2025 Volume 32 Issue 3 Pages 356-366
Aim: We aimed to elucidate the effect of a healthy diet containing adequate amounts of protein and vegetables on metabolic indices.
Methods: In this randomized crossover study, twenty-two healthy Japanese participants ingested two different test meals: fish diet (F) or fish diet with adequate vegetable content (FV). Each 5-day diet load test was separated by a washout period of at least seven days. Metabolic indices were measured in fasting blood and 24-h urine samples.
Results: The delta (Δ) plasma glucose and Δserum low-density lipoprotein (LDL) cholesterol concentrations were significantly larger in the participants in group FV than in group F (p=0.042, p=0.013, respectively). The urinary pH in participants in group F on day 6 was significantly lower than on day 1 (p=0.008), and the Δurinary pH and Δnet gastrointestinal absorption of alkali of participants in group FV tended to be smaller than in group F (p=0.070, p=0.075, respectively).
Conclusions: This study showed that a healthy diet containing adequate protein and vegetables reduced the dietary acid load and improved plasma glucose and serum LDL concentrations in healthy Japanese participants.
The trial registration number: UMIN000050761
High dietary acid load induces mild metabolic acidosis1, 2). The potential renal acid load (PRAL), a valid indicator of dietary acid load, could be calculated from the dietary intake of five nutrients (protein, phosphorus, potassium, calcium, and magnesium)3-5). It has been reported that high dietary acid load is associated with several cardiometabolic risk factors, such as high adiposity measures6), high low-density lipoprotein (LDL) cholesterol levels7), high blood pressure or hypertension6-9), type 2 diabetes10), and hyperuricemia11). In addition, previous studies suggest that a high dietary acid load is associated with an increased risk of total mortality and mortality from cardiovascular disease (CVD)12) and that both high acidity and high alkalinity of the diet are associated with an increased risk of total mortality and CVD-related mortality13).
A Western-type diet, characterized by a high intake of animal products and a low intake of fruits and vegetables, has become more prevalent globally. Although a previous study suggests that higher intake of vegetables is associated with a lower risk of CVD mortality14), the 2019 National Health and Nutrition Survey, Japan (NHNSJ) revealed that the mean daily intake of vegetables for Japanese adults did not meet the target value of 350 g recommended in the national plan for health promotion named “The second term of National Health Promotion Movement in the twenty first century (Health Japan 21 (the second term))”15, 16). On the other hand, the dietary protein intake for Japanese adult males and females attains the Recommended Dietary Allowances (RDAs) of protein in the Dietary Reference Intakes (DRIs) for Japanese, 2020 15, 17). Additionally, with the recent popularity of carbohydrate-restricted diets, excessive protein intake is a concern. Thus, to reduce the dietary acid load of the modern Japanese diet, increased vegetable (cation-rich food) intake and adequate protein (anion-rich food) intake should be considered. Although a previous study reported that urinary pH for alkaline diet was significantly higher than those for acidic diet18), few reports have documented the outcomes of dietary intervention tests under strictly controlled diet conditions. Thus, we conducted the rigorously controlled dietary intervention test to determine the effect of lower dietary acid load on metabolic indices.
This study aimed to elucidate the effect of a healthy diet containing adequate proteins and vegetables on metabolic indices.
Twenty-two healthy Japanese volunteers were recruited for this study. The inclusion criteria were age 20−59 y, and those who have the capacity to consent, voluntarily participate with sufficient understanding. The participants confirmed they were not on medical treatment or nutritional supplementation. Other exclusion criteria were allergies to the content of the test meals, heavy alcohol consumption, being currently pregnant or breastfeeding, and smoking. This study was conducted after obtaining written informed consent from all participants. It was approved by the ethics committee of the University of Shizuoka (Approval no. 3-58, approved 5 January 2021) and retrospectively registered with the University Hospital Medical Information Network (Registration no. UMIN000050761, retrospectively registered 5 April 2023). The study was performed in accordance with the Helsinki Declaration.
Study ProtocolWe used a randomized crossover study design that was stratified by sex and conducted from January 2021 to July 2021 at the University of Shizuoka, Shizuoka, Japan. As far as we checked, this study was the first human trial to elucidate the effect of a healthy diet containing adequate amounts of protein and vegetables in a controlled feeding setting. Therefore, no formal power calculation could be undertaken. We calculated the sample size on the serum LDL cholesterol concentration within a group using 0.80 power and a two sided α of 0.05 based on a previous study19) (calculated using G*Power 3.1). As a result, a minimal sample size of 10 participants was required. Taking into account a possible drop out rate of up to 20% and a sex difference, the number of participants was increased to 12 per sex. However, the previous study19) differs from this study in the type and duration of the intervention and should be treated with caution, as it is not appropriate as a formal power calculation.
