Metabolic syndrome (MetS) is characterized by a cluster of high blood pressure, central obesity, high serum triglyceride levels, low serum high-density lipoprotein (HDL) cholesterol levels, and high fasting blood glucose, with insulin resistance as an underlying condition [1
]. People who have MetS are at increased risk of the future development of Type 2 diabetes and cardiovascular diseases [2
In addition to excessive energy intake and a sedentary lifestyle, dietary quality is closely linked to MetS. A prudent/healthy diet pattern, characterized by high intake of vegetables, fruits, whole grains, and legumes, was inversely associated with MetS, while a Western/unhealthy pattern, characterized by a high intake of meat, processed meat, refined grains, and sweets, was positively associated with MetS in meta-analysis of 28 cross-sectional studies [4
]. The dietary-approaches-to-stop-hypertension (DASH) [5
] and Mediterranean diets [6
] were associated with reduced risk of MetS.
Recently, positive associations of low-grade metabolic acidosis, assessed by a high dietary acid load [7
] or high animal-protein intake [10
], with increased risk of Type 2 diabetes, were reported from Europe, Japan, and the United States (U.S.). In addition, low serum bicarbonate levels, a high anion gap, and a high dietary acid load were reported to be associated with insulin resistance [11
]. Dietary acid load was also associated with high blood pressure [13
], obesity and high serum low-density lipoprotein (LDL) cholesterol levels [16
As indices of dietary acid load, potential renal acid load (PRAL) [17
] and net endogenous acid production (NEAP) [18
] have been frequently used in epidemiological studies. Generally, PRAL and NEAP are positively correlated with intake of meat, fish, and eggs, and inversely correlated with an intake of vegetables, fruits, and dairy products [7
]. Therefore, it is thought that PRAL and NEAP are positively correlated with a Western/unhealthy diet pattern, and negatively correlated with a prudent/healthy diet pattern. This suggests that dietary patterns become potential confounders when assessing the relationship between dietary acid load and MetS. Another concern is whether observed associations of dietary quality with MetS are mediated by dietary acid load.
Only a small number of studies have examined the association between dietary acid load and MetS, except for small studies on patients with Type 2 diabetes [19
] or Iranian women [20
]. In these two earlier studies, lifestyle factors, including dietary habits, were not taken into consideration. In the present study, we examined the associations of NEAP with MetS and its components among approximately 28,000 participants in a baseline survey of the Japan Multi-Institutional Collaborative Cohort (J-MICC) study, considering lifestyle factors including nutrient patterns.
shows the characteristics of the study subjects according to NEAP scores. Subjects with higher NEAP scores were more likely to be men, current or past smokers, current drinkers, and less physically active during leisure time. Participants with higher NEAP scores had higher BMI, systolic and diastolic blood pressure, serum triglycerides, and fasting blood glucose, and lower serum HDL cholesterol levels.
also shows the median (with 25% and 75%) intake of nutrients and foods according to NEAP quartiles. NEAP scores were negatively correlated with carbohydrate, soluble and insoluble dietary fiber, n-6 polyunsaturated fatty acids, potassium, calcium, and sodium, and positively correlated with protein, fat, and cholesterol. Spearman’s rank correlation of NEAP scores with Nutrient Pattern 1 (fiber, iron, potassium, and vitamin pattern) and Nutrient Pattern 2 (fat and fat-soluble vitamins pattern) was −0.56 and 0.18, respectively. Regarding food intake, NEAP scores were positively correlated with fish, but negatively correlated with milk, yogurt, leafy green vegetables, green/yellow vegetables, cabbages, oranges, other fruits, and rice.
shows the OR and 95% CI for the associations of NEAP scores with MetS and its components. After adjustment for sex, age, study site, smoking and drinking habits, physical-activity levels during leisure time, total energy intake, and school career (Model 1), higher NEAP scores were associated with significantly higher OR of MetS, obesity, high blood pressure, high serum triglycerides, and high blood glucose (p
for trend < 0.001). After additional adjustment for Nutrient Pattern 1 scores (Model 2), associations were attenuated but remained statistically significant for MetS, obesity, high blood pressure, and high blood glucose (p
for trend < 0.001). The OR of MetS in a subgroup with the highest NEAP scores (Q4), relative to that with the lowest scores (Q1), was 1.25 (95% CI 1.12–1.39). When Nutrient Pattern 2 scores (Model 3) or carbohydrate intake (Model 4) were adjusted instead of Nutrient Pattern 1 scores, results were similar to those in Model 1. In these analyses, Nutrient Pattern 1 and 2 scores were significantly associated with MetS, independent of NEAP scores (Table S2
After stratifying by sex, age, or BMI, significant trends for the association between NEAP scores and MetS were observed for men and women, subjects aged ≥55 and <55 years, and obese subjects (≥25 kg/m2
), with no significant effect modifications (Table 3
). The trend for the association between NEAP scores and MetS did not greatly vary according to study site (Table S3
In the present study, there were significant positive trends between NEAP scores, and MetS and its components except for serum lipids. The associations remained significant after adjustment for carbohydrate intake and for two nutrient-pattern scores significantly associated with MetS in our previous study [25
]. Results were similar after stratifying by sex, age, and BMI.
