Association of serum uric acid with all-cause and cardiovascular mortality in obstructive sleep apnea

The study investigated the association between Serum Uric Acid (SUA) levels and all-cause as well as cardiovascular mortality in patients with Obstructive Sleep Apnea (OSA). This prospective cohort study enrolled participants with OSA from four cycles of the National Health and Nutrition Examination Survey (NHANES) conducted between 2005 and 2008, and 2015–2018. A weighted Cox proportional hazards model was used to assess adjusted hazard ratios (aHRs) and their corresponding 95% confidence intervals (CI) for all-cause and cardiovascular mortality. Additionally, multivariable logistic regression and restricted cubic splines (RCS) models were employed to examine nonlinear relationships between SUA and all-cause and cardiovascular mortality. Among the 5,584 OSA participants included in the study, covering the four NHANES cycles and with a median follow-up of 4.333 years, a total of 537 deaths were observed, including 108 deaths attributed to cardiovascular disease. Comparing the fourth quartile (Q4) of uric acid levels, both the fifth quartile (Q5) (aHRs = 1.51, 95% CI [1.08, 2.12]) and the second quartile (Q2) (aHRs = 1.53, 95% CI [1.04, 2.25]) of uric acid levels were independently associated with an increased risk of all-cause mortality. Furthermore, comparing the fourth quartile (Q4) of uric acid levels, the second quartile (Q2) (aHRs = 2.40, 95% CI [1.08, 5.35]) of uric acid levels were independently associated with an increased risk of cardiovascular mortality. The RCS model demonstrated a U-shaped pattern in the association between SUA and all-cause mortality in OSA, with an inflection point observed at 5.83 mg/dl. The findings of this study suggest a U-shaped association between serum SUA levels and all-cause mortality and nonlinearity association between serum SUA levels and all-cause mortality. Further studies are warranted to determine the causal relationship between SUA levels and all-cause and cardiovascular mortality.

Obstructive sleep apnea (OSA) is a symptomatic, recurrent obstruction of the upper airway that occurs during sleep, resulting in reduced airflow or complete cessation 1 .Globally, 936 million people suffer from OSA, with 425 million experiencing moderate to severe OSA 2 .The incidence of mild OSA in the general adult population ranges from 9 to 38%, reaching as high as 84% in certain elderly populations.Severe OSA affects 6% to 17% of the total adult population and is more prevalent in the elderly group (36%) 3 .
Serum uric acid (SUA) is a degradation product of purine.Intriguingly, experiments have demonstrated the beneficial antioxidant properties of uric acid 4 .In epidemiological studies, elevated SUA levels are widely acknowledged as a risk factor for both macrovascular and microvascular complications 5 .
Patients with obstructive sleep apnea (OSA) endure chronic intermittent hypoxemia and metabolic disruptions resulting from upper airway obstruction, hypopnea, and apnea during sleep.This leads to increased serum uric acid (SUA) levels 6 .Hyperuricemia is frequently observed in patients with OSA 7 .The severity of OSA correlates positively with SUA levels 8 , which may explain the heightened vulnerability of OSA patients to cardiovascular complications [9][10][11][12] .Cardiovascular disease (CVD) stands as the predominant global cause of mortality 13 .Hyperuricemia often coexists with other cardiovascular risk factors, including dyslipidemia, obesity, kidney disease, and uncontrolled glucose regulation.Nevertheless, recent research has proposed that low SUA

Variables
The exposure variable in this study was SUA, while the outcome variables were all-cause mortality and cardiovascular mortality.These mortality data were collected from NHANES-linked National Death Index public access files through December 31, 2019.
OSA is ascertained through the responses to three dichotomous questions.These questions are as follows: (1) Do you snore for 3 or more nights per week?(2) Do you experience snorting, gasping, or stopping breathing for 3 or more nights per week?(3) Do you feel excessively sleepy during the day, ranging from 16 to 30 times per month, despite getting around 7 or more hours of sleep per night on weekdays or work nights?If individuals respond positively to any of these three questions, they are considered to exhibit symptoms indicative of OSA 23 .
Covariates were classified as demographics and medical conditions.Demographic information, including age, sex, marital status, education, and ethnicity, as well as medical conditions such as hypertension (defined as a history of physician-diagnosed hypertension, a measured average systolic blood pressure ≥ 140 mmHg, diastolic blood pressure ≥ 90 mmHg, or current use of antihypertensive medication), and DM (defined as a history of physician-diagnosed DM, hemoglobin A1c level ≥ 6.5%, fasting glucose (mmol/l) ≥ 7.0, random blood glucose

