Dietary total, plant, and animal protein intake in relation to cardiovascular outcomes and inflammatory factors in elderly men: A cross‐sectional study

Abstract The source and amount of protein intake may influence cardiovascular and inflammatory risk, especially in elders who are often more vulnerable. However, findings on elders have been contradictory. Therefore, we examined the association between dietary total, plant, and animal protein intake in relation to cardiovascular outcomes and inflammatory factors in elderly men. The present cross‐sectional study included 357 elderly men. A validated and reliable food frequency questionnaire (FFQ) was used to assess dietary intake. All biochemical factors including triglycerides (TG), fasting blood sugar (FBS), high‐sensitivity C‐reactive protein (hs‐CRP), interleukin 6 (IL6), and tumor necrosis factor‐α (TNF‐α) were measured. Waist circumference (WC) and blood pressure (BP) were also assessed. A significant inverse association was found between animal protein intake and systolic blood pressure (SBP; OR: 0.62; 95% CI: 0.42, 0.91; ptrend = .014). There were significant inverse associations between plant protein intake and WC (OR: 0.34; 95% CI: 0.17, 0.68; ptrend < .001), FBS (OR: 0.51; 95% CI: 0.29, 0.89; ptrend = .018) and Hs‐CRP (OR: 0.39; 95% CI: 0.21, 0.70; ptrend = .002). Moreover, significant inverse associations were also found between total protein intake and SBP (OR: 0.54; 95% CI: 0.33, 0.86; ptrend = .010) and total protein and Hs‐CRP (OR: 0.50; 95% CI: 0.28, 0.88; ptrend = .015). In elderly men, a high dietary intake of plant protein was associated with lower odds of having high WC, FBS, and Hs‐CRP. In addition, high dietary intake of animal protein was associated with higher odds of having a high SBP level, which was explained by higher intake of dairy products.

especially in older populations (Mozaffarian et al., 2015).Worldwide, the number of people with CVDs has increased from 271 million in 1990 to 523 million in 2019 (Roth et al., 2020).Approximately twothirds of CVD-related deaths occur in the elderly (American Heart Association, 2013).Estimates predict that CVD will be the cause of more than 23 million (about 30.5%) deaths by 2030 worldwide (Lozano et al., 2012;Mozaffarian et al., 2015).According to The Global Burden of Disease (GBD's) previous reports in 2010 and 2015, CVD is responsible for 20%-23% of the burden of diseases in Iran (Naghavi et al., 2014;Namazi Shabestari et al., 2015;Shams-Beyranvand et al., 2017).Moreover, based on national statistics in 2015, the prevalence of CVDs among elderly from Iran was 39.9% (Shamsi et al., 2017).
The increased prevalence of CVDs is thought to be related to socioeconomic and cultural changes, nutrition, insufficient physical activity, and increased metabolic and physical risk factors (Sarrafzadegan & Mohammadifard, 2019).Therefore, much interest exists in the role of nutrition in prevention and treatment of CVDs (Sala-Vila, 2015;Willet, 1999).According to Tappia et al., nutritional factors are responsible for approximately 40% of all cases of CVDs (Willet, 1999).The source and amount of protein in one's diet could affect cardiovascular risk factors (Chesney, 2015;Naghshi, 2020).
According to a review, CVD risk could be decreased by dietary patterns that provide more plant protein sources (including unprocessed animal proteins), compared with the typical American diet (Chesney, 2015).Another study suggested that both processed and red meat were associated with increased CVD risk (Riccardi et al., 2022).In addition, a cohort study of 29,682 participants revealed that consuming plant proteins and fish were associated with lower risk of CVD incidence (Zhong et al., 2021).A recent meta-analysis on the effect of animal and plant proteins on CVD risk factors showed that consumption of plant proteins may have protective effects against CVD risk factors (Lamberg- Allardt et al., 2023).
However, it has been suggested that some groups of animal proteins like poultry and dairy products may be more effective in controlling CVD risk factors than some plant proteins like nuts (Zhubi-Bakija et al., 2021).
Thus, the association between source and amount of protein intake and CVD risk factors are still unclear.In addition, few studies have been conducted among elderly.The aforementioned reasons justify evaluating the association between the amount and sources of protein intake and cardiovascular outcomes and inflammatory factors in an elderly population.In addition, cardiovascular disease predominantly affects men; that is, for most age groups, rates of cardiovascular disease in men are higher than in women (Albrektsen et al., 2016).Therefore, the objective of this study was to examine the cross-sectional association between the amount and sources of protein intake and cardiovascular risk factors in elderly men.

