The Negative Association Between Non-alcoholic Fatty Liver Disease Severity and Gouty Nephropathy in a Non-diabetic Gouty Population: A Cross-Sectional Study

Background: Nonalcoholic fatty liver (NAFLD) and chronic kidney disease (CKD) share common pathogenic mechanisms and risk factors, but the specic relationship between NAFLD and gouty nephropathy has not been well understood. We aim to evaluate the association between NAFLD and gouty nephropathy in a non-diabetic gouty population. Methods: The retrospective cross ‐ sectional study was performed on 1049 non-diabetic gouty participants, who were hospitalized between 2014 and 2020, across 4 districts in Shandong, China. Demographic and clinical characteristics of the study population were collected. The odds ratios (OR) and corresponding 95% condence intervals (CI) in relation to the NAFLD severity determined by ultrasonography were obtained by multiple logistic regression analysis. Results: An unexpectedly inverse relationship was found between NAFLD severity and the risk of gout nephropathy. Multivariate logistic regression analysis demonstrated that higher degree of NAFLD severity is independently associated with lower risk of gouty nephropathy, after adjusted for age, sex, smoking, gout duration, and metabolic risk factors including obesity, hypertension, hyperglycemia, hyperuriacemia and dyslipidemia, with OR 0.392 (95 % CI 0.248–0.619, P < 0.001), 0.379 (95 % CI 0.233–0.616, P < 0.001) and 0.148 (95 % CI 0.043–0.512, P = 0.003) in participants with mild, moderate, and severe NAFLD, respectively, compared to those without NAFLD. We also observed a weakened association of serum uric acid (SUA) with metabolic risk factors and NAFLD under circumstances of gouty nephropathy (r = -0.054, P = 0.466). Conclusions: The presence and severity of NAFLD was negatively associated with the risk of gouty nephropathy in the non-diabetic gouty populations. Further investigation of NAFLD will provide insights into the pathogenesis of gouty nephropathy. Ccr, Cystatin estimated glomerular ltration assessment index of insulin resistance; LDL-C, SCr, creatinine; circumference.

receptor protein 3 in ammasomes in macrophages, and promoting chemokine secretion in proximal tubular cells [11], hyperuricaemia constantly aggravates the renal damage. However, the clinical risk factors for gouty nephropathy remain controversial and are insu ciently studied at present.
The relationship between nonalcoholic fatty liver disease (NAFLD) and chronic kidney disease (CKD) has attracted much attention, since they share similar risk factors and pathogenic mechanisms, such as insulin resistance (IR), diabetes mellitus (DM), hyperlipidemia and obesity [12][13][14][15][16][17][18]. While numbers of previous studies indicated the presence and severity of NAFLD was a potential contributory risk factor for the development and progression of CKD, recent studies suggested no adverse association between NAFLD hepatic steatosis and renal function both in general populations and diabetic individuals [19][20][21][22]. Some researchers argued that NAFLD itself is not an independent risk factor for CKD, while it is the comorbidities of NAFLD such as obesity, hypertension, and hyperuricemia that are independently associated with CKD [23]. Some also argue that NAFLD was associated with early stages of CKD, but not the late stages of CKD [22]. In addition, although it is reasonable to assume that NAFLD may promote renal damage, it is still uncertain if NAFLD is associated with a speci c type of kidney disease [16]. Given that most of the studies were focused on patients with DM [24,25], the speci c relationship between NAFLD and gouty nephropathy has been poorly studied.
To bridge the evidence gap between NAFLD and gouty nephropathy, we conducted a hospital-based cross-sectional study in Shandong, China. We aimed to investigate whether the probable advanced NAFLD evaluated by ultrasound is independently associates with gouty nephropathy in non-diabetic gouty populations, the results of which may provide a novel theoretical basis and therapeutic target for the prevention and treatment of gouty nephropathy.