Nonlinear association of SUA with all-cause and CVD mortality
As shown in Table 2, after adjusting for age, sex, education status, BMI, ethnicity, smoking, eGFR, history of hypertension, DM, CKD, CVD, stroke, cancer, and hyperlipidemia, a U-shaped association was observed between SUA and all-cause mortality in OSA.The lowest aHRs were around the level of 5.83 mg/dl (P < 0.001 for nonlinearity).RCS models suggested that the inflection point for all-cause mortality (Fig. 2) was 5.83 mg/dl.On the left side of the inflection point, the aHRs for all-cause mortality were 0.87 (95% CI: 0.70, 1.07; p = 0.18) for every 1 mg/dl increase in SUA.On the right side of the inflection point, the aHRs for all-cause mortality were 1.29 (95% CI: 1.12, 1.49; p < 0.001).The RCS models revealed an inflection point for CVD mortality at 6.66 mg/dl (Fig. 3).On the left side of this inflection point, there was a significant inverse association between SUA levels and CVD mortality, with aHRs of 0.84 (95% CI: 0.72, 0.98; p = 0.032) for every 1 mg/dl increase in SUA.Conversely, on the right side of the inflection point, there was a positive association, with aHRs of 1.34 (95% CI: 1.14, 1.58; p < 0.001).All-cause and CVD mortality were observed before and after the inflection point in Table 3.
As depicted in the Kaplan-Meier survival curves, patients with OSA who had SUA levels of ≥ 6.8 mg/dl (Q5) exhibited significantly higher rates of all-cause mortality (Fig. 4) and CVD mortality (Fig. 5) during the follow-up period.
In the female group, our analysis revealed no association between SUA levels and all-cause mortality after adjusting for multiple variables (model 3), as shown in Table S1.Conversely, in the male group, we observed that both Q1 and Q5 SUA levels were associated with an increased risk of all-cause mortality after adjusting for multiple variables (model 3), as presented in Table S1.www.nature.com/scientificreports/ Utilizing RCS models, we determined that the inflection point for all-cause mortality in females was 5.277 mg/ dl (Fig. S1).Notably, there were no significant differences in female all-cause mortality before and after the inflection point, as demonstrated in Table S2.Conversely, for males, the RCS models indicated an inflection point of 6.219 mg/dl for all-cause mortality (Fig. S2).Substantial differences in male all-cause mortality were observed before and after the inflection point in Table S2.
In the elderly group, we observed that low (Q1, Q2) and high (Q5) SUA levels were associated with an increased risk of all-cause mortality after adjusting for multiple variables (model 3), as presented in Table S3.In contrast, in the younger group, there was no association between SUA levels and all-cause mortality after adjusting for multiple variables (model 3), as shown in Table S3.The RCS models indicated an inflection point of 6.32 mg/dl for all-cause mortality in the elderly (Fig. S3).On the left side of the inflection point, the aHRs for all-cause mortality were 0.74 (95% CI: 0.64, 0.86; p < 0.001) for every 1 mg/dl increase in SUA.On the right side of the inflection point, the aHRs for all-cause mortality were 1.20 (95% CI: 0.98, 1.48; p = 0.074).However, in the young group, a linear correlation was observed for all-cause mortality, as indicated by the RCS models (Fig. S4).www.nature.com/scientificreports/

Stratified analyses
Stratified analyses were conducted to assess the robustness of the relationship between SUA and all-cause mortality.In the adjusted model 3, the results of the stratified analysis demonstrated that the association between SUA and all-cause mortality remained consistent across most sub-populations, as presented in Table 4. Notably, a significant interaction was observed between SUA and the stratified variables, namely age and BMI.Specifically,   our findings revealed a stronger association between SUA and all-cause mortality among elderly individuals (≥ 60 years old) with OSA, as indicated by a distinct interaction test.However, no significant interaction test was observed for male individuals with OSA.

