1. Introduction

Uric acid and urate are common molecules in the bloodstream that are generated through the intake of exogenous dietary sources or endogenous nucleic acid synthesis and breakdown, with purines undergoing metabolism by xanthine dehydrogenase and xanthine oxidase.[1] In contrast, hyperuricemia results from an imbalance between uric acid production and excretion. Currently, hyperuricemia can be categorized into 3 main types: overproduction of uric acid, underexcretion of uric acid, and a combination of both, with underexcretion being the predominant type, accounting for approximately 60% of cases.[2] The increase in the incidence of hyperuricemia has become a global public health issue. In China, the prevalence of hyperuricemia has increased from 11% to 14%.[3] Similarly, in the United States, between 2007 and 2016, the prevalence of hyperuricemia increased from 14.6% to 15.9%,[4] imposing substantial burdens on people lives and society. Despite being the end product of metabolic processes, uric acid exerts distinct biological effects and can cause renal damage through various mechanisms, including the induction of urate deposition, oxidative stress, inflammation, endothelial dysfunction, and renal fibrosis.[5–9] Moreover, numerous investigations have proposed an association between hyperuricemia and an increased risk of hypertension, diabetes, and cardiovascular diseases.[10–12]

The monocyte-to-high-density lipoprotein ratio (MHR), a recently identified predictive marker, involves both monocytes and high-density lipoprotein. It is closely correlated with the inflammatory and oxidative stress levels in the body. Monocytes play a crucial role in innate immunity, and when activated, they generate a variety of mediators with both inflammatory and oxidative properties. They trigger the body inflammatory response, endothelial dysfunction, and thrombus formation through interactions between the vascular endothelium and platelets.[13] Conversely, high-density lipoproteins perform a diverse range of functions to maintain the normal functioning of the body. In addition to maintaining cholesterol homeostasis, studies have shown that high-density lipoprotein can bind to cholesterol crystals, reduce lysosomal rupture, suppress nuclear factor-kappa B signaling, and reduce the levels of interleukin-1 betaβ and nod-like receptor family, pyrin domain containing 3.[14] Monosodium urate crystals are critical for inducing acute inflammatory responses and triggering gout. When monosodium urate crystals and high-density lipoprotein are simultaneously injected into the backs of mice, they effectively reduce neutrophil infiltration and expression of other pro-inflammatory factors.[15]

Elevated serum uric acid levels are closely linked to inflammation, and monosodium urate crystals can induce neutrophil and macrophage infiltration. Upon engulfing the crystals, these phagocytic cells undergo lysosomal rupture, resulting in potassium efflux and release of tissue proteases, thus activating the nod-like receptor family, pyrin domain containing 3 inflammasome.[16] Moreover, epidemiological research has suggested a specific association between inflammation and increased serum uric acid concentrations.[17] Several studies have highlighted the potential of MHR as a predictive biomarker for diverse metabolic disorders, such as metabolic syndrome, diabetic retinopathy, and diabetic nephropathy.[18–21] However, research exploring the latent association between MHR and hyperuricemia remains limited. Hence, this study aimed to further elucidate the potential association between MHR and hyperuricemia by expanding the sample size, altering the study population, and utilizing data collected from the National Health and Nutrition Examination Survey (NHANES) between 2009 and 2018.

2. Methods 2.1. Research population

The NHANES, conducted by the National Center for Health Statistics in the United States, is an extensive cross-sectional survey. This ensured the representativeness of the study population through a multistage, complex, random sampling method. All participants provided informed consent, and the study was approved by the NCHS Research Ethics Review Board. This study involved a population spanning 5 survey cycles from 2009 to 2018, comprising a total of 49,693 participants. The exclusion criteria were as follows: individuals below 18 years of age (n = 19,341); lack of information on education, PIR, body mass index (BMI), hypertension, UACR, estimated glomerular filtration rate (eGFR), diabetes, smoking, alcohol consumption, heart failure, coronary heart disease, physical activity and stroke (n = 12,558); missing uric acid data (n = 9); missing data on monocytes and high-density lipoprotein cholesterol (HDL-C) (n = 72); outliers with MHR > 2 (n = 19). Ultimately, 17,694 participants were included, with 14,182 classified as normal and 3512 classified as patients with hyperuricemia, as shown in Figure 1.