Design and study participants
We reported the retrospective cross-sectional study according to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement [26]. Data were collected from 1545 nondiabetic gouty patients hospitalized between 2014 and 2020, across 4 districts in Shandong, China, by a trained staff group. All patients were diagnosed with the American College of Rheumatology/European League Against Rheumatism (ACR/EULAR) gout diagnostic criteria [27]. The diagnosis of gouty nephropathy was based on the diagnosis of primary gout [28], with one or more of the following parameters: Urinary protein > 150 mg/dl; urine white blood cells > 5/high power eld (HPF); urine red blood cells > 3/HPF; serum creatinine > 115 μmol/l; blood uric acid/creatinine ratio > 2.5. Additionally, an ultrasound or ureterography revealing renal calculus or a shrunken kidney was also considered. The exclusion criteria include the following: (1) age < 18 years or > 80 years; (2) with secondary hyperuricaemia; (3) receiving medical treatment for current chronic glomerulonephritis other than gouty nephropathy; (4) with DM, severe CVDs, heart failure, or cancer; (5) with ethanol intake per week that is more than 140 g in men and 70 g in women[29]; (6) with speci c diseases that could result in fatty liver, such as viral hepatitis and drug-induced liver disease; (7) with positive for hepatitis B or C viruses other types of liver diseases, including primary biliary cirrhosis, autoimmune hepatitis [30]. Patients with missing information on severity of NAFLD, or inadequate variables to calculate eGFR were also excluded.
The protocol was designed according to the Declaration of Helsinki and was approved by the ethics committee of the Qingdao University a liated hospital, and all participants provided written informed consent. The study was registered on http://www.chictr.org.cn/ under number ChiCTR2000035185.
Height and weight were measured with patients standing without shoes and with lightweight clothing. WC was measured in the horizontal plane midway between the lowest ribs and the iliac crest, as suggested by the World Health Organization and the International Diabetes Federation. Body mass index (BMI) was calculated as weight divided by height squared (kg/m 2 ), and BMI > 29.9 was used to de ne obesity [31]. Blood pressure was reported as the means of three consecutive measurements with an interval of ve minutes, and hypertension was de ned by systolic blood pressure (SBP) ≥ 140mmHg, or diastolic blood pressure (DBP) ≥ 90mmHg, or self-reported previous diagnosis of hypertension by physicians.
Blood samples were obtained between 6:00-9:00 a.m. after fasting for at least 8 h. Venipuncture was performed in the median cubital vein, and centrifugation and dispensing were completed within 1 h. All samples were cold-chained, stored and transported to a central laboratory for testing within 2-4 h.

Statistical analysis
The SPSS version 22.0 software (SPSS IBM Corporation, Armonk, NY, USA) and SATA (version 12.0, StataCorp, Lakeway Dr., TX, USA) was used to perform statistical analyses. Normally distributed continuous variables are presented as mean ± standard deviation (SD), while non-normally distributed continuous variable are presented as median (interquartile range, IQR), and categorical variables are presented as frequency (%). We compared different groups by Chi-square and one-way ANOVA on normal and continuous variables, respectively. When observe values did not approximate to the normal distribution, the Kruskal-Wallis test were replaced. The correlations between the levels of SUA and risk factor variables were calculated with the Spearman's correlation test. The odds ratios (OR) and the corresponding 95% con dence intervals (CI) for gouty nephropathy concerning NAFLD severity were obtained by multiple logistic regression analysis. All statistical analyses were two-sided, and p-values less than 0.05 were considered to be statistically signi cant.

Results
We initially enrolled 1545 individuals and a total of 1049 were in nal analysis after exclusion criteria ( Fig. 1). We excluded participants who were missing laboratory results (n =32) and ultrasound results (62), had a history of DM (129), severe CVDs or cancer (17), excessive consumption (male ≥ 140 g/week, female ≥ 70 g/week) of pure alcohol (n = 117), self-reported viral hepatitis (including hepatitis B and hepatitis C virus) (n = 53), was using medications associated with secondary NAFLD or autoimmune liver disease (n = 21), and was considered as secondary hyperuricaemia (8).

