Association Between Protein Intake and Mortality in Hypertensive Patients Without Chronic Kidney Disease in the OLD-HTA CohortNovelty and Significance
Protein intake may have some benefits on reducing blood pressure and cardiovascular events, but their effects are still debated. The objective of this study was to test the prognostic value of protein intake assessed by 24-hour urinary urea in a cohort of hypertensive patients with preserved renal function. A total of 1128 hypertensive patients were followed according to tertile of protein intake adjusted for ideal body weight: <0.70, 0.70 to 0.93, and >0.93 g/kg. Baseline characteristics (mean±standard deviation) were age 45.1±13.2 years, systolic/diastolic blood pressure 185±32/107±20 mm Hg, and estimated glomerular filtration rate 82±32 mL/min. After 10 years of follow-up, 289 deaths occurred, 202 of which were of cardiovascular cause. After adjustment for major cardiovascular risk factors, patients in the second and third tertiles of protein intake had a decreased risk of all-cause death (hazard ratio [95% confidence interval], 0.71 [0.56–0.91]) and cardiovascular death (0.72 [0.54–0.96]), but not of stroke death (0.72 [0.41–1.28]) in comparison to patients in the low protein intake tertile. Normal–high protein intake was associated with a better outcome in a subset of the population: younger patients, low salt intake, without aortic atherosclerosis, or previous cardiovascular events (Pinteraction<0.10 for all). Hypertensive patients having a protein intake >0.7 g/kg ideal body weight, particularly those at low risk, had lower all-cause and cardiovascular mortality rates. Physicians may encourage hyper tensive patients to have normal or high protein diet in addition to low salt consumption, moderate alcohol consumption, and regular physical activity.
Appropriate lifestyle changes are recommended in current practice guidelines as the cornerstone of prevention for hypertension.1 Salt restriction, moderation of alcohol consumption, high consumption of fruits and vegetables, weight reduction, and regular physical exercise have demonstrated their capacity to reduce blood pressure (BP).2 In patients with chronic kidney disease, a low protein diet (0.6–0.7 g/kg per day) reduces renal death.3,4 Much less is known about the effect of protein intake in hypertensive patients without chronic kidney disease; although it may be associated with changes in BP, the results are controversial. In cross-sectional studies and in short-term randomized controlled trials, high protein intake has been associated with BP reduction,5 but results testing the link between protein intake and incidence of hypertension are also controversial.6,7 Protein intake may also be associated with cardiovascular events, with a high consumption associated with a protective effect against stroke in the general female population.8 A frequent limitation of these studies is the various and often unreliable methods in which protein intake is quantified, with the use of food frequency questionnaires, 24-hour dietary recall, and urinary urea concentration on a spot urine,9 whereas the current gold standard to assess protein intake is 24-hour urinary urea excretion.
The recommended amount of protein to be consumed by healthy adults is currently 0.8 to 1.0 g/kg, but protein intake in western countries currently far exceeds normal requirements (1.3–1.4 g/kg).10 To our knowledge, the prognostic significance of protein intake in patients with hypertension has never been properly addressed. Thus, the objective of the present study was to test the prognostic value (in terms of all-cause, cardiovascular, and stroke death) of protein intake assessed by 24-hour urinary urea in a cohort of hypertensive patients. As some conditions, such as renal function and cardiovascular remodeling, may exert a modifying effect, the interaction with several metabolic or organ damage variables was also tested.
The OLD-HTA Lyon cohort has been described.11 Briefly, 1152 patients were hospitalized between 1969 and December 1976 in the Cardiology Department at Louis Pradel Hospital (Lyon, France) for a work-up of their hypertension. Twenty-four patients were lost to follow-up and were excluded, leaving a study population of 1128 hypertensive inpatients with a 24-hour urinary collection form. Of these patients, 1092 also had a spot urine collection and 539 completed a food frequency questionnaire.
All patients provided oral consent to participate in the study, in accordance with the French regulation prevailing in the 1970s. The study was approved by the Commission Nationale Informatique et Liberté. Under French law, as mentioned in several published technical notes in line with European directives, only the approval of the Commission is required for single-center observational usual-care studies, such as this one. The vital status query was approved by national authorities before data extraction by the Institut National de la Statistique et des Etudes Economiques.