| Study population
In this cross-sectional study, 365 men were included from health centers in southern Tehran, Iran (March to August 2017).Ethics approval for the study was obtained from Tehran University of Medical Sciences (TUMS), as health centers in south of Tehran are supervised by TUMS (grant number: 48040).Men who were referred to health centers for primary healthcare were contacted by staff for inclusion in the present study.All participants were asked to provide written informed consent to be included in the study.Inclusion Hypertension was the main dependent variable used to calculate study sample size (as the highest sample size would be obtained using this variable) (Mehrabani et al., 2017).A total sample size of 340 participants was estimated.Clustered random sampling was used to determine the number of participants selected from each health center.After data collection (n = 365), participants who had very low calorie intake (<800 kcal/day) or high calorie intake (>4200 kcal/day) were excluded and 357 participants remained for statistical analysis.

| Dietary assessment
All participant nutritional information was obtained using 168-item semiquantitative food frequency questionnaire (FFQ) through face- N-Squared Computing), which was designed for Iranian foods, was used to compute nutrient intake (Abshirini et al., 2019).Validity and reliability of the FFQ to use among Iranian (Bijani et al., 2018;Malekahmadi et al., 2016) and non-Iranian elderly (Smith et al., 1998) have been reported by many studies.We calculated the protein content of each food item based on USDA nutrition facts and then categorized them as animal protein (AP), plant protein (PP), and total protein (TP).Ltd., Belfast, UK) was applied to determine the plasma concentration of high-sensitive C-reactive protein (hs-CRP).Other inflammatory biomarkers were determined using the enzymelinked immunosorbent assay (ELISA) method (Boster Biological Technology for IL-6 and TNF-a, China).

| Anthropometric assessment
Anthropometric indices like body weight, height, and waist circumference were assessed by a trained assistant.A portable digital scale (SECA 813; Seca) was used to measure participant weight (measurement precision of 100 g) when wearing light clothing.To measure participant height, a tape with 0.5 cm of precision was used while participants were asked to stand against a wall in a normal position without moving.For waist measurements, a tape measure within 0.5 cm of precision was used to measure at the middle point between the top of the hips and bottom of the ribs was selected.BMI was calculated as participant's weight (kg) divided by height (m 2 ).