Baseline Characteristics of participants with gout
Clinical characteristics of the individuals strati ed by the presence of gouty nephropathy were shown in Table 1, and strati ed by the severity of NAFLD in Table 2. As shown in Table 1, among the 1049 hospitalized non-diabetic gouty patients, 17.4% were found to have gouty nephropathy. Participants with gouty nephropathy were older and had longer gout duration, more prone to be smokers, with lower BMI, higher SBP, higher FPG and SUA levels, than those without gouty nephropathy. Of note, participants with gouty nephropathy were more likely to have lower prevalence and severity of NAFLD by ultrasound, with lower levels of ALT and AST, than those without gouty nephropathy (P < 0.001). Meanwhile, insigni cant differences in sex, DBP, HOMA-IR, and lipid pro les (including LDL-c, HDL-c, TG, TC and FFA) between the groups were detected.
The trend analysis of demographic and biochemical features among different categories of NAFLD severity was presented in Table 2. As we can see, participants with higher severity of NAFLD seem to be younger, more likely to be smokers, with higher BMI and abdominal obesity, higher FPG and HOMA-IR, higher levels of SUA, LDL-c, TG, TC and FFA, and lower levels of HDL-c, than those with less severe or without NAFLD. Of note, preserved renal function were observed in individuals with more severe NAFLD, re ected by lower prevalence of gouty nephropathy, lower levels of indicators of renal damage (SCr, BUN and Cys C), as well as higher levels of eGFR and Ccr, compared to those with less severe or without NAFLD (P < 0.001).
Association of ultrasound-diagnosed NAFLD with gouty nephropathy We further determined the renal function based on eGFR and Ccr levels with respect to the degree of NAFLD severity by box plot (Fig. 2 A and B). As we can see, as the severity of NAFLD increased, the levels of eGFR and Ccr were higher (P < 0.001), with the lowest eGFR levels seen in individuals with non-NAFLD and highest eGFR levels seen in those with severe NAFLD. Box plots were also performed to demonstrated the relationship between liver function and eGFR categories ( Fig. 2 C and D). Consistently, with the decline of eGFR in gouty participants, we observed that the levels of liver enzymes including ALT and AST decreased (P < 0.001). Table 3, the risk of gouty nephropathy was signi cantly associated with NAFLD severity in the logistic regression analyses. In unadjusted analysis, the OR was 0.336 (95 % CI 0.218-0.517, P < 0.001), 0.320 (95 % CI 0.205-0.510, P < 0.001) and 0.321 (95 % CI 0.218-0.517, P = 0.001), respectively, in participants with mild, moderate, and severe NAFLD compared to those without NAFLD. On further adjustment for age, sex and smoking, NAFLD severity was still signi cantly associated with the risk of gouty nephropathy, with OR 0.394 (95 % CI 0.254-0.612, P < 0.001), 0.388 (95 % CI 0.245-0.614, P < 0.001) and 0.152 (95 % CI 0.046-0.499, P = 0.002) in participants with mild, moderate, and severe NAFLD compared to those without NAFLD. When further adjusted for metabolic parameters, including obesity, hypertension, hyperglycemia, hyperuricemia, and hyperlipidemia, the association of NAFLD severity with gouty nephropathy was still signi cant, with OR 0.392 (95 % CI 0.248-0.619, P < 0.001), 0.379 (95 % CI 0.233-0.616, P < 0.001) and 0.148 (95 % CI 0.043-0.512, P = 0.003) in participants with mild, moderate, and severe NAFLD, respectively, compared to those without NAFLD.

Risk factors for gouty nephropathy in the multiple logistic regression analysis
The complex nature of the variables that determine the gouty nephropathy risk was further studied in a forest plot. As shown in Fig. 3, gouty nephropathy was introduced as a dependent variable in the multiple factors logistic regression analysis, using old age (more than 60 years), long gout duration (more than 10 years), smoking, obesity, hypertension, hyperglycemia, hyperuricemia, high LDL-c, low HDL-c, high TG and high TC level, as well as NAFLD severity as independent variables. In addition to the severity of NAFLD, we observed the old age, hypertension, and hyperglycemia were signi cant risk factors for gouty nephropathy. However, sex, long gout duration, smoking, obesity, hyperuricemia, and dyslipidemia were insigni cantly associated with gouty nephropathy in the non-diabetic gouty participants.
Association of SUA with metabolic syndrome and NAFLD in participants with gout In the present study, we also examined the association between SUA and both metabolic syndrome and NAFLD severity in the gouty population. To detect different clinical signi cance of hyperuricaemia for metabolic disorders in patients with and without renal insu ciency, we divided the participants into two groups based on the presence of gouty nephropathy, and separately analyzed the association of SUA with both metabolism-related parameters and NAFLD severity. The Spearman correlation coe cients between SUA and metabolism-related parameters were given in Table 4. SUA was signi cantly associated with BMI (r = 0.200, P < 0.001), WC (r = 0.187, P < 0.001), DBP (r = 0.098, P = 0.004), HOMA-IR (r = 0.131, P = 0.004), LDL-c (r = 0.151, P < 0.001), TG (r = 0.230, P < 0.001), TC (r = 0.160, P < 0.001) and degree of NAFLD severity (r = 0.240, P < 0.001) in participants without gouty nephropathy. However, this association weakened and became insigni cant in participants with gouty nephropathy.