A standardized form was completed for all patients, which collected data on various morphometric characteristics, risk factors for cardiovascular events (smoking status, alcohol intake, salt consumption, etc), history of cardiovascular disease, current medication, and known symptoms. A food frequency questionnaire was also completed for half of the cohort.
Smoking status was based on current tobacco consumption or consumption stopped <5 years previously. BP was measured with a manual sphygmomanometer with the patient in the supine position. Systolic BP, diastolic BP, and pulse pressure were the average of 6 measurements. A 12-lead ECG was performed with the patient in the supine position. An ophthalmoscopic fundus examination of the right eye was performed after pharmacological pupil dilation to detect signs of hypertensive retinopathy. A consensual classification of hypertensive retinopathy was made by 2 trained cardiologists according to the 4-grade classification of Keith, Wagener, and Barker.12
An overnight fasting blood sample was drawn for hemogram and plasma measurements (electrolytes, creatinine, glucose, and total cholesterol). Diabetes mellitus was retrospectively defined as either a fasting glucose measuring ≥1.26 g/L (≥7.9 mmol/L) on 2 separate occasions or current use of antidiabetic medication. Renal function was estimated using the Modification in Diet in Renal Disease formula.
Between the second and the third day of hospitalization, 24-hour urine (n=1128) and a spot urine (n=1092) were collected and urinary sodium, potassium, urea, and albumin measured.
Previous cardiovascular diseases included history of heart failure (clinical or chest X-ray findings, such as dyspnea, edema, cardiomegaly, or pulmonary congestion), coronary artery disease (clinical findings, such as angina pectoris or myocardial infarction, or Q wave on ECG), peripheral artery disease (walking impairment or pain at rest), and stroke (clinical findings). Target organ damage was defined as electric left ventricular hypertrophy in the case of a Sokolow index >3.5 mV, a simplified 3-grade classification for hypertensive retinopathy (none for grade 0, mild for grades 1 and 2, and severe for grades 3 and 4)12 and albuminuria >300 mg/d assessed on 24-hour analysis.
Detailed data on aortography have been described.13,14 In brief, patients were classified according to a simplified 2-modality score: absence of atherosclerosis (aortic atherosclerosis) when aortography showed absent or mild atherosclerosis and presence of atherosclerosis for patients with moderate or severe lesions.
Assessment of Protein Intake
Total protein intake (PInt) was estimated from 24-hour urinary urea excretion by the method of Maroni et al.15 We used the following equation to determine PInt in g/24 hour:
Urinary urea is expressed in mmol/24 hour, proteinuria in g/24 hour, and body weight in kg. PInt was finally adjusted for ideal body weight (IBW) to standardize this protein intake and is expressed in g/kg per day. The IBW was derived from the real height using a body mass index (BMI) value of 22 kg/m2 as the reference. Multivariable Cox regression model was also performed with real body weight. In a subset of patients with spot urine available, protein intake was estimated based on crude urinary urea concentration in mmol/L.
The food frequency questionnaire used to assess diet at baseline encompassed 5 items. In this questionnaire, participants detailed how often they consumed bread, cheese, ham, or other delicatessen, meat, and milk each week. Protein intake was calculated according to the protein content and the frequency of consumption of these foodstuffs using composition values from the ANSES (Agence Nationale de Sécurité Sanitaire alimentation, environnement, travail) database (https://www.anses.fr/fr/content/les-proteines).
Assessment of Outcomes
Deaths at 10 years of follow-up were obtained from the Répertoire National d’Identification des Personnes Physiques (a directory maintained by the Institut National de la Statistique et des Etudes Economiques). A follow-up of 10 years was selected to balance the need to have enough events to provide the statistical power to assess a meaningful association, while avoiding a weakening of the determinant effect with time. Causes of death were then coded from the death certificates, as provided by INSERM SC8, according to the International Classification of Diseases, Ninth Revision. All subjects not officially declared dead were considered to be alive at the end of follow-up.
The end points used in this study were all-cause death (cardiovascular and noncardiovascular, including sudden death), cardiovascular death (from cerebrovascular disease, myocardial infarction, heart failure, or renal death), and stroke death as classified by the French national CépiDC (Center d’Epidémiologie sur les Causes Médicales de Décès).16
Continuous variables with close to normal distributions are summarized as mean±standard deviation. Continuous variables with skewed distributions are summarized as median (interquartile range). Categorical variables are expressed as percentages.
