Hyperuricemia is caused by the excessive secretion of uric acid in the liver, and/or by the decrease of uric acid excretion in the kidney or intestine, resulting in accumulation of uric acid in the body(Reference Merriman1). Approximately a quarter of patients with hyperuricemia will develop gout(Reference Campion, Glynn and DeLabry2), which is a common rheumatic disease characterised by the deposition of sodium simplex crystals around the joints(Reference Dalbeth, Merriman and Stamp3), affecting patient’s activities and daily life. Besides, an increasing body of evidence have demonstrated that hyperuricemia is a risk factor associated with chronic diseases, such as hypertension, diabetes, metabolic syndrome, CHD and chronic kidney disease(Reference Borghi, Agabiti-Rosei and Johnson4,Reference Ponticelli, Podestà and Moroni5) . Thus, it is urgent and necessary to identify the protective factors in the prevention of hyperuricemia. Plenty of factors have been reported to be associated with the initiation and progression of hyperuricemia, such as gene, gender, age, lifestyle, dietary habits, drug treatment and economic development. Regarding the changeable factors, dietary factors play a pivot role in the prevention of hyperuricemia. Several diets are associated with elevated levels of blood uric acid and might trigger hyperuricemia, such as excessive intakes of meat, meat broth, seafood, beer and sweets, while moderate intakes of vegetable and fruit, dietary fibre, and vitamin C have been reported to be negatively associated with hyperuricemia risk(Reference Choi and Curhan6–Reference Sun, Sun and Wang9).
Nuts are rich in essential fatty acids, high-quality protein, various vitamins and trace elements. Different types of nuts have different nutrient contents, which might provide differential effects in the prevention of hyperuricemia. So what is the effect of nut intake on serum uric acid? A randomised controlled trial (RCT) found that supplemental Pakistani almond and American almond had a reduced effect on serum uric acid levels in patients with CHD(Reference Jamshed, Gilani and Sultan10). Conversely, another randomised controlled trial demonstrated that supplemental walnut and salt-free cashew nut had no significant effect on serum uric acid concentration and other biochemical parameters in patients with metabolic syndrome(Reference Mukuddem-Petersen, Stonehouse Oosthuizen and Jerling11). To date, no study has assessed the relationship between different types of nut intake and hyperuricemia risk. Therefore, by using the data from the Health Check Program of Undergraduates (HCPU), the present study comprehensively investigated the relationship between different nut intake and hyperuricemia risk with a cross-sectional study.
Methods
Study population and design
We conducted a cross-sectional study of undergraduates who participated in the HCPU at Qingdao University from August 2018 to October 2019 to evaluate whether dietary intake, lifestyle and other factors are associated with the risk factors of chronic disease. A total of 16 211 undergraduates who took part in health check programme were recruited. Undergraduates who voluntarily completed the semi-quantitative FFQ were included in the cross-sectional study. Participants who did not complete the FFQ, with a history of malignancy, liver, CVD and metabolic diseases were excluded. Finally, 14 056 participants were included in the statistical analysis. This study was approved by the Ethics Committee of the Affiliated Hospital of Qingdao University (Qingdao University, China). All undergraduate participants signed written informed consent forms. This study was conducted in accordance with the criteria set out in the Declaration of Helsinki.
Dietary nut intake
The dietary intakes of different nuts were assessed by a semi-quantitative FFQ designed by trained dietitians from the Department of Nutrition, the Affiliated Hospital of Qingdao University. The type, frequency (daily, weekly and monthly) and dose (servings) of nut consumed by participants in the previous months were investigated. We have pre-defined a small handful without shells as one serving, namely 25 g. The dietary intakes of nuts were transformed into g/d. The semi-quantitative FFQ contained the following nut: peanut, melon seed, pine nut, pistachio, cashew, pecan, walnut, almond and daily nut. Among them, daily nut is a packaged food with mixed nuts. The mean intake of total nuts of the subjects is 35·63 ± 60·79 g/d, composed of peanut, melon seed, pine nut, pistachio, cashew, pecan, walnut, almond and daily nut.