Each test was conducted over six consecutive days and separated by a washout period of at least seven days. The flow diagram of participants through each test period are shown in Supplemental Fig.1. All participants were asked to avoid heavy exercise and any intake of alcohol and purine-rich foods (＞200 mg/100 g) for three days before each test period. All the participants were instructed to eat and drink the same prescribed meal until 20:00 on the night before day 1 of the test period. The prescribed meal comprised 658 kcal of energy, 15.4 g of protein, 10 g of fat, and 127 g of carbohydrate for males and 583 kcal of energy, 14.1 g of protein, 10 g of fat, and 110 g of carbohydrate for females. Participants were prohibited from eating or drinking anything other than water after ingesting the prescribed meal and instructed to drink 500 mL of water until 24:00 before day 1 of the test period. Fasting blood samples were collected at 7:30 on days 1 and 6. Participants ingested test meals at 8:00 (breakfast), 12:00 (lunch), and 18:00 (dinner) during the test period (day 1 through day 5). During the test period, they were instructed to drink 200 mL of water after waking up and 1300 mL of water from 8:00 to 24:00. During 24-h urine collection, urine just before breakfast (at 8:00) on day 1 was discarded and collected after breakfast on day 1 until just before breakfast (at 8:00) on day 2, which was denoted as day 1. Similarly, 24-h urine collection was performed on days 5 to 6, which was denoted as day 6. Anthropometric measurements were performed on days 1 and 6. Participants responded to a food intake frequency survey using the brief-type self-administered diet history questionnaire (BDHQ) before the intervention to investigate their habitual nutrient intake.
Flow diagram of participants through each test period
The two different test meals used were fish diet (F) and fish diet with adequate vegetable content (FV). Each test meal was designed to be within the energy ratio of protein:fat:carbohydrate (13–20:20–30:50–65) based on the DRIs for Japanese17) (Table 1, Supplemental Table 1). The main dish was fish, and each test meal was the same except for the amounts of vegetables (F; 95 g, FV; 380 g). The quantities of vegetables included in the meal provided in group FV met the recommendation of at least 350 g as recommended by Health Japan 21 (the second term)16). The participants were instructed to ingest each test meal within 30 minutes. The Japan Food Research Laboratories Foundation (Tokyo, Japan) analyzed the test meal components.
| A. Male | |||||||
| Group | Energy kcal/day | Protein g/day | Fat g/day | Carbohydrate g/day | Dietary fiber g/day | NaCl g/day |
PRAL mEq/day |
| F | 2344 | 76.0 | 74.4 | 354.3 | 22.0 | 5.1 | 19.7 |
| %Energy | 13.0 | 28.6 | 58.4 | ||||
| FV | 2496 | 83.0 | 81.8 | 372.9 | 29.7 | 6.8 | 11.1 |
| %Energy | 13.3 | 29.5 | 57.2 | ||||
| B. Female | |||||||
| Group | Energy kcal/day | Protein g/day | Fat g/day | Carbohydrate g/day | Dietary fiber g/day | NaCl g/day |
PRAL mEq/day |
| F | 2034 | 70.9 | 63.4 | 305.3 | 20.1 | 5.1 | 18.5 |
| %Energy | 13.9 | 28.1 | 58.0 | ||||
| FV | 2154 | 76.7 | 68.6 | 321.9 | 27.8 | 6.3 | 9.3 |
| %Energy | 14.2 | 28.7 | 57.1 | ||||
F: fish diet, FV: fish diet with adequate vegetable content, NaCl: sodium chloride, PRAL: potential renal acid load
| A. Male | ||||
| Group |
K mg/day |
Ca mg/day |
Mg mg/day |
P mg/day |
| F | 2060 | 707 | 290 | 1126 |
| FV | 2771 | 895 | 354 | 1307 |
| B. Female | ||||
| Group |
K mg/day |
Ca mg/day |
Mg mg/day |
P mg/day |
| F | 1871 | 637 | 240 | 996 |