Discussion
This retrospective cross-sectional study outlines the association between the presence and severity of NAFLD and the risk of gouty nephropathy in non-diabetic gouty patients. Interestingly, we detected a signi cant negative association between NAFLD severity and the risk of gouty nephropathy, which is an unexpected nding because the severity of hepatic steatosis in NAFLD is associated with increased CKD risk in previous studies [37]. Moreover, this association was independent of age, sex, smoking, gout duration, and metabolic risk factors (e.g. obesity, hypertension, hyperglycemia, hyperuriacemia and dyslipidemia). We also observed a weakened association of SUA with NAFLD and metabolic risk factors under gouty nephropathy. Therefore, our study suggests a complicated correlation between co-existing NAFLD and the development of renal dysfunction in non-diabetic gouty patients, which might be linked to the surrounding mellitus. Further investigation of NAFLD will provide insights into the pathogenesis of gouty nephropathy.
Up till now, numbers of studies have explored the relationship between NAFLD and CKD risk in different populations. While numbers of previous studies indicated that the presence of NAFLD was a potential contributory risk factor for the development and progression of CKD, recent studies suggested no adverse association between NAFLD hepatic steatosis and renal function both in general populations and diabetic individuals [19][20][21][22]. Some experts argued that NAFLD itself is not an independent risk factor for CKD, while the comorbidities of NAFLD such as obesity, hypertension, hyperglycemia and hyperuricemia are independently associated with renal dysfunction [23]. Some also argue that NAFLD was associated with early stages of CKD, but not the late stages of CKD [22]. However, most of previous studies had a much higher prevalence of diabetes in their study samples [24,25], raising the possibility of selection bias given the strong association between diabetes and CKD. Additionally, although it is reasonable to assume that NAFLD may promote renal damage, it is still uncertain if NAFLD is associated with a speci c type of kidney disease [16]. Anyhow, despite that gouty nephropathy and diabetic kidney disease (DKD) have many risk factors in common, they are actually distinct entities with differing risk pro les. Given that none of the studies have been focused on the speci c relationship between NAFLD and gouty nephropathy, we conducted this retrospective cross-sectional study to explore the association between the presence and severity of NAFLD and the risk of gouty nephropathy in Chinese gouty individuals without diabetes.
Inverse associations of NAFLD severity as well as liver enzymes, with the risk of gouty nephropathy were observed in non-diabetic gouty individuals in this study. Similar to our ndings, latest evidence showed that in CKD population, the prevalence of NAFLD was signi cantly higher in patients with preserved renal function, and they had more severe liver brosis than advanced CKD patients [38]. Similarly, inverse associations of NAFLD as well as liver disease biomarkers with the risk of ischemic stroke have also been reported in a large-scale case-cohort study [39]. Another research addressing the risk of composite cardiovascular endpoints also presented an inverse association of NAFLD in general population with old age, and a positive association in younger individuals [40], suggesting a complex relationship between NAFLD and CVD risk. Additionally, complicated associations between liver dysfunction and the risk of mortality have also been reported, with the observation that low ALT levels indicated a higher all-cause, CVD-related, and cancer-related mortality [41]. Some researchers further argued that ALT levels may exhibit a U-shaped association with cardiovascular and total mortality, and that the low ALT could be considered as a biomarker of an exaggerated hepatic aging process [42,43].
NAFLD and CKD share similar pathological mechanisms, therefore they are speculated to have some links. The aforementioned mechanisms included the role of obesity, insulin resistance, the reninangiotensin system, and dysregulation of glucose metabolism and lipogenesis in the development of both disorders[17, 44,45]. In this study, in line with the concept that NAFLD is the hepatic manifestation of the metabolic syndrome, the presence of NAFLD is associated with obesity and components of metabolic syndrome. However, the association of renal dysfunction and components of metabolic syndrome, such as obesity, insulin resistance and hyperlipidemia, became insigni cant in this nondiabetic gouty population. We concluded several responsible mechanisms regarding the negative association between NAFLD severity and risk of gout nephropathy as follows: (I) Lower degree of oxidative stress in gouty nephropathy. The intensity of antioxidant protection system and OS factors in patients with NAFLD has been shown depending on the form of CKD [46]. While the intensity of OS increases in the form of chronic pyelonephritis and DKD, in the comorbidity of NAFLD with gouty nephropathy and in conditions of asymptomatic hyperuricemia, the degree of OS is signi cantly lower due to the strong antioxidant properties of UA [46]. Notably, UA is considered as a double-edged sword in the context of human health, as it has both anti-and pro-oxidant properties depending on the surrounding environment [47][48][49]. In a physiological milieu, UA serves as an antioxidant, and when homeostasis is perturbed, divergent effects are observed depending on the clinical context [50]. Moreover, the effect of reducing hydroxyl and superoxide radicals is concentration-dependent, such that increasing serum concentrations of UA intensi es scavenging of reactive oxygen species (ROS) [51][52][53]. UA also has the ability to form stable complexes with iron ions, which can dramatically inhibits Fe 3+ -catalyzed ascorbate oxidation and lipid peroxidation [54]. Consistently, inverse relationship between UA and cardiovascular and mortality risk has also been seen in patients with ESRD on hemodialysis [50,[55][56][57], which was considered to be due to antioxidant properties of UA.