Appropriate tests (analysis of variance or χ2) were used to compare the characteristics of subgroups. Correlations were assessed with a linear regression analysis (Pearson’s coefficient of correlation r) after logarithmic transformation, if appropriate. Multiple regression analysis (forward stepwise) included variables with a P value <0.10 in univariate analysis.
The prognostic value of PInt/IBW was first examined in the whole cohort using tertiles (<0.70, 0.70–0.93, and >0.93 g/kg). The times to all-cause, cardiovascular, and stroke death were analyzed by Kaplan–Meier curves, and these curves were compared using the log-rank test. In a second step, adjusted hazard ratios for PInt/IBW were calculated as categorical variables (tertiles and the recommended threshold of protein intake at 0.8 g/kg by public health17) and as a continuous variable in univariate analysis and in a multivariable Cox regression model. The proportional hazard hypothesis was tested by introducing a variable-by-time interaction into the Cox regression model. Two multivariable Cox regression models were used: model 1 was built with potential confounding dietary variables such as plasma total cholesterol, plasma triglycerides, salt consumption, 24-hour urinary potassium excretion, and fasting glucose; model 2 was adjusted for the usual cardiovascular risk factors, including age, sex, systolic BP, smoking status, diabetes mellitus, total cholesterol, estimated glomerular filtration rate, BMI, antihypertensive treatment, and previous cardiovascular disease.
Sensitivity analyses were performed with the 2 same multivariable Cox regression models, after the exclusion of patients with secondary hypertension (n=856 remaining), and after exclusion of patients who died in the first 2 years (n=1034 remaining).
The modifying effect of the following variables on the prognostic value of PInt/IBW in model 2 was assessed with (1) demographic variables such as sex and age (median value at 46 years); (2) cardiovascular variables, that is, systolic BP (median value at 180 mm Hg), diastolic BP (median value at 107 mm Hg), aortic atherosclerosis condition, hypertensive retinopathy, previous cardiovascular disease; and (3) metabolic variables, that is, diabetes mellitus condition, BMI (normal or overweight patients with a threshold value at 25 kg/m2), salt intake (median value at 6 g/d), and total cholesterol (median value at 2.2 g/L). Interactions were tested for all these conditions.
The analyses were performed with the SPSS 20.0.0 program (SPSS, Chicago). A P value of <0.05 was considered as statistically significant, except for the test of interaction in which case a P value of <0.10 was retained for statistical significance.
Patient Baseline Characteristics and Outcomes
As indicated in Table 1, BP was markedly elevated (180/107 mm Hg) and a history of cardiovascular disease was present in 27.6% of patients. Half of the patients were receiving at least one antihypertensive drug at baseline: thiazide diuretics (34.5%), centrally acting drugs (30.3%), antialdosterone (13.2%), and β-blockers (3.7%). With respect to underlying cause, 76.8% of patients had essential hypertension, 5% had renal artery stenosis, and 5% had renal parenchymal disease, whereas primary aldosteronism and aortic coarctation were encountered in <3% of cases each.
Table 1 shows the classification into tertiles of patients according to their PInt/IBW: first tertile <0.70 (0.51±0.15), second tertile 0.70 to 0.93 (0.82±0.07), and third tertile >0.93 (1.24±0.37) g/kg/d. These subgroups illustrated, respectively, baseline low, normal, and high protein diets. Baseline characteristics in term of demographics, cardiac, and medical data were similar among tertiles. An increased protein intake was associated with higher salt and potassium intake consumption and a lower albuminuria.
PInt/IBW was slightly correlated with spot urine protein intake adjusted for IBW (r=0.218, P<0.001, n=1092) and with food frequency questionnaire protein intake assessment adjusted for IBW (r=0.105, P=0.015, n=539; Figure S1 in the online-only Data Supplement). PInt/IBW was statistically positively correlated with BMI, renal function, kalemia, total cholesterol, plasma urea, and 24-hour urinary sodium excretion and negatively correlated with plasma creatinine (Table S1). In multivariable regression analysis, all of these variables except kalemia remained correlated with PInt/IBW (Table S2).
After a 10-year follow-up, there were 289 deaths, 202 of which were from a cardiovascular cause (including 50 acute stroke deaths and 25 renal deaths).