| FV | 2550 | 782 | 304 | 1142 |
F: fish diet, FV: fish diet with adequate vegetable content, K: potassium, Ca: calcium, Mg: magnesium, P: phosphorus
PRAL was calculated according to the following formula3-5):
PRAL (mEq/day)=0.4888×protein (g/day)＋0.0366×phosphorus (P) (mg/day)−0.0205×potassium (K) (mg/day)−0.0125×calcium (Ca) (mg/day)−0.0263×magnesium (Mg) (mg/day)
Blood and Urine Analysis Methods and Anthropometric MeasurementsBlood samples were centrifuged at 940 g for 10 min at 4℃, separated into plasma or serum, and stored at −80℃ until the analysis. Clinical laboratory measures included the concentrations of plasma glucose, hemoglobin A1c (HbA1c), serum blood urea nitrogen, creatinine, aspartate aminotransferase, alanine aminotransferase, γ-glutamyl transpeptidase, uric acid (UA), insulin (IRI), triglyceride (TG), LDL cholesterol, and high-density lipoprotein (HDL) cholesterol. Urinary measurements included urinary pH and the concentrations of urinary UA, sodium (Na), chloride (Cl), K, Ca, Mg, and P. A clinical testing company (SRL Inc., Tokyo, Japan) analyzed blood and urine samples, except for the analysis of urinary pH. The pH was measured using a portable pH meter (LAQUA act, D-71; Horiba Scientific, Kyoto, Japan). The net gastrointestinal absorption of alkali (GIAA) was calculated using the following formula20):
GIAA (mEq/day)=[Urinary Na (mEq/day)＋Urinary K (mEq/day)＋Urinary Ca (mEq/day)＋Urinary Mg (mEq/day)]−[Urinary Cl (mEq/day) +1.8×Urinary P (mmol/day)]
We calculated the urinary UA excretion using the following formula:
Urinary UA excretion (mg/day)=Urinary UA (mg/dL)×Urinary volume (dL/day)
Anthropometric measurements were performed using bioelectrical impedance analysis (InBody 770; InBody Japan, Tokyo, Japan). Height was measured using a stadiometer (YL-65S; Yagami, Aichi, Japan).
Statistical AnalysisAll data are presented as mean ±SD or medians (25th to 75th interquartile range). One male participant’s urinary Cl concentration on day 6 in group F was below the detection limit and was excluded from the analysis. We employed all other data for the analysis and performed the allocation based on sex. We assessed the order effects on delta (Δ) serum LDL concentrations and confirmed the validity of the crossover design. The Shapiro−Wilk statistic was used for data normality testing. Parametric analysis was used for normally distributed data, and non-parametric analysis was used for data exhibiting a non-normal distribution. Differences between sexes were identified using an independent t-test or the Mann−Whitney U test. Differences in serum and urinary metabolic indices between days 1 and 6 were identified using a paired t-test or the Wilcoxon signed-rank test. Differences in serum and urinary metabolic indices between groups F and FV were identified using a paired t-test or the Wilcoxon signed-rank test. P-values ＜0.05 were considered statistically in all analyses. We performed the statistical analysis using the SPSS software for Windows, release 26 (IBM Corp., Armonk, NY).
Of 24 participants obtained written informed consent, 23 participants were enrolled the study. Twenty two participants completed the study and were included in the analysis (Supplemental Fig.1). The clinical and biological characteristics of the participants included in the analysis are shown in Table 2. According to the data calculated from the BDHQ, habitual intake of energy, the energy ratio of protein, and minerals (e.g., K, Ca, and Mg) before the intervention did not reach the RDAs or tentative dietary goal for preventing life-style related diseases (DG) (Supplemental Table 2).