Discussion
In this prospective analysis, we examined the relationship between SUA and mortality in patients with OSA using data from the NHANES database.Our findings reveal a nonlinear association between SUA and the risk of all-cause death and cardiovascular death in the OSA population, as well as the identification of an inflection point.Both high and low levels of SUA were associated with increased risks of all-cause and cardiovascular mortality.After adjusting for demographics and disease variables, our analysis demonstrated a U-shaped curve for the association between SUA levels and the risk of all-cause and cardiovascular death, with patients having SUA levels of 5.9-6.799mg/dl serving as the reference group.OSA patients with higher SUA levels (≥ 6.8 mg/dl) exhibited a higher risk of all-cause mortality, while those with lower SUA levels (4.4-5.2 mg/dl) had an elevated risk of both all-cause and cardiovascular mortality compared to the reference group.Given the simplicity, convenience, and cost-effectiveness of measuring SUA, this study underscores the importance of managing SUA levels to help identify high-risk groups and improve the scientific management of OSA patients' uric acid levels, ultimately reducing mortality rates.
It is worth noting that the observed all-cause mortality risk in OSA patients aligns with previous studies on the mortality risk associated with SUA levels in other populations 5,18,25 .Our results demonstrate a U-shaped curve, indicating that both excessively high and low uric acid levels increase the risk of all-cause mortality in OSA patients.Specifically, we found a statistically significant difference when the uric acid level exceeded 5.83 mg/dl.For every 1 mg/dl increase in uric acid levels, the risk of all-cause mortality increased by 28.6% in OSA patients with high uric acid levels.Elevated SUA levels have also been associated with increased mortality risk in patients with liver disease 23 and DM 21 .Conversely, our results did not show a statistically significant effect on mortality in the OSA population when uric acid levels decreased below 5.83 mg/dl.Lower uric acid concentrations leading to increased mortality may be attributed to nutritional factors such as low body weight and muscle loss 26 .Considering that obesity is a major risk factor for obstructive sleep apnea 10 , it is possible that OSA patients are more prone to overnutrition rather than malnutrition, which may explain the lack of significance in patients with low-uric acid OSA.Conflicting results across studies may also be influenced by differences in cohort characteristics, sample size, and adjustment for confounders.Additionally, there is inconsistency in the reference values for SUA levels in previous studies 18 .
In general, men exhibited significantly higher levels of serum uric acid (SUA) compared to women.Previous studies have described SUA levels as an independent risk factor for mortality, irrespective of gender, while other studies have highlighted distinct disparities between men and women 16,18,27,28 .Within different gender groups, a non-linear relationship between uric acid concentration and mortality risk was observed, displaying a U-shaped pattern was similar to the findings reported by Cheng 28 .Specifically, men with OSA displayed a significant inflection point at 6.219 mg/dl, whereas women exhibited a U-shaped inflection point at 5.277 mg/ dl.Interestingly, there was no significant difference in uric acid concentration and mortality risk among women.However, Mazidi et al. 27 demonstrated that SUA influenced mortality in women aged ≥ 50, whereas Zhang et al. 29 established SUA as an independent risk factor for all-cause mortality in women of normal weight.The proportion of men and women among OSA patients differs, with men experiencing an incidence rate approximately three times higher than that of women 30 .Additionally, OSA patients generally exhibit higher uric acid levels compared to the general population 31 , with men having higher uric acid levels than women.Consequently, divergent data results between men and women among patients with OSA can be reasonably expected.
Age is another crucial factor to consider when examining the relationship between uric acid and health outcomes.Uric acid levels tend to increase with age, which can contribute to the development of age-related diseases such as gout and cardiovascular disease.Elderly individuals with high uric acid levels are more susceptible to  gout and comorbidities like hypertension and DM 32 .Our analysis revealed a J-shaped curve in the relationship between SUA and death among OSA patients aged 60 and above, with an inflection point at 6.324 mg/dl.Among OSA patients aged ≥ 60 with uric acid levels below 6.32 mg/dl, each 1 mg/dl increase in uric acid was associated with a 25% decrease in mortality rate.Uric acid levels ≥ 6.32 mg/dl indicated that the increase in high uric acid had little effect on mortality.Which was similar with the study by Tseng et al. 17 , low SUA levels increased the risk of all-cause and cardiovascular mortality, potentially influenced by nutritional status.There was no statistical difference in uric acid concentration and mortality risk among OSA patients younger than 60 years old, and the underlying mechanism for this observation remains unclear and requires further investigation.Uric acid is a complex substance that plays a significant role in the body.One theory suggests that it may promote inflammation, leading to cellular and tissue damage throughout the body and increasing the risk of chronic diseases and CVD.Another theory proposes that high uric acid levels contribute to the development of metabolic syndrome, which encompasses conditions such as obesity, hypertension, and insulin resistance, all of which are associated with increased mortality risk 33 .Numerous studies have elucidated the relationship between uric acid and mortality risk in other condition 5,[14][15][16][17][18][19] , demonstrating that both high and low uric acid levels are associated with increased mortality risk.When discussing serum uric acid, it is often associated with gout.Serum uric acid levels play a crucial role in diagnosing gout.Moreover, uric acid in urine is recognized as a cardiovascular disease (CVD) risk factor, as previous research suggests 34 .However, it is essential to acknowledge that uric acid levels can vary in different diseases or populations.The traditional cut-off values used to categorize uric acid levels as high or low are not necessarily effective in predicting disease-related mortality.Monitoring uric acid levels may serve as an important tool for predicting and preventing chronic diseases and premature death.
It is important to acknowledge the potential limitations of our study.Firstly, the diagnosis of OSA relied mainly on medical history and questionnaires, introducing potential biases, larger studies are needed to address these limitations.Secondly, the study participants were members of the United States population, so the generalizability of the results to other populations remains to be verified.Despite these limitations, our research possesses several strengths.Firstly, we employed RCS regression to explore the nonlinear relationship between SUA and OSA mortality risk, a method utilized for the first time in this context.Secondly, we conducted subgroup and sensitivity analyses for all-cause mortality data to assess effect sizes.Thirdly, we assessed death risk across the full range of SUA levels in the OSA population while considering both low and high SUA levels.Finally, in prior research, diuretics have been well-documented as factors associated with both elevated uric acid (UA) levels and increased mortality rates 35 .In this study, diuretics were included as an adjustment factor, which serves to mitigate potential biases in the results related to cardiovascular deaths.