Survival Analysis in the Whole Cohort
As shown in the Kaplan–Meier curves, after 10 years of follow-up, the survival rates decreased for patients in the lowest tertile of PInt/IBW for all-cause mortality (P=0.008; Figure 1A) and for cardiovascular mortality (P=0.029; Figure 1B). No significant difference was observed for stroke mortality (P=0.29; Figure 1C).
Table 2 shows the results from the univariate and multivariable Cox regression analyses for all-cause, cardiovascular, and stroke death. Overall, patients in the second and third tertiles had a decreased risk of all-cause and cardiovascular death in comparison with those in tertile 1. No significant effect was observed for stroke deaths. After adjustment, the results were essentially the same, albeit some limited statistical differences were observed for stroke mortality (only in multivariable model 1). Similar results were observed after adjustment for the real body weight of patients (Table S3).
After exclusion of patients with secondary hypertension, we observed similar results for tertiles 3 and 2 versus 1: multivariable model 1, hazard ratio (95% confidence interval for all-cause mortality 0.67 (0.50–0.91), P=0.009; cardiovascular mortality 0.60 (0.42–0.85), P=0.004; and stroke mortality 0.38 (0.19–0.74), P=0.004; multivariable model 2, all-cause mortality 0.68 (0.52–0.90), P=0.007; cardiovascular mortality 0.64 (0.46–0.90), P=0.009; and stroke mortality 0.54 (0.29–1.01), P=0.055. The exclusion of patients who died in the first 2 years of follow-up gave similar results: multivariable model 1, all-cause mortality 0.69 (0.52–0.92), P=0.012; cardiovascular mortality 0.68 (0.48–0.97), P=0.036; and stroke mortality 0.45 (0.23–0.89), P=0.021; multivariable model 2, all-cause mortality 0.70 (0.53–0.92), P=0.010; cardiovascular mortality 0.73 (0.52–1.02), P=0.064; and stroke mortality 0.67 (0.35–1.28), P=0.229.
Prognostic Value of Protein Intake According to Various Conditions
To test the interaction of various clinical conditions, tertiles 3 and 2 were grouped because of their similar survival rates, which led to a comparison between 2 groups: <0.7 versus ≥0.7 g/kg. Table 3 shows the prognostic value of tertiles 3 and 2 versus 1 according to various conditions. Overall, the protective effect of a PInt/IBW >0.7 g/kg was observed, particularly for all-cause death in young patients and in the absence of previous cardiovascular disease or aortic atherosclerosis. For cardiovascular mortality, the protective effect of a PInt/IBW >0.7 g/kg was more marked in absence of previous cardiovascular disease and to a lesser extent in young patients and in absence of aortic atherosclerosis (albeit the P value for interaction was >0.1). For stroke mortality, we observed a better prognostic value for tertiles 2 and 3 in the absence of aortic atherosclerosis and in patients with a low salt intake, albeit the P value for interaction was not statistically significant. We noted a marked interaction for hypertensive retinopathy and stroke.
Based on a large cohort with a long follow-up, our results demonstrate for the first time that hypertensive patients with a normal–high protein intake had decreased all-cause and cardiovascular mortality rates and, to a lesser extent, stroke deaths. More specifically, a PInt/IBW >0.7 g/kg/d was associated with a good prognosis in the absence of overt cardiovascular disease (heart failure, aortic atherosclerosis, or coronary artery disease).
Effect of Protein Intake on BP and Outcomes
Our results clearly indicate a risk reduction for all-cause mortality and cardiovascular mortality with a normal–high protein intake because this was observed in 2 multivariable models adjusting for a series of metabolic or risk factors. This prognostic effect was consistent in several sensitivity analyses, confirming the robustness of the finding.
Evidence concerning the cardiovascular effect of protein intake is scant and controversial, especially in hypertensive patients without chronic kidney disease. Concerning the effect on BP, the INTERSALT study showed a significant inverse association between urinary urea nitrogen (estimated from 24-hour urinary urea) and BP.18 On the other hand, PREVEND study,6 among others,19,20 did not report such an association. A recent meta-analysis of 40 randomized controlled trials demonstrated that dietary protein intake (median protein supplementation of 40 g/d) was associated with a significant decrease in mean systolic BP and diastolic BP of ≈2 mm Hg.5 There was no heterogeneity in BP reduction based on animal or vegetable protein sources. In our study, none of the BP variables appeared correlated with 24-hour urinary urea, emphasizing the fact that the BP effect of protein intake was limited, if any.