| Assessment of other covariates
A valid and reliable questionnaire was used to assess the socioeconomic status (SES; Mozaffari et al., 2019).The questionnaire considered occupation, education, vehicle and house ownership, modern household appliances, the number of rooms and family members, and trips through the previous year.For other covariates, a demographic questionnaire was used, which considered age, marital, and smoking status.Participants were asked about chronic diseases such as diabetes, hyperlipidemia, myocardial infarction, hypertension, stroke, angina, and thyroid disease.They were also asked about their drugs such as heart disease drugs, diabetes drugs, and lipid-lowering and thyroid drugs.
Participants were asked to be seated for about 10 minutes, and blood pressure (BP) was measured twice within a one-minute interval.Their bladders were emptied.Moreover, no caffeinated beverages consumed and no exercises were done within 1 h before the measurement.The average time for activities was presented as (MET-h/week).
General characteristics were described for animal, plant, and total protein turtles.Chi-squared test was used to test the distribution of elderly men with continuous variables, and one-way analysis of variance (ANOVA) was used to test the distribution of elderly men with categorical variables within the tertiles of animal, plant and total protein intakes.In addition, one-way analysis of covariance (ANCOVA) with energy intake adjustment was used to compare the distribution of nutrients and food group intakes across tertiles of animal, plant, and total protein intake.One-way analysis of covariance (ANCOVA) was also used to compare cardiovascular outcomes and inflammatory parameters across tertiles of animal, protein, and total protein intake in both crude and adjusted models.
To conduct binary logistic regression, participants were categorized for cardiovascular risk factors following the reported cutoffs from the Adult Treatment Panel III (ATP-III; Maas & Appelman, 2010): WC > 88 cm, BP ≥ 130/85 mmHg, FBS ≥ 110 mg/dL, and TG ≥ 150 mg/ dL.For serum hs-CRP level, previous cutoffs for Iranian elderly were used (hs-CRP >2 mg/L) (Rashidi Pour Fard et al., 2015).To assess the association between animal, plant, and total protein intake and cardio-metabolic risk factors, binary logistic regression in crude and two adjusted models were applied.The following confounders were adjusted in the final model: age, marital status, physical activity, socioeconomic status, smoking status, diseases and drugs, body mass index (BMI), fat, and carbohydrate intake.Plant protein was included as an additional cofounder in models with animal protein, and animal protein was included as an additional cofounder in the models with plant protein.Adjusted cofounders were selected based on previous relevant studies as well as univariate models, which were conducted for each outcome, and associations with p < .2 were considered as possible cofounders.The first tertiles of dietary intake were the reference group.SPSS software version 26.0 was used to analyze the data, and p values < .05were considered statistically significant.had higher age (p = .02),WC (p = .001),and BMI (p = .03)were more likely to be nonsmokers (p = .04)and not consume certain drugs such as lipid-lowering drugs (p = .016)and heart disease drugs (p < .001).

| RE SULTS
Participants in the highest tertile of PP had lower age (p < .001),lower WC (p = .02),and lower weight (p = .01).They also were less likely to use certain drugs like antidiabetic drugs (p = .007)and heart disease drugs (p = .020).In addition, participants who were in the highest tertile of TP were more likely to be nonsmokers (p < .001) and not consume certain medications such as antidiabetic drugs (p = .015),lipid-lowering drugs, and heart disease drugs (p < .001).
As a significant inverse association between AP and SBP was found, we conducted another logistic regression to find the potential reason.Odds ratios (OR) and 95% confidence intervals (CI) for SBP and DBP across tertiles of different animal protein groups are presented in Table 5 and Table 6.We found that the consumption of dairy products had a significant inverse association with SBP (OR: 0.61; 95% CI: 0.41, 0.91; ptrend = .017).In addition, there was a  significant inverse association between egg and DBP (OR: 0.57; 95% CI: 0.37, 0.87; ptrend = .010).

