contained a total of 14 919 participants aged ≥20 years with normal uACR (7605 female and 7314 male) (Figure 1).

| Urinary albumin-to-creatinine ratio
Trained researchers collected 5 mL of spotted urine from each participant and sent frozen urine samples (−20°C) to the laboratory.
The sample's stability was demonstrated at 5°C and temperatures less than or equal to −20°C. 14 The solid-phase fluorescent immunoassay was used to test urine albumin, while the enzymatic technique was used to quantify urine creatinine, in accordance with the NHANES's recommendations, the gold standard method was used to standardize and calibrate the amounts of urine albumin and creatinine. 15,16 uACR was calculated and reported in milligrams per gram (uACR = urine albumin/urine creatinine).

| Outcome ascertainment
The main outcome variable, blood pressure (BP), was measured with a mercury sphygmomanometer by trained personnel following standardized protocols, 17 following a 5-min quiet rest in the seated position, three BP readings were taken consecutively, and the participant's maximum inflation level was determined. If the blood pressure measurement is interrupted or incomplete, a fourth measurement can be taken.
All systolic and diastolic blood pressure measurements were taken at the Mobile Examination Center. In the present study, we calculated the average of systolic and diastolic BP readings for further analyses, if they had only one BP reading, then, it was the final record. Hypertension was defined as a mean SBP ≥130 mmHg, or a mean DBP ≥80 mmHg, or were taking hypertension medication or were informed of a hypertension diagnosis by a physician/health professional. nonsmoker; 0.05-10 ng/mL: exposed but not an active smoker; >10 ng/mL: active smoker). 18,19 The drinking status was divided into three categories: never (less than 12 drinks in a year), former (more than 12 drinks in a year but no longer drinking), and current (more than 12 drinks in a year but still drinking). 20 Physical activity status was classified as vigorous, moderate and inactive. 21 Diabetes was defined as being informed by doctor/health professional about the diagnosis of diabetes and/or a glycosylated hemoglobin measurement of ≥6.5%. 22

| Statistical analysis
The participants were characterized using descriptive statistics, where continuous variables were expressed as mean standard deviation and categorical variables as frequency or as a percentage.
The one-way ANOVA (normal distribution) and Kruscal-Wallis H (skewed distribution) tests were used to compare continuous variables between the various groups. The χ 2 test or Fisher's exact test was used to compare categorical variables between the various groups. Multivariate logistic regression analysis was used to evaluate blood pressure and hypertension based on the uACR (continuous and categorical variables). The covariate adjustment was determined by the following principle: when covariates were added to this model, the matched odds ratio changed by at least 10%. P for trend tests were performed by rerunning the corresponding regression models with the quartiles of uACR as a continuous variable. In addition, to characterize the shape of the relationship between uACR and the main outcome, a generalized additive model and fitted smoothing curve were used. The log-likelihood ratio test was used to determine   Abbreviations: BMI, body mass index; DM, diabetes mellitus; DP, diastolic blood pressure; eGFR, estimated glomerular filtration rate; PIR, poverty income ratio; SBP, systolic blood pressure; TC, total cholesterol; UA, uric acid; uACR, urinary albumin-to-creatinine ratio.
group. There was a trend for higher β of blood pressure and OR of hypertension among participants in the higher quartile of uACR relative to the lower quartile (p for trend <.001).
uACR and the prevalence of hypertension were positively correlated, according to the generalized additive model's results and fitted smoothing curve ( Figure 2). The nonlinear relationship between hypertension prevalence and uACR was found by visually assessing images (p for nonlinearity<.001), further analysis found that the inflection point of the threshold effect was 7.65 mg/g of uACR, below the threshold, the prevalence of hypertension increased significantly for each 1 mg/g increase in uACR (OR, 1.14; 95% CI 1.11-1.17) and then slowly increased above the threshold (OR, 1.04; 95% CI 1.02-1.05).
A stratified analysis of participants was performed to assess the relationship between uACR (continuous variable) and hypertension in various subgroups (Figure 3). An interesting finding was that the lower the BMI, the higher the OR of uACR to the incidence of hypertension (p for interactions = 0.012), other than BMI, age, gender, smoking status, drinking status, physical activity and diabetes did not significantly change the association between uACR and hypertension prevalence (p for all interactions >.05).

| DISCUSSION
The present study utilized the population-based cross-sectional resources of NHANES to detect the correlation between uACR levels and prevalence of hypertension in adults, which, to our knowledge, was the first time to examine such a relationship in this population. We found that uACR within the normal range was positively correlated with systolic and diastolic blood pressures and the prevalence of hypertension even after multivariable adjustment.
The nonlinear relationship between hypertension prevalence and uACR was found by visually assessing images. The optimal cutoff T A B L E 2 Weighted β/OR (95% confidence intervals) for relationship between uACR and blood pressure/hypertension in different models among US adult. Note: Values are weighted β/OR (95% CIs) unless otherwise indicated. uACR, albumin-to-creatinine ratio. Model 1 was adjusted for none. Model 2 was adjusted for age, sex, race, education, BMI, Energy, Sodium, UA, TC, eGFR. Model 3 was adjusted for age, sex, race, BMI, Energy, Sodium, UA, TC, eGFR, DM, alcohol intake, smoke exposure, total physical activity, lipoprotein-lowering drugs, antihypertensive drugs.
MING ET AL.
| 627 point of the association of uACR with hypertension was 7.65 mg/g, on each side of the optimal uACR value, the magnitude of the correlation between uACR and the prevalence of hypertension was different. Furthermore, the stratified analysis showed that the association between uACR levels with hypertension was much stronger among nonobese individuals compared with obese people.
Although F I G U R E 3 Stratified analyses by potential modifiers of the association between urine albumin-to-creatinine ratio and the prevalence of hypertension.
glomeruli are raised to make up for the kidney's lower glomerular filtration rate, which causes albuminuria to rise. 29 Elevated albuminuria might therefore in turn be a sign of reduced nephron numbers.
There has been growing evidence in recent years that urine albumin excretion, even when it is within the normal range, is linked to an increased risk of hypertension. The Nurse's Health Study 30  Finally, the average blood pressure that is determined by taking readings continuously for a short period of time might not accurately reflect the true state of blood pressure. In addition, genetic and environmental factors can be confounding factors that may affect the association results, we cannot ignore the influence of these factors.

| CONCLUSION
In conclusion, our results suggest that even within the normal range, subtly elevated uACR was associated with an increased prevalence of hypertension in the USA general population. For individuals with high uACR, closer monitoring of blood pressure is highly recommended.
However, whether uACR is a good predictor of hypertension, especially in populations with different BMI, needs further studies.

ACKNOWLEDGMENTS
We want to thank all patients who participated in this study. The study was conducted without any external funding and was driven by the investigators' scientific interest and collaboration.

CONFLICT OF INTEREST STATEMENT
The authors declare no conflict of interest.

DATA AVAILABILITY STATEMENT
The data sets generated and analyzed during the current study are available in the nhanes repository, (https://wwwn.cdc.gov/nchs/ nhanes/Default.aspx).