Figure 1.:

The study population flowchart. NHANES (National Health and Nutrition Examination Survey), PIR (poverty income ratio), BMI (body mass index), UACR (urine albumin-to-creatinine ratio), eGFR (estimated glomerular filtration rate), DM (diabetes mellitus), HF (heart failure), CAD (coronary artery disease), HDL-C (high-density lipoprotein cholesterol), MHR (monocyte-to-high-density lipoprotein ratio).

2.2. Exposure and outcome

Blood sample collection and processing followed by the NHANES laboratory or medical technology expert. HDL-C levels were determined using direct immunoassay or the precipitation method,[22] and MHR was computed by taking the ratio of the monocyte count (103 cells/µL) to the level of high-density lipoprotein (mmol/L).[23] Serum uric acid levels equal to or exceeding 7 mg/dL in males and 6 mg/dL in females were used as criteria for defining hyperuricemia.[24]

2.3. Covariates

Through standardized questionnaires, participants provided demographic information, including gender, age (categorized as 18–39, 40–59, 60–80), race, education level, and family income to poverty ratio (PIR). Smoking habits were assessed (categorized as never, former, and current), as were alcohol consumption habits (classified as never, moderate, heavy, and binge). For males, moderate drinking was defined as consuming 1 to 2 drinks per day, heavy drinking was defined as consuming 3 to 4 drinks per day, and binge drinking was defined as consuming 5 or more drinks per day. In the case of females, moderate drinking was set at 1 drink per day, heavy drinking was considered to be 2 to 3 to drinks per day, and binge drinking was identified as consuming 4 or more drinks per day. BMI was measured and categorized as < 25, 25 to 30, or > 30 kg/m². The CKD-EPI creatinine equation was employed to calculate the eGFR,[25] which was classified into 2 categories, <60 or ≥ 60 mL/minute/1.73 m2, guided by the diagnostic standards of chronic kidney disease (CKD) and insights from prior research.[26–28] The UACR classification was based on a threshold of 30 mg/g. The determination of whether patients have had a stroke, heart failure, or coronary artery disease (CAD) was based on self-reported information obtained through surveys. The diagnosis of hypertension was based on self-reported information regarding the use of antihypertensive medications or measurements indicating blood pressure equal to or exceeding 140/90 mm Hg, as reported in surveys. Diabetes was determined based on survey reports, fasting blood glucose (≥126 mg/dL), glycated hemoglobin (≥6.5%), the use of antidiabetic medications, or a documented history of diabetes. The level of physical activity was represented in Metabolic Equivalent of Task (MET) minutes, referring to the Physical Activity Questionnaire (PAQ) section of the NHANES database. Weekly MET minutes were computed based on this questionnaire, and individuals were subsequently grouped according to adult physical activity guidelines,[29] specifically based on whether they achieved ≥ 600 MET minutes per week.