With respect to the prognostic value of protein intake in relation to cardiovascular events, particularly stroke, the results also seem inconsistent. Data were obtained mainly in the general population. Some studies indicated a reduction in risk of stroke with high protein intake8,21–23 while others did not.24,25 The main limitation of these studies was the assessment of protein intake based on food frequency questionnaire, 24-hour dietary recall, or urinary urea concentration on spot urine. This aspect is a particular strength of our study because 24-hour urinary urea excretion performed in a hospital ward was used to quantify protein consumption and currently represents the gold standard. Albeit statistically significant, the correlation with the food frequency questionnaire and urinary urea concentration on spot urine was extremely weak in our study, emphasizing the limitations of these approaches, as already known.
Similarly, the prognostic value of protein intake for all-cause mortality is also controversial.26 In addition to the benefit on cardiovascular outcome, protein intake may also affect cancer deaths, particularly in patients >65 years.27 Younger age, no previous cardiovascular event (stroke, heart failure, coronary artery disease, or peripheral artery disease), and absence of hypertensive retinopathy or aortic remodeling amplified the protective effect of protein intake. This is an important point for the practical effect of our results because it confirms the current recommendations of a >0.8 g/kg protein consumption in hypertensive patients, provided that they have normal cardiac and vascular function.
Putative Mechanisms for the Protective Effect of Protein Intake
Studies in healthy subjects demonstrate that an increase in dietary protein from 15% to 30% of energy at a constant carbohydrate intake produces a sustained decrease in appetite and spontaneous caloric intake by increasing central nervous system leptin sensitivity and results in weight loss.28,29 Moreover, fat mass decreased significantly in healthy subjects. On the contrary, data in renal failure patients suggest that a reduced protein intake is associated with a decreased total caloric intake.30 Thus, the implication of changes in the total caloric intake on outcomes cannot be ruled out, albeit difficult to address because of the fact that this parameter was not monitored in our study. Yet, these metabolic effects probably do not account entirely for the better outcome associated with high protein diet.
The profile of patients in whom a high protein intake had the greatest favorable influence (ie, younger patients, without detectable aortic atherosclerosis or previous cardiovascular disease) is consistent with a rather preserved endothelial function. Mainly via nitric oxide production, a functional endothelium has anti-inflammatory, antithrombotic, vasorelaxant, and antihypertrophic properties. Arginine and tryptophan, 2 amino acids contained in most proteins, are involved in nitric oxide production; in addition, tryptophan decreases plasma concentration of epinephrine and norepinephrine.31,32 These mechanisms, present when endothelial function is preserved (Figure 2), may be in the case of endothelial dysfunction, explaining why the protective effect of protein intake is no longer observed with aging, hypertensive retinopathy, and overt cardiovascular disease. High salt intake may also play a role by reducing vascular nitric oxide bioavailability.33 In clinical practice, one can suppose that high protein intake coming from animal sources will be associated also with an increased salt consumption, leading to less positive nutritional consequence in hypertensive patients.
Other noncardiovascular benefits are less clear. Studies demonstrating an increased risk of overall mortality and cancer death in patients with a high protein intake made the hypothesis that protein restriction decreases growth hormone receptor/insulin-like growth factor-1 activity. This metabolic pathway displays a major role in age-related diseases, such as cancer, and in overall mortality.27 But, once again, protein intake assessment was based on reports of food and beverage intake and not indexed on BMI.
The main limitations of our study are its observational design, the absence of data regarding the frequency of visits, changes in antihypertensive medication, and achievement of BP control during follow-up. Moreover, interaction between renal death and protein intake could not specifically be explored in our cohort because of a low number of outcomes after 10 years of follow-up (n=25). Another limitation is the absence of urinary creatinine measurement to control for a complete 24-hour urinary collection. However, as mentioned, urinary collection was done during the in-hospital stay using a standardized protocol. The patients’ characteristics differ from today’s hypertensive patients, with a higher BP level, a high prevalence of severe retinopathy, and different kinds of antihypertensive treatment. However, there is no indication that the current disease differs from that in the 1970s in terms of its pathophysiology and that an interaction exists between the kind of treatment and protein consumption. Finally, we did not have information on the origin of protein intake and whether it came from animal or vegetable sources.