There have been very few studies on the relationship between dietary acid load and MetS, but Iwase et al. [19
] reported that higher PRAL and NEAP scores were associated with a higher prevalence of MetS in 149 patients with Type 2 diabetes. In a cross-sectional study on 371 Iranian women, NEAP was significantly associated with large waist circumference and high serum triglyceride (TG) levels, but not with the prevalence of MetS (OR = 3.77, 95% CI 0.77–18.4) [20
]. However, in these two studies, the number of subjects was small, and lifestyle factors such as smoking and drinking habits, physical activity, and dietary quality were not considered.
Positive relationships between dietary acid load and the risk of Type 2 diabetes were reported for women enrolled in the E3N-EPIC cohort [7
], male participants in the JPHC study [8
], and male and female health professionals in a meta-analysis of three cohort studies performed in the United States [9
]. On the other hand, in a Swedish cohort study, neither PRAL nor NEAP was associated with the incidence rate of Type 2 diabetes mellitus (DM) in men [26
]. Regarding other components of MetS, a significant association between higher serum anion gap and hypertension was reported in the National Health and Nutrition Examination Survey (NHENES) of the U.S. [27
]. A prospective study also reported a positive trend between NEAP and the risk of hypertension in U.S. women [13
], while a cross-sectional study showed a positive trend between NEAP and hypertension among Japanese workers with BMI < 23 kg/m2
or no shift work [14
]. Recent meta-analysis of one prospective and eight cross-sectional studies showed a significant linear dose–response relationship between PRAL and hypertension [15
]. However, the effect size was rather small, and the risk of hypertension associated with a 20 unit increase in PRAL was 1.03 (95% CI 1.00–1.06). In the National Health and Nutrition Survey of Japan, PRAL and NEAP (assessed on one-day weighted diet record) were positively correlated with BMI and systolic blood pressure, but not with serum HDL cholesterol or glycated hemoglobin concentrations [16
]. In recently published meta-analysis, dietary acid load was positively associated with BMI/central obesity and serum triglycerides [28
]. However, most studies included in this meta-analysis did not adjust for diet patterns, which are important confounders, as discussed later. Taken together, significant positive associations between dietary acid load and Type 2 DM, hypertension, obesity, and serum lipids have been reported by several researchers, but the results were not always consistent or conclusive.
Dietary acid load is positively correlated with a Western/high fat dietary pattern [7
], and inversely correlated with a healthy/prudent dietary pattern [7
]. In our study, NEAP scores were positively correlated with fat and fat-soluble vitamin pattern scores, while they were negatively correlated with fiber, iron, potassium, and vitamin pattern scores, both of which were significantly associated with the prevalence of MetS in our previous study [25
]. Therefore, we adjusted for these nutrient pattern scores as potential confounders. After adjustment, the association between NEAP scores and MetS was attenuated, but still statistically significant. In these analyses, two nutrient patterns were significantly associated with MetS, independent of NEAP. The results suggested that the observed association between NEAP and MetS was not fully explained by healthy or high fat nutrition patterns, and that NEAP and these nutrition patterns reflected different aspects of dietary quality.
Regarding underlying biological mechanisms, chronic low-grate metabolic acidosis is known to result in insulin resistance, an underlying condition of MetS. In the NHENES, serum bicarbonate concentration and serum anion gap were inversely and positively associated with fasting insulin levels and Homeostasis Model Assessment for Insulin Resistance (HOMA-IR), respectively [11
]. The authors speculated that increased cortisol excretion mediated the relationship between metabolic acidosis and insulin resistance. In a cross-sectional study in a Japanese population, PRAL and NEAP were also positively associated with HOMA-IR [12
]. Acidosis induces the transcription and expression of Transforming Growth Factor (TGF)-β, which inhibits insulin secretion and reduces the affinity of insulin to its receptor [29
]. Acidosis also increases the urinary excretion of calcium and magnesium [30
] that play important roles in insulin action [31
]. These two minerals, together with potassium, reduce blood pressure by counterbalancing the effects of sodium [33
]. Dietary calcium increases intracellular calcium concentration and reduces vascular resistance.
The strengths of the present study are the large number of study subjects, and adjustment for potential confounding factors, including nutrient patterns and school career. On the other hand, this study had several limitations. First, because of the cross-sectional study design, the temporal relationship between dietary acid load and MetS onset was not clear. Second, only NEAP was used as an index of dietary acid load. We could not calculate PRAL scores due to a lack of data on magnesium and phosphorus intake. However, in general, NEAP and PRAL scores have high positive correlation in Japanese populations (r ≥ 0.95) [8
]. Third, because data on abdominal circumference were not available for every individual, BMI was used instead. When the BMI cut-off point of 25 kg/m2
was used, the sensitivity and specificity levels of our diagnosis of high waist circumference were 83% and 82% for men (no. = 633), and 45% and 98% for women (no. = 623), respectively, in the subsample of the J-MICC study in the Tokushima Prefecture [25
]. Therefore, misclassification of abdominal obesity may have occurred, especially in women. However, misclassification may have been nondifferential with regard to NEAP scores, with the effect tending towards null results. Fourth, even though several potential confounders were adjusted, there may have been unknown confounding factors, as with all observational studies. Finally, since almost all the study participants were Japanese, the obtained results may not be directly applicable to other ethnic groups.