| Total participants | Male | Female | p-value | |
|---|---|---|---|---|
| n | 22 | 11 | 11 | |
| Age (year) | 23.5 (22.0, 34.3) | 23.0 (21.0, 26.0) | 24.0 (23.0, 35.0) | 0.274 |
| BMI (kg/m2) | 20.8 (19.2, 23.7) | 22.0 (17.5, 25.1) | 20.6 (19.5, 23.1) | 0.870 |
| BUN (mg/dL) | 10.9±2.2 | 12.2±1.7 | 9.5±1.9 | 0.003 |
| Cre (mg/dL) | 0.8±0.2 | 1.0 (0.9, 1.1) | 0.7 (0.6, 0.7) | ＜0.001 |
| UA (mg/dL) | 5.6±1.3 | 6.4 (5.7, 7.3) | 4.7 (4.0, 5.0) | 0.001 |
| HbA1c (%) | 5.2 (5.0, 5.3) | 5.2 (5.0, 5.3) | 5.2 (5.0, 5.4) | 0.549 |
| AST (U/L) | 18.0 (14.8, 21.5) | 18.0 (15.0, 23.0) | 17.0 (14.0, 20.0) | 0.598 |
| ALT (U/L) | 14.0 (11.0, 24.3) | 16.0 (14.0, 25.0) | 11.0 (8.0, 17.0) | 0.032 |
| γ-GTP (U/L) | 18.0 (15.0, 24.3) | 20.0 (18.0, 24.0) | 15.0 (12.0, 27.0) | 0.121 |
Values are presented as mean±standard deviation or medians (25th to 75th interquartile range). Differences between sexes were identified using an independent t-test or the Mann–Whitney U test. BMI: body mass index, BUN: blood urea nitrogen, Cre: creatinine, UA: uric acid, HbA1c: hemoglobin A1c, AST: aspartate aminotransferase, ALT: alanine aminotransferase, γ-GTP: γ-glutamyl transpeptidase
| Total participants | Male | Female | |
|---|---|---|---|
| Energy (kcal) | 1599.8±552.3 | 1687.4±415.9 | 1512.3±671.4 |
| Protein %Energy (%) | 9.3 (7.8, 10.3) | 9.5±4.0 | 9.2±2.6 |
| Fat %Energy (%) | 28.6±7.4 | 29.4±8.4 | 27.8±6.5 |
| Carbohydrate %Energy (%) | 62.1±10.0 | 61.1±11.5 | 63.0±8.6 |
| K (mg) | 2015.1±809.2 | 2064.3±1039.3 | 1966.0±538.1 |
| Ca (mg) | 380.2±167.4 | 422.4±208.2 | 337.9±107.7 |
| Mg (mg) | 200.7±76.6 | 212.8±92.0 | 188.5±59.5 |
| P (mg) | 875.8±337.7 | 956.7±390.7 | 794.9±269.0 |
Values are presented as mean±standard deviation or medians (25th to 75th interquartile range).
BDHQ: brief-type self-administered diet history questionnaire, K: potassium, Ca: calcium, Mg: magnesium, P: phosphorus
Table 3 shows the values of the evaluated metabolic indices in group F. These results by sex are shown in Supplemental Table 3. Plasma glucose, serum IRI, TG, LDL cholesterol, HDL cholesterol concentrations, and urinary pH for group F on day 6 were significantly lower than on day 1. Serum UA concentrations for group F on day 6 were significantly higher than on day 1. Urinary GIAA on day 6 tended to be lower than on day 1 in group F (p=0.086). Urinary UA excretion on day 6 tended to be higher than on day 1 in group F (p=0.077).
| Group F | Group FV | p-value Change (Δ) | |||
|---|---|---|---|---|---|
| Change (Δ) | Change (Δ) | ||||
| Plasma glucose (mg/dL) | |||||
| Day 1 | 98.4±6.5 | 97.5 (93.0, 104.3) | |||
| Day 6 | 90.8±4.6＊＊ | −7.0 (−10.3, −3.0) | 88.0 (85.0, 92.5)＊＊ | −8.5 (−14.0, −6.8) | 0.042 |
| Serum insulin (mIU/mL) | |||||
| Day 1 | 7.8 (5.7, 9.8) | 7.8 (5.3, 12.4) | |||
| Day 6 | 5.5 (3.2, 7.5)＊＊ | −2.0 (−4.3, −1.1) | 5.4 (4.5, 7.0)＊＊ | −2.3 (−5.6, −0.8) | 0.721 |
| Serum triglyceride (mg/dL) | |||||
| Day 1 | 91.3±40.9 | 94.7±35.7 | |||
| Day 6 | 72.0±28.9＊ | −19.4±33.5 | 71.0±29.3＊＊ | −23.7±30.1 | 0.606 |
| Serum LDL cholesterol (mg/dL) | |||||
| Day 1 | 106.5±25.3 | 107.0 (88.0, 131.3) | |||
| Day 6 | 98.3±27.5＊＊ | −8.2±12.3 | 80.5 (68.8, 115.0) ＊＊ | −17.8±10.4 | 0.013 |
| Serum HDL cholesterol (mg/dL) | |||||
| Day 1 | 63.5±14.2 | 63.0±12.3 | |||
| Day 6 | 60.6±14.2＊＊ | −2.9±4.6 | 61.3±12.8 | −1.7±4.3 | 0.330 |
| Serum uric acid (mg/dL) | |||||
| Day 1 | 5.3±1.1 | 5.6±1.4 | |||
| Day 6 | 6.3±1.4＊＊ | 1.0±0.8 | 6.2±1.6＊＊ | 0.7±0.5 | 0.015 |
| Urinary pH | |||||
| Day 1 | 6.5 (6.2, 6.6) | 6.4±0.3 | |||
| Day 6 | 6.1 (6.0, 6.4)＊＊ | −0.2±0.3 | 6.4±0.4 | 0.0±0.3 | 0.070 |
| GIAA (mEq/day) | |||||
| Day 1 | 12.8±10.9 | 12.0±12.9 | |||
| Day 6 | 8.1±9.6 | −4.9±12.5 | 14.1±13.4 | 2.1±12.6 | 0.075 |
| Urinary uric acid excretion (mg/day) | |||||
| Day 1 | 472.0 (411.6, 534.3) | 503.4 (476.2, 573.7) | |||
| Day 6 | 505.3 (454.3, 560.9) | 42.1 (−7.7, 103.0) | 567.4 (549.4, 630.9)＊＊ | 59.9 (−0.1, 94.7) | 0.390 |
Values are presented as mean±standard deviation or medians (25th to 75th interquartile range). ＊p＜0.05; ＊＊p＜0.01 vs. Day 1. Differences between days 1 and 6 were identified using a paired t-test or the Wilcoxon signed-rank test. Differences between groups F and FV were identified using a paired t-test or the Wilcoxon signed-rank test.