2.4. Statistical examination

Statistical analyses were performed using SPSS 26.0, EmpowerStats 4.1, and Stata 16, and statistical significance was set at P < .05. The examinations were adjusted using weights as outlined in the NHANES guidelines, taking into consideration the 10-year data period and primary focus on blood samples. To create a weighted estimate, we referred to the “WTMEC2YR” weight variable and sampled 1-fifth of the 2-year weights from 2009 to 2018 for each individual. Initially, univariate analyses were performed to compare baseline characteristics among the groups. Nominal variables are depicted as proportions (%) and quantitative variables are presented as mean values and standard deviations. Appropriate statistical tests, such as t-tests or chi-squared tests, were employed, depending on the nature of the data. Subsequently, MHR was categorized into quartiles, and its association with the risk of hyperuricemia was investigated using a multivariable logistic regression model. The lowest quartile was defined as the reference group. Model 1 was unadjusted, Model 2 was adjusted for demographic factors (gender, age, race, education, and PIR), and Model 3 included additional adjustments for health-related variables (BMI, UACR, hypertension, diabetes, eGFR, smoking, alcohol use, coronary artery disease, heart failure, physical activity, and stroke). Subgroup analyses were conducted to investigate how MHR correlates with the prevalence of hyperuricemia across different subgroups, stratified by factors such as age, gender, race, PIR, BMI, eGFR, diabetes, hypertension, UACR, and physical activity. To investigate the potential nonlinear links between MHR and hyperuricemia prevalence, we utilized a smooth curve fitting method to identify threshold effects and inflection points. Two-piecewise logistic regression models were used to assess differences at various threshold points.

3. Result 3.1. Initial characteristics

The baseline characteristics of 17,694 participants are summarized in Table 1. Among them, 14,182 were nonhyperuricemic, and 3512 presented with hyperuricemia. In comparison to nonhyperuricemic participants, those with hyperuricemia were predominantly male and older, with a higher proportion in the 60–80 age group. They exhibited higher MHR and BMI. Additionally, their eGFR levels were lower, and there were a greater number of individuals with hypertension (P < .05).

Table 1 - Baseline characteristics of the study participants. Parameters Total (N = 17694) Nonhyperuricemia (N = 14182) Hyperuricemia (N = 3512) P Age (yr), % 48.16 ± 17.43 47.15 ± 17.21 52.25 ± 17.72 <.001  18–39 6380 (36.06) 5382 (37.95) 998 (28.42)  40–59 5936 (33.55) 4833 (34.08) 1103 (31.41)  60–80 5378 (30.39) 3967 (27.97) 1411 (40.18) Gender, % <.001  Male 8683 (49.07) 6700 (47.24) 1983 (56.46)  Female 9011 (50.93) 7482 (52.76) 1529 (43.54) Race, % <.001  Mexican American 2486 (14.05) 2107 (14.86) 379 (10.79)  Other Hispanic 1748 (9.88) 1476 (10.41) 272 (7.74)  NonHispanic White 7447 (42.09) 5915 (41.71) 1532 (43.62)  NonHispanic Black 3576 (20.21) 2740 (19.32) 836 (23.80)  Other race 2437 (13.77) 1944 (13.70) 493 (14.04) Education, % .233  High school and below 7395 (41.79) 5896 (41.57) 1499 (42.68)  Above high school 10,299 (58.21) 8286 (58.43) 2013 (57.32) PIR, % .612  Below 1.3 5343 (30.20) 4299 (30.31) 1044 (29.73)  1.3–3.5 6646 (37.56) 5302 (37.39) 1344 (38.27)  Over 3.5 5705 (32.24) 4581 (32.30) 1124 (32.00) BMI (kg/m2), % 29.26 ± 7.00 28.45 ± 6.55 32.52 ± 7.79 <.001  BMI < 25 5108 (28.87) 4636 (32.69) 472 (13.44)  25 ≤ BMI < 30 5769 (32.60) 4710 (33.21) 1059 (30.15)  BMI ≥ 30 6817 (38.53) 4836 (34.10) 1981 (56.41) Hypertension, % <.001  No 10,442 (59.01) 9041 (63.75) 1401 (39.89)  Yes 7252 (40.99) 5141 (36.25) 2111 (60.11) UACR (mg/g), % <.001  <30 15,747 (89.00) 12,836 (90.51) 2911 (82.89)  ≥30 1947 (11.00) 1346 (9.49) 601 (17.11) eGFR (mL/min/1.73 m2), % 96.19 ± 22.04 98.97 ± 20.20 84.96 ± 25.35 <.001  <60 1147 (6.48) 545 (3.84) 602 (17.14)  ≥60 16,547 (93.52) 13,637 (96.16) 2910 (82.86) DM, % <.001  No 14,748 (83.35) 12,067 (85.09) 2681 (76.34)  Yes 2946 (16.65) 2115 (14.91) 831 (23.66) Smoke, % .053  Never 10,219 (57.75) 8299 (58.52) 1920 (54.67)  Former 3983 (22.51) 3003 (21.17) 980 (27.90)  Now 3492 (19.74) 2880 (20.31) 612 (17.43) Alcohol use, % .620  Never 3497 (19.76) 2770 (19.53) 727 (20.70)  Moderate 6984 (39.47) 5615 (39.59) 1369 (38.98)  Heavy 4748 (26.83) 3859 (27.21) 889 (25.31)  Binge 2465 (13.93) 1938 (13.67) 527 (15.01) HF, % <.001  No 17,237 (97.42) 13,927 (98.20) 3310 (94.25)  Yes 457 (2.58) 255 (1.80) 202 (5.75) CAD, % <.001  No 17,062 (96.43) 13,743 (96.90) 3319 (94.50)  Yes 632 (3.57) 439 (3.10) 193 (5.50) Stroke, % <.001  No 17,165 (97.01) 13,808 (97.36) 3357 (95.59)  Yes 529 (2.99) 374 (2.64) 155 (4.41) ≥600 MET min/wk, % <.001  No 6488 (36.67) 5063 (35.70) 1425 (40.58)  Yes 11,206 (63.33) 9119 (64.30) 2087 (59.42) MHR 0.44 ± 0.22 0.43 ± 0.21 0.50 ± 0.24 <.001