Physical examination
Anthropometric information, including systolic blood pressure and diastolic blood pressure (mmHg), height (m) and weight (kg), was measured by well-trained nurses in accordance with standard procedures. BMI (kg/m2) was calculated by dividing weight in kilograms by height in square metres.
Biochemical measurement
Participants were required to fasting for at least 10 h, and their venous blood was collected in a vacuum sampling vessel for laboratory examination. The serum was obtained by centrifugation (4000 rpm for 10 min at 4°C) for biochemical analysis. The outcome of interest was uric acid, which was determined by enzyme colorimetry and uricase-peroxidase method (Accute TBA-40FR autoanalyser, Toshiba). Male participants with serum uric acid ≥ 416·0 μmol/l (7·0 mg/dl) and female participants with serum uric acid ≥ 357·0 μmol/l (6·0 mg/dl) were defined as hyperuricemia(Reference Yu, Zheng and Zhang12). In addition, automatic biochemical analyser was adopted to measure blood glucose and lipid profiles.
Statistical analysis
Shapiro–Wilk test was conducted to check whether the distribution of continuous variables was normal. If the variable was a normal distribution, the data were expressed as the mean and standard deviation. For skewed variables, the data were presented as median (quartile range). The continuous variables were compared by using t test, and the categorical variables were analysed with chi-square test (χ 2).
Dietary exposure to different types of nuts (peanut, melon seed, pine nut, pistachio, cashew, pecan, walnut, almond, daily nut and total nut) was classified by quartiles (quartile 1: < 25th percentile, quartile 2: ≥ 25th–50th percentile, quartile 3: ≥ 50th–75th percentile and quartile 4: ≥ 75th percentile), and each quartile intake was compared with the lowest category. Logistic regression model was implemented to calculate OR and 95 % CI, by using the lowest category as reference. Multivariable-adjusted models were adopted to assess the association between individual nut intake and hyperuricemia, the model 1 was adjusted for age, gender, BMI, blood pressure, serum concentrations of glucose, total cholesterol, TAG and HDL-cholesterol, model 2, as model 1 added physical activity, dietary intakes of total energy, total carbohydrate, total protein and total fat, dietary intakes of candies, meat (pork, beef and mutton), soup (poultry and cattle broth, fish and shrimp soup) and seafood. Besides, the individual nut items were included in the multivariable-adjusted model and adjusted for each other. The trend test was performed by assigning a median value to each category and modelling the variable as a continuous variable(Reference Larsson, Virtamo and Wolk13). Subgroup analyses stratified by gender and BMI were conducted to assess potential interaction with this variable. Statistical analyses were conducted with STATA 15.0 (Stata Corp.), and two-tailed P-value < 0·05 was considered as statistically significant.
Results
Baseline characteristics of the participants are shown in Table 1. A total of 16 211 undergraduates at Qingdao University participated in the HCPU. Among them, 14 056 participants (6862 males and 7194 females) aged 15–25 years completed the semi-quantitative FFQ and were included for data analysis. Among them, 5767 participants were identified as hyperuricemia (41 %). The prevalence of male participants (53·4 %) was significantly higher than that of female participants (29·3 %) (P < 0·001). The BMI and fasting blood glucose levels of hyperuricemia participants were significantly higher than those of normal subjects. Participants with hyperuricemia had higher levels of TAG and total cholesterol and lower levels of HDL-cholesterol than normal participants (P < 0·001). There was no significant difference in age between the participants (P < 0·001).
Table 1. The baseline characteristics of the participants
P for difference between genders was conducted by chi-square test, and the other parameters were analysed with t test.