In our historic cohort, we demonstrated that a PInt/IBW >0.7 g/kg per day had a protective effect for all-cause, cardiovascular, and stroke mortality in hypertensive patients. This may be because some amino acids (arginine and tryptophan) may influence the bioavailability of nitric oxide. This is supported by the fact that the beneficial effect of protein intake apparently needs a preserved cardiovascular and probably endothelial function to be observed (such as in young hypertensive patients without overt cardiovascular disease). A major practical implication of these data is to encourage hypertensive patients without chronic kidney disease to have a normal (0.8–1.0 g/kg) or high (>1.0 g/kg) protein diet in addition to low salt consumption, moderate alcohol consumption, and regular physical activity, particularly those with low cardiovascular risk.
We thank Sophie Rushton-Smith, PhD, for thorough editing of this article.
The online-only Data Supplement is available with this article at http://hyper.ahajournals.org/lookup/suppl/doi:10.1161/HYPERTENSIONAHA.116.07409/-/DC1.
- Received February 23, 2016.
- Revision received March 2, 2016.
- Accepted March 12, 2016.
- © 2016 American Heart Association, Inc.
- 1.↵ESH ESC Task Force for the Management of Arterial Hypertension. 2013 Practice guidelines for the management of arterial hypertension of the European Society of Hypertension (ESH) and the European Society of Cardiology (ESC): ESH/ESC Task Force for the Management of Arterial Hypertension. J Hypertens. 2013;31:1925–1938.
- Fouque D,
- Laville M
- Rebholz CM,
- Friedman EE,
- Powers LJ,
- Arroyave WD,
- He J,
- Kelly TN
- Tielemans SM,
- Geleijnse JM,
- van Baak MA,
- Engberink MF,
- Brink EJ,
- de Jong PE,
- Gansevoort RT,
- Bakker SJ
- Buendia JR,
- Bradlee ML,
- Singer MR,
- Moore LL
- Rand WM,
- Pellett PL,
- Young VR
- Harbaoui B,
- Courand PY,
- Milon H,
- Fauvel JP,
- Khettab F,
- Mechtouff L,
- Cassar E,
- Girerd N,
- Lantelme P
- Pavillon G,
- Laurent F
- 17.↵Rapport AFSSA Apports en protéines: consommation, qualité, besoins et recommandations. 2007;2015:http://www.afssa.fr/Documents/NUT-Ra-Proteines.pdf.
- Stamler J,
- Elliott P,
- Kesteloot H,
- Nichols R,
- Claeys G,
- Dyer AR,
- Stamler R
- Iso H,
- Sato S,
- Kitamura A,
- Naito Y,
- Shimamoto T,
- Komachi Y
- Iso H,
- Stampfer MJ,
- Manson JE,
- Rexrode K,
- Hu F,
- Hennekens CH,
- Colditz GA,
- Speizer FE,
- Willett WC
- Preis SR,
- Stampfer MJ,
- Spiegelman D,
- Willett WC,
- Rimm EB
- Sauvaget C,
- Nagano J,
- Hayashi M,
- Yamada M
- Pedersen AN,
- Kondrup J,
- Børsheim E
- Weigle DS,
- Breen PA,
- Matthys CC,
- Callahan HS,
- Meeuws KE,
- Burden VR,
- Purnell JQ
- Lejeune MP,
- Westerterp KR,
- Adam TC,
- Luscombe-Marsh ND,
- Westerterp-Plantenga MS
- Watts SW,
- Morrison SF,
- Davis RP,
- Barman SM
Novelty and Significance
What Is New?
This is the first cohort study of hypertensive patients establishing the prognostic value of protein intake using the current gold standard, 24-hour urinary urea, concerning all-cause, cardiovascular, and stroke mortality.
What Is Relevant?
After adjustment for major cardiovascular risk factors, hypertensive patients with protein intake >0.7 g/kg per day of ideal body weight had a decreased risk of all-cause and cardiovascular death, but not of stroke death.
Normal–high protein intake had a more marked positive influence in a subset of our population: younger patients, low salt intake, absence of aortic atherosclerosis, or previous cardiovascular events.
Our data demonstrate that a protein intake >0.7 g/kg per day of ideal body weight has a favorable effect, particularly in low-risk hypertensive patients. Physicians may encourage hypertensive patients to have normal or high protein diet in addition to the usual lifestyle changes (low salt consumption, moderate alcohol consumption, and regular physical activity).