Results are expressed as mean ± standardized differences or as counts and percentages.

BMI = body mass index, CAD = coronary artery disease, DM = diabetes mellitus, eGFR = estimated glomerular filtration rate, HF = heart failure, MHR = monocyte-to-high-density lipoprotein ratio, PIR = ratio of family income to poverty, UACR = urine albumin-to-creatinine ratio

3.2. The correlation between hyperuricemia and MHR

Table 2 displays the outcomes of the weighted multivariate logistic regression analysis investigating the connection between MHR and hyperuricemia. Compared to the lowest MHR quartile, all models (except for the second quartile in Model 3) showed a statistically significant positive correlation with hyperuricemia (P < .05). The risk of hyperuricemia increased with increasing MHR. Additionally, there may be a linear trend in hyperuricemia prevalence across MHR quartiles (P for trend < .001).

Table 2 - Association between MHR quartiles and the hyperuricemia in participants. Model Model 1: OR (95% CI) P value Model 2: OR (95% CI) P value Model 3: OR (95% CI) P value MHR 4.50 (3.66–5.54)
<.001 4.03 (3.22–5.05)
<.001 1.98 (1.54–2.54)
<.001 MHR (Quartile) Q1 (0.04–0.29) Reference Reference Reference Q2 (0.29–0.40) 1.39 (1.17–1.64)
<.001 1.38 (1.16–1.62)
<.001 1.11 (0.94–1.33)
.219 Q3 (0.40–0.55) 1.95 (1.66–2.28)
<.001 1.89 (1.61–2.21)
<.001 1.40 (1.18–1.65)
<.001 Q4 (0.55–2.00) 2.76 (2.37–3.21)
<.001 2.61 (2.23–3.06)
<.001 1.62 (1.37–1.92)
<.001 P for trend <.001 <.001 <.001

Model 1: No covariate adjustments. Model 2: Demographic factors, such as age, race, gender, education, and PIR, were taken into account for adjustments. Model 3: Comprehensive adjustments included age, race, gender, education, PIR, hypertension, BMI, UACR, eGFR, DM, smoking, alcohol use, heart failure, coronary artery disease, physical activity, and stroke.

MHR = monocyte-to-high-density lipoprotein ratio.