F: fish diet, FV: fish diet with adequate vegetable content, LDL: low-density lipoprotein, HDL: high-density lipoprotein, GIAA: net gastrointestinal absorption of alkali
| Male | Female | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Group F | Group FV | p-value Change (Δ) | Group F | Group FV | p-value Change (Δ) | |||||
| Change (Δ) | Change (Δ) | Change (Δ) | Change (Δ) | |||||||
| Plasma glucose (mg/dL) | ||||||||||
| Day 1 | 97.4±4.7 | 97.0 (93.0, 106.0) | 99.4±8.0 | 98.0 (93.0, 101.0) | ||||||
| Day 6 | 90.4±4.2 ＊＊ | -5.0 (-11.0, -3.0) | 89.0 (85.0, 94.0) ＊＊ | −9.0 (−15.0, −6.0) | 0.168 | 91.3±5.1 ＊＊ | −7.0 (−10.0, −2.0) | 88.0 (85.0, 92.0) ＊＊ | −8.0 (−13.0, −7.0) | 0.152 |
| Serum insulin (mIU/mL) | ||||||||||
| Day 1 | 7.0±2.0 | 7.0 (5.4, 10.3) | 9.1 (5.9, 12.9) | 8.0 (5.1, 13.4) | ||||||
| Day 6 | 5.1±2.0 ＊＊ | −1.7 (−2.9, −0.9) | 5.2 (4.1, 5.8) ＊ | −1.5 (−5.2, −0.5) | 0.894 | 5.2 (4.3, 8.3) ＊＊ | −2.7 (−5.3, −1.3) | 6.2 (4.7, 8.5) ＊＊ | −3.1 (−5.7, −0.9) | 0.477 |
| Serum triglyceride (mg/dL) | ||||||||||
| Day 1 | 104.4±43.1 | 102.2±29.0 | 78.3±35.6 | 87.3±41.4 | ||||||
| Day 6 | 82.0±26.9 | −22.4±38.8 | 81.7±28.2 | −20.5±38.2 | 0.898 | 61.9±28.4 | −16.4±28.9 | 60.3±27.4 ＊＊ | −27.0±20.7 | 0.248 |
| Serum LDL cholesterol (mg/dL) | ||||||||||
| Day 1 | 118.0±26.0 | 120.4±25.6 | 95.1±19.4 | 96.1±23.1 | ||||||
| Day 6 | 107.4±27.4 ＊ | −10.6±13.3 | 102.3±30.7 ＊＊ | −18.1±12.5 | 0.224 | 89.3±25.8 | −5.8±11.2 | 78.6±23.7 ＊＊ | −17.5±8.4 | 0.021 |
| Serum HDL cholesterol (mg/dL) | ||||||||||
| Day 1 | 57.0±10.2 | 59.5±11.9 | 69.9±15.1 | 66.6±12.3 | ||||||
| Day 6 | 53.2±10.9 ＊ | −3.8±5.0 | 56.3±11.4 ＊ | −3.2±3.8 | 0.723 | 68.0±13.6 | −1.9±4.3 | 66.4±12.7 | −0.3±4.5 | 0.313 |
| Serum uric acid (mg/dL) | ||||||||||
| Day 1 | 6.2±0.7 | 6.4±1.1 | 4.3 (4.0, 4.7) | 4.6 (4.0, 5.0) | ||||||
| Day 6 | 7.3±1.0 ＊＊ | 1.1±0.9 | 7.1±1.2 ＊＊ | 0.7±0.6 | 0.056 | 5.2 (4.6, 5.7) ＊＊ | 0.9±0.7 | 5.0 (4.6, 5.6) ＊＊ | 0.6±0.5 | 0.170 |
| Urinary pH | ||||||||||
| Day 1 | 6.2±0.3 | 6.3±0.3 | 6.6 (6.5, 6.7) | 6.5±0.3 | ||||||
| Day 6 | 6.1±0.3 | −0.2±0.3 | 6.2±0.3 | −0.1±0.3 | 0.817 | 6.1 (6.1, 6.5) ＊ | −0.2±0.3 | 6.6±0.4 | 0.1±0.3 | 0.004 |
| GIAA (mEq/day) | ||||||||||
| Day 1 | 9.0±8.3 | 8.3±11.2 | 16.6±12.2 | 15.8±13.9 | ||||||
| Day 6 | 5.7±4.9 | −3.3±11.2 | 6.5±10.2 | −1.8±10.1 | 0.830 | 10.2±12.3 | −6.4±14.0 | 21.7±12.2 | 5.9±14.0 | 0.023 |
| Urinary uric acid excretion (mg/day) | ||||||||||
| Day 1 | 501.4±98.5 | 562.5 (494.9, 578.7) | 460.9±66.7 | 479.8 (464.2, 503.8) | ||||||
| Day 6 | 537.9±95.2 | 31.0 (−19.9, 109.5) | 570.3 (540.5, 629.4) | 34.3 (−42.7, 114.3) | 0.790 | 489.8±106.2 | 46.9 (14.3, 102.1) | 559.0 (549.8, 635.4) ＊ | 76.7 (44.1, 90.9) | 0.286 |
Values are presented as mean±standard deviation or medians (25th to 75th interquartile range). ＊p＜0.05; ＊＊p＜0.01 vs. Day 1. Differences between days 1 and 6 were identified using a paired t-test or the Wilcoxon signed-rank test. Differences between groups F and FV were identified using a paired t-test or the Wilcoxon signed-rank test. F: fish diet, FV: fish diet with adequate vegetable content, LDL: low-density lipoprotein, HDL: high-density lipoprotein, GIAA: net gastrointestinal absorption of alkali