| DISCUSS ION
The present study showed that the total amount and sources of protein may have effective roles on cardiovascular outcomes and inflammatory factors as higher intake of protein had a significant inverse association with SBP and Hs-CRP.In addition, higher intake of PP was significantly associated with lower WC, FBS, and Hs-CRP.
Higher intake of AP was associated with lower SBP, and the association was due to higher intake of dairy products (Table 6).
Associations between the amount and source of protein intake and cardiovascular outcomes and inflammatory factors have been explored in studies that were mostly conducted in adults; a tape within 0.5 cm of precision was used but not specific to the elderly.
Available evidence about the association between diet and disease in young adults is not appropriate for the elderly because of the differences in distribution and quantity of body fat mass (Guarner & Rubio-Ruiz, 2015).
In the present study, a significant positive association between higher dairy intake and SBP was found.In line with our finding, respectively (FitzGerald et al., 2004;FitzGerald & Meisel, 2000).
In addition, the amino acid content of dairy products may be another possible factor responsible for their hypotensive effects (Jennings et al., 2015;Poggiogalle et al., 2019;Teymoori et al., 2017).
According to Landi et al. (2021), the most abundant amino acids in milk are glutamic acid and tyrosine.In a study conducted by Stamler et al. (2009), a significant relationship between dietary glutamic acid and decreased blood pressure was found.Glutamate is included in the glutathione molecule, which has antioxidant activity and may play a significant role in blood pressure homeostasis (Vasdev et al., 2006).
Moreover, Altorf-van der Kuil et al. (2013) found that higher intake of was tyrosine related to a 2.4 mm Hg lower SBP.Tyrosine can be a precursor for norepinephrine synthesis and modulate norepinephrine levels, which then has effects on the sympathetic tone in the vasculature (Siomkajło et al., 2017).
In a present study, we found a significant inverse association between higher intake of PP and WC, FBS, and Hs-CRP.In accordance with our findings, Park et al. (2018) conducted a study on Korean elders and found a significant inverse association between consuming plant protein and WC.In another study conducted by Lin et al. (2011) in Belgium, in both males and females, a significant inverse association between plant protein intake and WC was found.Furthermore, in a cohort study by Shang et al. (2017) also found an inverse association between plant protein intake and WC.In another cohort study by Hruby and Jacques (2018), a favorable association between PP intake and WC was found.However, in another study conducted with both men and women, Halkjaer et al. (2011) found no significant association between plant protein intake and WC.In line with our findings of lowered FBS by PP intake, Comerford and Pasin (2016) found that higher intake of plant protein was significantly associated with a decreased risk of type 2 diabetes.Another study indicated that replacing sources of animal with plant protein may improve glycemic control in people with diabetes (Viguiliouk et al., 2015).Plantbased proteins are high in amino acids like arginine and pyruvigenic amino acids (glycine, alanine, and serine) (Krajcovicova-Kudlackova et al., 2005).In a study conducted by Vangipurapu et al. (2019), they found that glycine and serine were associated with improved insulin sensitivity.Moreover, they were associated with significant increases in insulin secretion (DI), which may be responsible for lowered blood glucose.In addition, another study found that a plant-based diet which leads to limited leucine and histidine can improve body composition and reduce insulin resistance (Kahleova et al., 2018).
Hruby and Jacques (2019) found that there was a significant inverse association between PP intake and Hs-CRP.Moreover, they observed the same association between total protein intake and Hs-CRP, which is also in line with the results of the present study.
In a meta-analysis conducted by Haghighatdoost et al. (2017), a significant inverse association between vegetarian diet and Hs-CRP was found.Another study conducted on adults with chronic kidney disease showed an inverse association between consuming plant protein and levels of Hs-CRP (Aycart et al., 2021).Various biological factors affect the levels of inflammatory factors, among which are the compounds found in plant sources.Bioactive peptides from plant sources have the ability to reduce inflammatory factors by regulating metabolic cycles that cause inflammation, such as the MAPK (mitogen-activated protein kinase) pathway and NF-Κb, which is the key pathway of protein phosphorylation (Liu et al., 2022).

| CON CLUS IONS
In the present study among elders, it was found that higher intake of PP was more beneficial than AP, because it had more favorable associations with cardiovascular and inflammatory outcomes including WC, FBS, and Hs-CRP compared with AP which was associated with lower blood pressure.However, it should be considered that the higher dairy product intake was responsible for the association.

| LIMITATI ON AND S TRENG TH S
To the best of our knowledge, this is the first study to investigate the relationship between dietary total, animal, and plant protein criteria were being: (a) an Iranian male; (b) over 60 years old; (c) no self-reported previous diagnosis of malignant diseases (e.g., cancer); and (d) no change in usual diet as a result of disease or per dietitian recommendation.
to-face interviews by a trained nutritionist.Dietary intakes were changed from serving sizes and household measures to grams.A modified version of the NUTRITIONIST IV software (version 7.0;
Hidayat et al. (2017) conducted a meta-analysis about the effects of milk proteins on blood pressure and found that milk protein significantly decreased SBP by −3.33 mm Hg.The possible mechanism for the hypotensive effects of milk proteins may be the inhibition of angiotensin-converting enzyme activity(Fekete et al., 2013;Nonaka et al., 2014;Pal & Radavelli-Bagatini, 2013), which catalyzes the conversion of angiotensin I to angiotensin II and leads to arterial vasodilation.Lactokinins and casokininsare angiotensin-converting enzyme inhibitory peptides present in whey and casein (proteins of milk),