(II) Weakened association between UA and metabolic risks under renal insu ciency. An interesting observation was that although the independent association of serum UA with metabolic syndrome was recognized [58], this association no longer existed in participants with advanced CKD after adjustment for other metabolic risk factors and renal eGFR [59]. It was also showed that when there are more late stages of CKD, the relationship of NALFD and CKD attenuates during multivariate analysis [22]. In this study, we further con rmed that in subjects with gouty nephropathy, hyperuricemia is no longer associated with the high prevalence of obesity, insulin resistance, hyperlipidemia or NAFLD severity, compared with those without gouty nephropathy. Consistently, researchers also observed a weak predictor of CVD in people at low risk of atherosclerotic CVD, while UA is recognized as a signi cant independent predictor of CVD among high-risk individuals[60]. This renal function-dependent association might be related to the different properties and effects of UA on metabolic disorders under different physicochemical conditions [50]. Another possible explanation for this phenomenon is that it is the generation of oxygen radicals by xanthine oxidase (XO) in the process of generating UA, rather than hyperuricaemia per se, that causes endothelial dysfunction, oxidative stress and eventually multiple metabolic disorders[61].
(III) Monocyte-associated immunosuppression in ESRD. Despite prolonged and severe hyperuricaemia, gouty arthritis is considered to be rare in patients with ESRD[62, 63]. Further exploration indicated that monocytes from gouty patients with ESRD produced signi cantly lower amounts of proin ammatory cytokines such as IL-1β, IL-6 and TNF-α, in response to stimuli of MSU crystals than did monocytes from healthy subjects, which may be partly account for the infrequent gout episodes in chronic renal failure patients [64]. Similarly, in gouty patients with severe renal dysfunction, the monocyte-associated immunosuppression, with reduced secretion of proin ammatory cytokines in response to MSU crystals, might be partly account for the less severity of NAFLD.
(IV) Frequent use of xanthine oxidase (XO) inhibition provides the potential to attenuate liver damage. Recent evidence indicated that XO inhibition could attenuate diet-induced steatohepatitis in mice[65, 66]. Specially, febuxostat, with its non-purine structure, selectively inhibits both the oxidized and reduced form of XO[67], has been shown to signi cantly decrease hepatic XO activity and UA levels in the NAFLD model mice, accompanied by attenuation of insulin resistance, lipid peroxidation, and classically activated M1-like macrophage accumulation in the liver [65]. Experts further found that in NAFLD patients with hyperuricemia, treatment with febuxostat for 24 weeks signi cantly decreased the serum levels of liver enzymes, alanine aminotransferase and aspartate aminotransferase. In our cross-sectional study, more frequents uses of febuxostat were found in gouty patients with nephropathy compared with those without. Therefore, the more frequent use of febuxostat in gouty patients with nephropathy may partly account for the lower severity of NAFLD and lower levels of liver enzymes.
To the best of our knowledge, the current study is the rst to evaluate the relationships between the presence and severity of NAFLD and risk of gouty nephropathy in hospitalized gouty population without diabetes. Even though previous studies have reported positive associations of NAFLD with CVD and CKD, no study have investigated the separate association with gouty nephropathy. Our present study also has several limitations. Firstly, owing to the cross-sectional study design, we are unable to draw conclusions about the causality of NAFLD and gout nephropathy. Secondly, liver biopsy is the best diagnostic tool for quanti cation of NAFLD, and it is most sensitive and speci c for providing important prognostic information. However, because it is an invasive procedure, it cannot be easily implemented in normal people. Thus, our NAFLD diagnosis was based on ultrasound imaging. It has been shown that only when liver fatty of > 33 % on liver biopsy can radiological imaging optimal for detecting steatosis [68]. Third, eGFR was estimated by age and creatinine regardless of the in uence of some drugs. Some drugs, including trimethoprim and cimetidine, inhibit creatinine secretion, thereby reducing creatinine clearance and elevating the serum creatinine level without affecting the GFR[69]. However, no history of above drug uses was observed in the participants. Finally, patients of different races and ethnics were not included in this study, making it di cult to avoid selection bias. In the future, randomized studies with larger cohorts of patients and longer follow-up and histologically con rmed NAFLD are needed to verify a causal relationship between NAFLD and gouty nephropathy.