In crude logistic regression model, compared with the lowest quartile, the highest quartile intakes of peanut (OR = 1·18; 95 % CI (1·07, 1·29); P for trend < 0·001), melon seed (OR = 1·17; 95 % CI (1·07, 1·28); P for trend < 0·001), cashew (OR = 1·11; 95 % CI (1·01, 1·22); P for trend = 0·001) and total nut (OR = 1·10; 95 % CI (1·00, 1·21); P for trend = 0·011) were positively associated with the hyperuricemia risk (see online Supplemental Figs. 1–3 and 8). Non-significant association was found for other nuts (Figs. 1–2 and Supplemental Figs. 4–7) in the crude model. The positive association between hyperuricemia risk and peanut, melon seed, cashew, and total nut were disappeared in model 1, adjusting for age, gender, BMI, blood pressure, serum concentrations of glucose, total cholesterol, TAG, HDL-cholesterol and individual nut intake, but walnut (OR = 0·77; 95 % CI (0·57, 1·03); P for trend = 0·039) intake was marginally negatively associated with the hyperuricemia risk (Fig. 2). After further adjustment for dietary information (intakes of energy, carbohydrate, protein, fat, candies, meat, soup and seafood) and physical activity in model 2, the highest quartile of pine nut (OR = 0·71; 95 % CI (0·51, 0·98); P for trend = 0·061) intake, compared with the lowest quartile, was inversely associated with the risk of hyperuricemia (Fig. 1). Walnut (OR = 0·78; 95 % CI (0·58, 1·05); P for trend = 0·047) intake remained the marginally negatively association with hyperuricemia risk in model 2 (Fig. 2).
Fig. 1. Association of pine nut intake with hyperuricemia risk. Number of participants in each quartile: Q1, 4401; Q2, 4827; Q3, 1437; Q4, 3391
Fig. 2. Association of walnut intake with hyperuricemia risk. Number of participants in each quartile: Q1, 3573; Q2, 4926; Q3, 1631; Q4, 3926
The associations between hyperuricemia and dietary individual and total nut in stratified analyses by gender are shown in Supplemental Table 1. Only pine nut intake (OR = 0·63; 95 % CI (0·39, 1·03); P for trend = 0·087) showed a marginally inversely association with hyperuricemia risk in model 2 stratified by female. No significant associations were found between hyperuricemia risk and other dietary nut intake in crude and adjusted model stratified by gender.
The associations between hyperuricemia and dietary individual and total nut in stratified analyses by BMI are shown in Supplemental Table 2. For participants with BMI ≤ 24, the highest quartile of peanut (OR = 1·23; 95 % CI (1·11, 1·37); P for trend = 0·001), melon seed (OR = 1·21; 95 % CI (1·09, 1·34); P for trend < 0·001), pistachio (OR = 1·13; 95 % CI (1·02, 1·26); P for trend = 0·004), cashew (OR = 1·18; 95 % CI (1·06, 1·32); P for trend = 0·001), almond (OR = 1·12; 95 % CI (1·01, 1·24); P for trend = 0·029), daily nut (OR = 1·16; 95 % CI (1·04, 1·29); P for trend = 0·010) and total nut (OR = 1·17; 95 % CI (1·05, 1·31); P for trend = 0·012) intake, compared with the lowest quartile, were positively associated with hyperuricemia risk in crude model. After further adjustment in model 1 and model 2, the positive correlations between hyperuricemia risk and peanut, melon seed, pistachio, cashew, almond, daily nut and total nut were disappeared. In model 2, dietary intakes of pine nut (OR = 0·71; 95 % CI (0·49, 1·03); P for trend = 0·081) and walnut (OR = 0·73; 95 % CI (0·52, 1·03); P for trend = 0·047) were marginally inversely associated with the hyperuricemia risk. No significant associations were found between hyperuricemia risk and dietary individual and total nut intake in crude and adjusted model for participants with BMI ≤ 24.
Discussion
To the best of our knowledge, this study was the first to comprehensively investigate the relationship between individual nut intake and hyperuricemia risk with a cross-sectional study. The results showed that dietary higher intakes of pine nut and walnut were negatively associated with hyperuricemia risk.