Table 3 shows the values of the evaluated metabolic indices in group FV. These results by sex are shown in Supplemental Table 3. Plasma glucose, serum IRI, TG, and LDL cholesterol concentrations for group FV on day 6 were significantly lower than on day 1. Serum UA concentrations and urinary UA excretion for group FV on day 6 were significantly higher than on day 1. No significant differences in urinary pH and GIAA were observed in group FV between days 1 and 6.
Changes from Day 1 in Metabolic Indices between Groups F and FVTable 3 shows the changes from day 1 in the metabolic indices of participants in groups F and FV. These results by sex are shown in Supplemental Table 3. ΔPlasma glucose and Δserum LDL cholesterol concentrations for group FV were significantly larger than those for group F (p=0.042, p=0.013, respectively). ΔSerum UA concentration for group FV was significantly smaller than those for group F (p=0.015). ΔUrinary pH and Δurinary GIAA for group FV tended to be smaller than those for group F (p=0.070, p=0.075, respectively). There were no significant differences in Δurinary UA excretion between groups F and FV.
In this study, we investigated the effect of a healthy diet containing adequate amounts of protein and vegetables on metabolic indices in healthy Japanese participants.
This study showed that Δplasma glucose and Δserum LDL cholesterol concentrations for group FV were significantly larger than those for group F (Table 3). The primary sources of dietary fiber are grains, fruits, and vegetables. In the DRIs for the Japanese, the RDAs of dietary fiber are 21 g/day and 18 g/day for males and females, respectively, for the age group of 18-64 years17). The Japan atherosclerosis society guidelines for the prevention of atherosclerotic cardiovascular diseases 2022 recommend increasing dietary fiber intake21). Previous studies have reported that higher intake of dietary fiber should be promoted to prevent CVD14) and that dietary fiber intake is associated with reduced incidence and mortality from several non-communicable diseases (e.g., coronary heart disease, stroke, and type 2 diabetes)22) and a reduced risk of mortality from CVD23). A previous study also reported that the risk reduction associated with a range of clinical outcomes was greatest when daily dietary fiber intake was 25-29 g22). In our study, the dietary fiber contents in group F were 22.0 g/day for males and 20.1 g/day for females, and those in group FV were 29.7 g/day for males and 27.8 g/day for females. Our study demonstrated that the adequate dietary fiber intake brought about by the increased consumption of vegetables in group FV contributed to the reduced plasma glucose and serum LDL cholesterol concentrations in the healthy participants. In addition, previous studies reported that intake of whole grains was associated with lower LDL cholesterol concentrations24) and that increasing oat sources of whole grain might be beneficial for lipid management25). Although this study focused on the effect of increased fiber intake from vegetables on metabolic indices, increased fiber intake not only from vegetables but also from whole grains may contribute to more improvement of metabolic indices.