Conclusions
In conclusion, in this retrospective cross-sectional study of non-diabetic gouty individuals, NAFLD was independently associated with a lowered risk for gouty nephropathy, and this association was independent of age, sex, gout duration, and metabolic risk factors (obesity, hypertension, hyperglycemia, hyperuriacemia and dyslipidemia). We also observed a weakened association of UA with NAFLD and metabolic risk factors under gouty nephropathy. Our study suggests a complicated correlation between co-existing NAFLD and the development of renal dysfunction in non-diabetic gouty patients, which might be linked to the surrounding mellitus. Although prospective studies are still needed to gure out the causal relationship, our ndings provide novel insights into the potential roles of NAFLD in the pathogenesis of gouty nephropathy.

Declarations
Ethics approval and consent to participate This study was approved by the ethics committee of A liated Hospital of Medical College Qingdao University, and conducted in accordance with good clinical practice. All participants gave their informed consent to participate.

Consent for publication
All participants gave their informed consent to publication.

Availability of data and materials
The datasets generated and/or analyzed during the current study available from the corresponding author on reasonable request.

Figure 2
The box plots of eGFR and Ccr levels strati ed by degree of NAFLD severity, and serum transaminases levels strati ed by eGFR categories. A. eGFR levels strati ed by degree of NAFLD severity; B. Ccr levels strati ed by degree of NAFLD severity; C. ALT levels strati ed by eGFR categories; D. AST levels strati ed by eGFR categories. Data are presented as medians, 25th and 75th percentiles (boxes), and 10th and 90th percentiles (whiskers). Abbreviations: ALT, alanine transaminase; AST, aspartate aminotransferase; Ccr, creatinine clearance; eGFR, estimated glomerular ltration rate; NAFLD, non-alcoholic fatty liver disease.

Figure 3
Multivariate logistic regression analysis of risk factors for gouty nephropathy. Obesity was de ned as BMI > 29.9 kg/m2. Hypertension was de ned by systolic blood pressure ≥ 140mmHg or diastolic blood pressure ≥ 90mmHg, or self-reported previous diagnosis of hypertension by physicians. Hyperglycaemia