Conclusions
The findings of this study suggest that a U-shaped association between serum SUA levels and all-cause mortality and nonlinearity association between serum SUA levels and all-cause mortality.Further studies are warranted to determine the causal relationship between SUA level and all-cause, CVD mortality.

Figure 1 .
Figure 1.Flow chart of the study population.

Figure 3 .
Figure 3. Multivariable-adjusted HRs for CVD mortality by uric acid level.

Figure 4 .
Figure 4. Kaplane-Meier survival analysis was used to evaluate between SUA and all-cause mortality.The p-value was obtained by log rank test.Log rank P < 0.0001.

Figure 5 .
Figure 5. Kaplane-Meier survival analysis was used to evaluate between SUA and all-cause mortality.The p-value was obtained by log rank test.Log rank P = 0.033.

Table 1 .
Baseline characteristics of study participants based on the SUA quintiles.Mean ± SEs for continuous variables: P value was calculated by weighted Student's t test.Number (%) for Categorical variables: P value was calculated by weighted chi-square test.Abbreviations: BMI, body mass index; CVD, cardiovascular disease; DM, diabetes mellitus; IFG, Impaired Fasting Glucose.IGT, Impaired Glucose Tolerance; CKD, chronic kidney disease; GFR, Glomerular filtration rate.RDW, Red cell distribution width.COPD, Chronic Obstructive Pulmonary Disease; HB, hemoglobin.* CKD-EPI, Chronic Kidney Disease Epidemiology Collaboration.

Table 4 .
Stratification analysis of SUA with all-cause mortality in OSA.Adjusted for age, sex, education status, BMI, eGFR, race, smoke, history of hypertension, DM, CKD, CVD, stroke, cancer and Hyperlipidemia, drug of diuretic.*p < 0.05.