Journal of Human Hypertension (2014), 1–6

& 2014 Macmillan Publishers Limited All rights reserved 0950-9240/14

www.nature.com/jhh

ORIGINAL ARTICLE
Effect of a high-protein diet on maintenance of blood
pressure levels achieved after initial weight loss: the DiOGenes
randomized study
MF Engberink1,2, JM Geleijnse1,2, SJL Bakker1,3, TM Larsen4, T Handjieva-Darlesnka5, A Kafatos6, JA Martinez7, AFH Pfeiffer8,
M Kunešová9, SA Jebb10, C Holst11, A Astrup12, WHM Saris13, EJ Brink1 and MA van Baak1,13

Randomized trials have shown significant blood pressure (BP) reductions after increased protein compared with carbohydrate
intake, but the effect on BP maintenance after initial weight loss is unclear. We examined the effect of a high-protein diet on the
maintenance of reduced BP after weight loss in 420 overweight adults from the Diet, Obesity and Genes study. After an 8-week
weight-loss period (48% BW), subjects (42±6 years) were randomized to either a high-protein diet (23–28 en% protein) or a lower-
protein control diet (10–15 en% protein) for 26 weeks. BMI after weight loss was 30.3±4.3 kg m  2, BP was 118/73 mm Hg and 28
subjects (6.5%) used antihypertensive agents. Systolic BP during 26 weeks of weight maintenance dietary intervention increased in
both treatment groups, but it was 2.2 mm Hg less (95% CI:  4.6 to 0.2 mm Hg, P ¼ 0.08) in the high-protein group than in the
lower-protein control group. In 191 (pre)hypertensive subjects (baseline systolic BPX120 mm Hg), a larger difference was observed
(  4.2 mm Hg (  7.7,  0.7), P ¼ 0.02). The effect was attenuated after adjustment for initial BP (  3.4 mm Hg (  6.9,  0.03),
P ¼ 0.048), and after additional adjustment for weight change (  2.7 mm Hg (  6.1, 0.4), P ¼ 0.11). Adjustment for 24-h urinary
excretion of sodium and potassium did not change the results. Diastolic BP yielded similar results. These findings suggest that a BP
reduction after weight loss is better maintained when the intake of protein is increased at the expense of carbohydrates. This effect
is partly mediated by body weight.

Journal of Human Hypertension advance online publication, 24 April 2014; doi:10.1038/jhh.2014.30

INTRODUCTION for (imbalances in) baseline BP, other potential confounding

Hypertension is the leading risk factor worldwide for total
and cardiovascular mortality.1 Diet and lifestyle have an
important role in blood pressure (BP) control. Established dietary
measures to lower BP are weight reduction in overweight
and obese individuals, reduced sodium intake, moderation of
alcohol intake (among those who drink) and an increased
potassium intake.2,3 More recently, interest has grown in diet
composition and macronutrient intake, including protein. In the
OmniHeart study, a randomized cross-over trial among 164 US
adults with untreated (pre)hypertension, a 6-week protein-rich
diet reduced BP by 1.6/1.4 mm Hg when compared with a
carbohydrate-rich diet.4 When compared with a diet rich in
monounsaturated fat, however, no effect of dietary protein on BP
was observed.4 Subsequent randomized controlled trials that
investigated the BP-lowering effect of diets with a higher protein
content at the expense of carbohydrates showed inconsistent
results.5–9 These trials primarily focussed on weight loss8,9 or
weight maintenance after initial weight loss.5–7 In these trials,
sample size was relatively small, and no adjustments were made
factors or change in body weight, which may partly explain the
variable findings.
In the pan-European Diet, Obesity and Genes (DiOGenes) study,
the effect of diets differing in protein content and glycaemic
index on body weight was studied in 773 overweight adults aged
o65 years.10,11 Within the DiOGenes study, Gogebakan et al.12
explored whether weight-loss-induced improvements in
cardiovascular risk factors were subsequently affected by ad
libitum diets differing in protein content and glycaemic index.
They showed a beneficial effect of a low-glycaemic index diet and/
or low-protein diet on high-sensitive C-reactive protein, whereas
lipid profiles and BP were not differentially affected. Initial BP
levels and weight change, however, were not taken into account
in their analysis, and no adjustments were made for salt intake.
We therefore examined the effect of the high-protein diet on
maintenance of reduced BP after weight loss in the DiOGenes
study, with stratification for (pre)hypertensive status and
adjustment for potential confounders such as body weight
changes.