TA B L E 1 General characteristics of participants across the tertiles of AP, PP, and TP. Characteristics AP PP TP T1 T2 T3 p value a T1 T2 T3 p value a T1 T2 T3 p value a <34.76 34.76, 47.06 >47.06 <21.57 21.57, 27.60 >27.6 <66.76 66.76, 90.84 >90.84 n 121 117 119 119 122 115 118 120 119
Participants' general characteristics across tertiles of AP, PP, and TP are shown in Table1.Participants in the highest tertile of AP Energy-adjusted dietary intakes across tertiles of AP, PP, and TP.Cardiovascular risk factors across tertiles of AP, PP, and TP.Value < .05 was considered significant.Values are based on Mean ± SE.Crude: Not adjusted for any variables.Model 1: The model was adjusted for age, energy intake, marital status, socioeconomic status, physical activity, smoking, and BMI.Model 2: Model 1 + diseases and drugs, fiber, total fat, and total carbohydrate.Abbreviations: AP, animal protein; DBP, diastolic blood pressure; FBS, fasting blood sugar; HDL-C, high-density lipoprotein cholesterol; hs-CRP, high-sensitive C-reactive protein; IL-6, interleukin-6; LDL-C, low-density lipoprotein cholesterol; PP, plant protein; SBP, systolic blood pressure; TC, total cholesterol; TG, triglyceride; TNF-α, tumor necrosis factor-alpha; TP, total protein; WC, waist circumference.Multivariate adjusted odds ratios and 95% confidence intervals for blood glucose, lipid profile, and inflammatory factors across tertiles of AP, PP, and TP.Adjusted for age, smoking, physical activity, socioeconomic status, marital status, and energy.Adjusted for age, smoking, physical activity, socioeconomic status, marital status, energy, disease, antidiabetic drugs, thyroid drugs, and heart disease drugs.Multivariate adjusted odds ratios and 95% confidence intervals for blood pressure across tertiles of eggs, poultry, and fish.Adjusted for age, smoking, physical activity, socioeconomic status, marital status, and energy.Adjusted for age, smoking, physical activity, socioeconomic status, marital status, energy, disease, antidiabetic drugs, thyroid drugs and heart disease drugs.Multivariate adjusted odds ratios and 95% confidence intervals for blood pressure across tertiles of red meat, processed meat, and dairy products.
Note: p value < .05 was considered significant.Values are based on average ± standard deviation or reported percentage.Analysis of variance ANOVA for quantitative data and Chi-2 test for qualitative data have been used.Abbreviations: AP, animal protein; BMI,body mass index; PP,plant protein; TP, total protein; WC, waist circumference.aCalculated by analysis of variance ANOVA for quantitative data and Chi-2 test for qualitative data.TA B L E 1 (Continued)TA B L E 2Note: p Value < .05 was considered significant.All the variables, except energy, adjusted for energy intake.Abbreviations: AP, animal protein; PP, plant protein; TP, total protein.aCalculated by analysis of variance (ANOVA) for energy intake and multivariate analysis of covariance (ANCOVA) for other dietary variables.bMean ± SE.TA B L E 3 TA B L E 3 (Continued) TA B L E 4 a p For trend Calculate by logistic regression.bc TA B L E 5 a p For trend Calculate by logistic regression.bc TA B L E 6 a p For trend Calculate by logistic regression.bAdjusted for age, smoking, physical activity, socioeconomic status, marital status, and energy.cAdjustedfor age, smoking, physical activity, socioeconomic status, marital status, energy, disease, antidiabetic drugs, thyroid drugs, and heart disease drugs.