The prevalence of hyperuricemia varies greatly in different geographic regions, which is associated with local eating habits. Previously, a national cross-sectional survey of 13 provinces in Chinese, including rural and urban areas, demonstrated that the prevalence of hyperuricemia was 8·4 % from 2009 to 2010(Reference Liu, Zhang and Wang14), while another study conducted in a rural area of Henan Province showed that the prevalence of hyperuricemia in adults was 12·6 % from 2015 to 2017(Reference Dong, Zhang and Wang15). In the present study, the prevalence of hyperuricemia was as high as 41 %, which was significantly higher than previous studies. It could be explained by the following reasons. First, Qingdao is a coastal city, rich in all kinds of seafood. Second, Qingdao’s specialty, Tsingtao Beer is well known, and both foods are rich in purines. Therefore, diet composition is an important variable factor contributing to hyperuricemia. Different kinds of nuts contain differential fatty acids, vitamins and trace elements. Therefore, it may have differential effects in preventing hyperuricemia.
In the present study, higher intakes of walnut and pine nut were negatively associated with hyperuricemia risk. Also both walnut and pine nut are nutritious foods, but each has its own unique ingredients. Generally speaking, these nuts are rich in MUFA and PUFA, protein, dietary fibre, vitamins, minerals, polyphenols and antioxidants(Reference Tas and Gokmen16,Reference Jahanban-Esfahlan, Ostadrahimi and Tabibiazar17) . The specific nutrient composition and content are shown in Table 2. It is generally accepted that nuts are high energy density food, because their fat content is as high as 42–76 g per 100 g. However, diet rich in nuts does not appear to be obesity but was associated with reduced risk of obesity(Reference Rajaram and Sabaté18,Reference O’Neil, Fulgoni and Nicklas19) , which might be due to the low content of SFA and high content of MUFA and PUFA in nuts(Reference Souza, Gomes and Naves20), as shown in Table 2, walnut and pine nut contain more PUFA than other nuts. In the renal tubules of the human kidney, the urate in the crude urine is mainly reabsorbed into the blood through the pathway mediated by urate transporter 1 (URAT1). It has been reported that unsaturated fatty acids, especially long-chain unsaturated fatty acids, can inhibit the uric acid reabsorption role of URAT1(Reference Saito, Toyoda and Takada21). Additionally, several studies have found that minerals in nut, such as Mg, are negatively correlated with hyperuricemia, and its mechanism might be related to inflammation(Reference Zhang and Qiu22). Mg deficiency was related to acute inflammatory response mediated by Ca, N-methyl-D-aspartate, IL-6, and TNF-α, and that C-reactive protein as biomarkers of inflammation(Reference Nielsen23). Hyperuricemia was positively correlated with C-reactive protein, while dietary Mg intake was negatively correlated with C-reactive protein content. Therefore, we inferred that Mg intake was negatively correlated with hyperuricemia, and circulating uric acid played an important role in inflammation(Reference Dibaba, Xun and He24–Reference King, Mainous and Geesey26). In addition, walnut can inhibit the activity of xanthine oxidase and oxidative stress, which is rich in the polyphenols and antioxidants, also reduce the production of uric acid and accelerate the excretion of uric acid(Reference Jahanban-Esfahlan, Ostadrahimi and Tabibiazar17,Reference Nijveldt, van Nood and van Hoorn27) . Besides, dietary fibre in nut might also contribute to prevent hyperuricemia(Reference Sun, Sun and Zhang8) and the reasons have been summarised as follows: (1) the viscosity and water-holding capacity of dietary fibre are consistent with the absorption of purine or adenine by digestive system(Reference Koguchi, Nakajima and Wada28,Reference Koguchi, Koguchi and Nakajima29) and (2) dietary fibre could promote intestinal emptying and have the potential role of binding intestinal uric acid excretion(Reference Zhang and Qiu22). From what has been mentioned, the synergistic effect provided by nuts could confer a lower risk of hyperuricemia. In addition, walnut and pine nut also contain bioactive components such as tocols, phystrols and sphingolipids. These components might influence the production of uric acid by different pathways, but there are few studies on the relationship between bioactive components in nuts and hyperuricemia(Reference Alasalvar and Bolling30). Besides, elevated serum uric acid is positive correlation with obesity(Reference Poletto, Harima and Ferreira31). With the increase in BMI, the level of uric acid would be increased(Reference Li, Zhang and Zeng32). Stratified analysis showed that BMI was a factor influencing the relationships between nuts and hyperuricemia risk. Therefore, it is necessary to control the intakes of nuts to avoid excessive energy intake, and the recommended nut intake is less than 25 g/d(Reference Stuetz, Schlörmann and Glei33). Although the intake of pine nut and walnut is negatively correlated with hyperuricemia risk, due to the high oil content of nuts, people with hyperuricemia should correspondingly reduce the intake of fat from other foods to reduce complications.