3.3. Subgroup analysis

To ensure the robustness of the link between MHR and hyperuricemia across diverse population subgroups, stratified analyses were conducted along with Model 3, as outlined in Table 3. When stratified by age, gender, PIR, diabetes mellitus, hypertension, and physical activity, the findings indicated a significant positive correlation between MHR and hyperuricemia (P < .05). In the race, BMI, eGFR, and UACR-stratified analysis, a more pronounced positive association between MHR and hyperuricemia was observed in individuals with eGFR ≥ 60 mL/minute/1.73 m², BMI ≥ 25 kg/m², UACR < 30 mg/g, and participants identified as nonOther Hispanic (P < .05).

Table 3 - Subgroup analysis of the association between MHR and hyperuricemia. Character OR (95%CI) P value P for interaction Age <.0001  18–39 2.36 (1.59–3.50) <.001  40–59 1.63 (1.03–2.59) .037  60–80 2.05 (1.31–3.20) .002 Gender <.0001  Male 1.66 (1.23–2.24) .001  Female 3.20 (2.09–4.88) <.001 Race .1422  Mexican American 3.26 (1.71–6.21) <.001  Other Hispanic 2.08 (0.98–4.40) .055  NonHispanic White 1.85 (1.32–2.58) <.001  NonHispanic Black 1.80 (1.09–2.99) .022  Other race 2.56 (1.36–4.83) .004 PIR .8453  Below 1.3 2.63 (1.75–3.96) <.001  1.3–3.5 1.88 (1.29–2.75) .001  Over 3.5 1.79 (1.13–2.85) .014 BMI .8691  BMI < 25 1.87 (0.92–3.82) .084  25 ≤ BMI < 30 1.87 (1.22–2.86) .004  BMI ≥ 30 2.08 (1.48–2.91) <.001 eGFR .0052  <60 1,49 (0.69–3.24) .312  ≥60 2.04 (1.57–2.65) <.001 DM <.0001  No 2.04 (1.54–2.70) <.001  Yes 1.97 (1.14–3.40) .016 Hypertension <.0001  No 2.54 (1.79–3.61) <.001  Yes 1.66 (1.19–2.32) .003 UACR <.0001  <30 2.16 (1.65–2.82) <.001  ≥30 1.21 (0.62–2.37) .573 ≥600 MET min/wk .022  No 1.78 (1.17–2.71) .007  Yes 2.20 (1.60–3.01) <.001

Subgroup analysis of the association between MHR and hyperuricemia. Subgroup analysis was conducted using weighted multivariable logistic regression.

BMI = body mass index, DM = diabetes mellitus, eGFR = estimated glomerular filtration rate, PIR = ratio of family income to poverty, UACR = urine albumin-to-creatinine ratio.

3.4. Nonlinear association between MHR and hyperuricemia

Figure 2 shows the results of the smoothed curve adjustment. Using a logistic regression model with 2 segments, the inflection point for MHR was identified as 0.826. When the MHR was ≤ 0.826, a notable positive association with the prevalence of hyperuricemia was observed, with an OR of 3.10 (95% CI: 2.15–4.49; P < .001), while no significant association was found beyond 0.826. Furthermore, in subgroups with eGFR ≥ 60 mL/minute/1.73 m², females, overweight and obese and nonHispanic white individuals, distinct inflection points for MHR were observed at 0.795, 0.246, 0.694, 0.86 and 0.777, respectively (Table 4).

Table 4 - Two-piecewise logistic regression model analysis results of the threshold relationship between MHR and hyperuricemia. Model OR (95%CI) P value  Inflection point (K) 0.826  MHR ≤ 0.826 3.10 (2.15–4.49) <.001  MHR > 0.826 0.72 (0.29–1.79) .481  Log-likelihood ratio <0.001 eGFR ≥ 60(mL/min/1.73 m2)  Inflection point (K) 0.795  MHR ≤ 0.795 3.19 (2.11–4.82) <.001  MHR > 0.795 0.95 (0.41–2.20) .909  Log-likelihood ratio <0.001