A previous study identified the gut microbiota as an additional contributing factor to the pathophysiology of obesity26). Dietary fibers are metabolized by the microbiota in the cecum and colon, and the major products from the microbial fermentative activity in the gut are short chain fatty acids (SCFAs) (e.g., acetate, propionate, and butyrate)27). A previous study reported that the level of HbA1c decreased significantly and butyric acid concentrations increased significantly in the high fiber diet compared to the control diet28). Thus, SCFAs may also be associated with improved plasma glucose concentrations due to adequate dietary fiber intake from vegetables observed in this study.
We hypothesized that differences in PRAL between groups F and FV might also affect serum LDL cholesterol reduction. Mild metabolic acidosis is caused by eating patterns associated with a high dietary acid load1). Metabolic acidosis enhances glucocorticoid secretion and increases plasma and urine cortisol concentrations2). Previous studies reported that metabolic acidosis resulted in impaired tissue sensitivity to insulin and impaired glucose metabolism29), and that cortisol excess caused insulin resistance30, 31). The mechanism that increased cortisol production due to mild metabolic acidosis may affect metabolic indices has been suggested in previous reports2, 6, 30, 31) but remains poorly established. Furthermore, it has been reported that treatment of metabolic acidosis using base-producing fruits and vegetables rather than sodium bicarbonate was associated with lower LDL cholesterol concentrations32). Thus, adequate vegetable intake may have prevented metabolic acidosis in group FV, decreasing serum LDL cholesterol concentration.
A previous study has reported that a higher fish intake was associated with a reduced risk of coronary heart disease among middle-aged Japanese33). In this study, plasma glucose, serum IRI, TG, and LDL cholesterol concentrations on day 6 were significantly lower than on day 1 in both groups F and FV (Tables 3). One possible explanation for the improvement of these metabolic indices is the effect of the fish diet. In addition, most participants did not attain the RDAs or DGs for energy, the energy ratio of protein, and minerals (e.g., Ca, K, and Mg) (Supplemental Table 2). Thus, in group F, consuming the test meals may not have met the RDAs or DG for minerals but attained the RDA or DG for energy and energy ratio of protein, which may have improved these metabolic indices. However, we would like to emphasize that this study showed a significant improvement in metabolic indices (plasma glucose and serum LDL concentrations) in group FV, which attained the RDA for minerals, compared to group F. A previous study has reported that the combination of dietary factors (fruits, vegetables, fish, and salt) was associated with variation in CVD mortality risk14). Thus, this study’s results indicate that improving diet quality is effective in improving metabolic indices.
Previous studies have reported that alkalization of urine occurs by consuming alkaline foods (such as vegetables and fruits)18, 34, 35) and that urinary pH is an indicator of dietary acid-base load, i.e., fruit and vegetable intake and meat intake36). Additionally, it has been reported that adhering to dietary recommendations (increased GIAA and decreased acid load) may normalize urinary pH in stone formers37). In this study, Δurinary GIAA for group FV tended to be smaller than those for group F (p=0.075) although no significant difference in urinary GIAA was observed in group FV between day 1 and day 6 (Table 3). Thus, the changes in urinary GIAA following dietary intervention observed in this study are consistent with those reported previously. Our result suggests that adequate vegetable intake, even over five days, increases GIAA and improves urinary pH. However, the intervention was already started on day 1 and it is necessary to assess urinary GIAA before the intervention or during an equilibrium period to examine the effect of vegetable loading on GIAA.
In this study, there was a significant decrease in urinary pH and a nonsignificant change in urinary UA excretion after 5 days intervention in group F (diet with insufficient vegetables and more acidic diet), resulting in a significant increase in serum UA concentration. A previous study reported that urinary pH decreased and serum UA concentrations increased in acidic diet compared with those for alkaline diet18), and the changes observed in this study are consistent with the previously reported study. Thus, our result indicates that decrease in urinary pH due to acidic diet intake may be responsible for the increase in serum UA concentration.