Table 2. Average nutrient composition of nut (per 100 g)
LA, linoleic acid; ALA, α-linolenic acid.
Data for raw nuts, except where specified.
Source: US Department of Agriculture Nutrient Data Base at: https://fdc.nal.usda.gov/fdc-app.html#/?query=nut (accessed November 2020).
The present study has several advantages worth mentioning. As far as we know, this study was the first to systematically evaluate the relationship between different nut intake and hyperuricemia risk, which showed that special nuts are beneficial for the prevention of hyperuricemia. The semi-quantitative FFQ was designed by professional dietitians in Affiliated Hospital of Qingdao University, and it has been used for more than three decades and proved to be reliable. Simultaneously, the present study had some shortcomings that needed to be addressed. First, the participants were undergraduates about 18 years of age, and the results were not representative of the general Chinese population. Second, due to the cross-sectional design of this study, it cannot reveal the causal and temporal relationship between nut intake and hyperuricemia risk. Therefore, further prospective cohort studies or intervention trials are warranted to explore these causal relationship. Third, since the nuts consumption by volunteers was not weight each time, semi-quantitative FFQ was not accurate in estimating nuts intake. In addition, misclassification was unavoidable when using FFQ to assess dietary information. Fourth, there was no design about liquor and alcoholic beverages intake in the questionnaire. Thus, alcohol intake could not be adjusted in the multivariate adjustment model, and that might influence the final results.
Conclusion
The present study indicated that the relationships between different intakes of nuts and hyperuricemia risk are different. Increasing intakes of pine nut and walnut were negatively associated with the hyperuricemia. Further prospective studies and intervention studies are warranted to confirm the findings of the present study.
Acknowledgements
Acknowledgements: The authors acknowledge the participants and staff members who helped with this study. Financial support: This work was supported by National Natural Science Foundation of China (NSFC: 81773433 and 82073538); by the Key scientific Research Projects in Shandong Province China (2017YYSP007); and by the 2018 Chinese Nutrition Society (CNS) Nutrition Research Foundation-DSM Research Fund (CNS-DSM2018A30). Conflict of interest: There are no conflicts of interest. Authorship: C.W. was responsible for the investigation, formal analysis and roles/writing-original draft. X.F.G. was responsible for the conceptualisation, formal analysis and writing-review and editing. T.Y. was responsible for the investigation. T.Z. was responsible for the investigation. D.L. was responsible for the conceptualisation, writing-review and editing. Ethics of human subject participation: This study was conducted according to the guidelines laid down in the Declaration of Helsinki, and all procedures involving research study participants were approved by the Ethics Committee of the Affiliated Hospital of Qingdao University. Written informed consent was obtained from all subjects.
Supplementary material
For supplementary material accompanying this, paper visit https://doi.org/10.1017/S1368980021002998