In this study, Δserum UA concentration for group FV was significantly smaller than those for group F although serum UA concentration on day 6 was significantly higher than on day 1 in groups F and FV (Table 3). As purine contents are not listed in Japan’s Standard Tables of Food Composition38), we estimated the purine content of the test meals using the values listed in the “List of purine content in foods” section of the Guidelines for the Treatment of Hyperuricemia and Gout, 3rd Edition39). Purine content in group FV was about 70 mg/day higher than in group F, although the purines added to the test meals of group FV were vegetable-derived. The predominant purines in vegetables are adenine and guanine40). Although hypoxanthine increases serum UA concentrations41), vegetable-derived purines contain less hypoxanthine. A previous study reported that serum UA concentrations increased and urinary UA excretion decreased in acidic diet and that urinary UA excretion was higher in individuals on alkaline diet than in those on acidic diet, despite the higher purine content in the alkaline diet (acidic diet, 351 mg/day; alkaline diet, 494 mg/day)18). In group FV, urinary UA excretion on day 6 was significantly higher than that on day 1 (Table 3). Thus, we speculated that lowering PRAL through adequate vegetable intake might increase urinary UA excretion and suppress the increase in serum UA concentrations.
The PRAL of the test meals in group FV for both sexes (male; 11.1 mEq/day, female; 9.3 mEq/day) were generally within the range of values observed in other Japanese populations (4-11 mEq/day)6, 7, 10, 42). Although we limited the main dish of the test meal to fish, various foods (e.g., meat, soy products) are consumed as a main dish daily. A previous study reported that PRAL showed inverse associations with potatoes, pulses, vegetables, fruit, and dairy products6). Future evaluations based on daily diets will help to assess the effect of PRAL on metabolic indices.
Another interesting point is that there were sex differences in the changes of metabolic indices due to vegetable intake, which was less variation in male compared to female (Supplemental Table 3). According to the data calculating from the BDHQ (Supplemental Table 2), the PRAL was lower in female than in male (male; 14.8±5.9 mEq/day, female; 7.3±9.2 mEq/day), which is in broad agreement with several previous studies6, 36). These differences in daily eating habits between the sexes may be responsible for the differences in the changes of metabolic indices. In addition, a previous study reported that urinary pH and GIAA were higher in female than in male after the intake of identical meals43), and future studies are needed to investigate sex differences in response to acid–base load.
The strength of this study is that it evaluated the effects of differences in PRAL on metabolic indices under strictly diet-controlled conditions, using two test meals that differed only in vegetable content. In addition, this study was conducted under a realistic alkaline load with dietary content per the DRIs. This study exhibited several limitations that should be considered. Firstly, we could not undertake formal power calculation. We calculated the sample size based on a previous study19), but the type and duration of the intervention was different from this study. Although we confirmed sufficient power (1−β=1.00) by a post hoc power analysis for the serum LDL concentrations within group FV (α=0.05) (calculated using G*Power 3.1), formal power analysis should be performed in future studies. Secondly, the intervention was already started on day 1 and the 24-h urine data on day 1 have already been affected by the intervention. We did not assess urinary metabolic indices (urinary pH, GIAA, and UA excretion) before the intervention as baseline data in this study. Therefore, future studies need to assess urinary metabolic indices (urinary pH, GIAA, and UA excretion) before the intervention or during an equilibrium period to examine the effect of vegetable loading on urinary metabolic indices (urinary pH, GIAA, and UA excretion). Thirdly, 72.7% of the participants were in their 20s, and we need to evaluate individuals of a wide range of ages in the future. Lastly, our study population included healthy individuals. It is necessary to explore the effects of PRAL on metabolic indices in participants with impaired glucose tolerance or dyslipidemia. A large cohort is required to determine the effect of a healthy diet containing adequate amounts of protein and vegetables.
This study demonstrated that a healthy diet containing adequate protein and vegetables reduced acid load and improved plasma glucose and serum LDL concentrations in healthy Japanese participants, despite a realistic alkaline load (vegetables at 350 g/day). Furthermore, this study suggests that a healthy diet containing adequate amounts of protein and vegetables may be one of the effective dietary interventions for reducing the risk of cardiovascular diseases.
We are grateful to the volunteers who participated in the study.
The authors received funding support from the Food and Healthcare Open Innovation Project of Shizuoka Prefecture (to TG). This study was also funded by the Grant-in-Aid for Young Scientists 20K13802 from Japan’s Ministry of Education, Culture, Sports, Science, and Technology (to YK-S).
The authors received funding support from the Food and Healthcare Open Innovation Project of Shizuoka Prefecture (to TG). This study was also funded by the Grant-in-Aid for Young Scientists 20K13802 from Japan’s Ministry of Education, Culture, Sports, Science, and Technology (to YK-S). The authors declare that they have no competing interests.