1
Top Institute Food and Nutrition, Wageningen, The Netherlands; 2Division of Human Nutrition, Wageningen University, Wageningen, The Netherlands; 3Lifestyle Medicine
Program, University of Groningen, University Medical Center Groningen, Groningen, The Netherlands; 4Department of Nutrition, Exercise and Sports (NEXS), Faculty of Science,
University of Copenhagen, Copenhagen, Denmark; 5Department of Pharmacology and Toxicology, Medical Faculty, National Transport Hospital, Sofia, Bulgaria; 6Department of
Social Medicine, Preventive Medicine and Nutrition Clinic, University of Crete, Crete, Greece; 7Department of Physiology and Nutrition, University of Navarra, Pamplona, Spain;
8
Department of Clinical Nutrition, German Institute of Human Nutrition Potsdam-Rehbrücke, Nuthetal, Germany; 9Obesity Management Center, Institute of Endocrinology,
Prague, Czech Republic; 10Medical Research Council Human Nutrition Research, Elsie Widdowson Laboratory, Cambridge, UK; 11Institute of Preventive Medicine, Centre for Health
and Society, Copenhagen, Denmark; 12Department of Human Nutrition, Faculty of Life Sciences, University of Copenhagen, Copenhagen, Denmark and 13Department of Human
Biology, NUTRIM School for Nutrition, Toxicology and Metabolism, Faculty of Health, Medicine and Life Sciences, Maastricht University Medical Centre, Maastricht, The
Netherlands. Correspondence: Professor MA van Baak, Department of Human Biology, NUTRIM School for Nutrition, Toxicology and Metabolism, Faculty of Health, Medicine and
Life Sciences, Maastricht University Medical Centre, PO Box 616, Maastricht 6200 MD, The Netherlands.
E-mail: m.vanbaak@maastrichtuniversity.nl
Received 9 October 2013; revised 17 March 2014; accepted 21 March 2014
 Dietary protein and maintaining blood pressure
MF Engberink et al
2
SUBJECTS AND METHODS
The DiOGenes study design
The present analysis used data from the pan-European DiOGenes study, a
randomized controlled trial on the effects of diets varying in protein
content and glycaemic index on weight-loss maintenance conducted
between November 2005 and April 2007 (http://www.diogenes-eu.org).
The DiOGenes study was approved by the local ethics committees of each
European centre, and has been described in more detail elsewhere.10 In
brief, the design of the multicentre DiOGenes study imposed an initial low
calorie diet-induced weight loss of at least 8% of body weight in
overweight and obese subjects, followed by a 26-week dietary intervention
period serving as a weight maintenance phase. During this weight
maintenance phase, four different ad libitum diets were compared with
either high or low glycaemic index and/or protein content for the
maintenance of body weight.11 BP was also monitored during this
intervention period.
Study population
The population of the present analysis included overweight and obese
adults from eight European countries (that is, The Netherlands, Denmark,
United Kingdom, Greece, Spain, Germany, Bulgaria and the Czech
Republic) who successfully lost weight. Mean reduction in body weight
during 8 weeks was 11.0±3.3 kg, which was accompanied by a BP
decrease of 7.8±11.1 mm Hg systolic and 5.0±8.0 mm Hg diastolic (all
Po0.001). In the present analysis, we examined the effect of dietary
protein on maintaining a reduced BP during 26 weeks of follow-up.
Changes in glycaemic index were not taken into account because there
was no significant interaction with protein intake in relation to BP
(P interaction ¼ 0.61). Of the 773 adults who were randomized, 619
subjects were assigned to either a high-protein diet or a diet with a lower
protein content (that is, lower-protein control diet). The original DiOGenes
study also included a group of 154 subjects who consumed a regular
‘background diet’, but this group was not included in the present analysis.
In addition, 185 subjects (30%) dropped out during the 26-week dietary
intervention period, and 13 subjects were excluded because of missing BP
data, leaving 420 subjects for the present analysis (Figure 1).
Dietary intervention
Subjects were advised to maintain their achieved weight during the
intervention. All subjects completed a 3-day dietary record at weeks 4 and
26 of the dietary intervention period. During the intervention period,
subjects received careful and intensive dietary and behavioural guidance
773 adults succeeded the criteria for successful
weight loss and were randomized for the
maintenance intervention
(represents baseline)
305 assigned to
Lower-protein control diet
102 dropped out (33.4 %)
8 with missing BP data
195 adults included in the
primary analysis
Figure 1. Flow diagram of subjects from the DiOGenes randomized study included in the present analyses on the effect of dietary protein
content on maintaining achieved blood pressure level. BP denotes blood pressure.
Journal of Human Hypertension (2014) 1 – 6
on macronutrient intake (and glycaemic index) every two weeks up to
week 6, and once a month thereafter. In addition to dietary guidance,
participants from The Netherlands and Denmark could choose free foods
with prespecified protein content from a shop, resulting in a more
controlled intervention (that is, ‘trial shop’ intervention).10
The target protein content was 23–28 energy percent (en%) for the
high-protein diet and 10–15 en% for the lower-protein control group. The
study was ad libitum for total energy intake, but was carefully controlled for
macronutrient composition based on a points system.13
Body weight and BP
Height was measured using a stadiometer to the nearest 0.5 cm. Body
weight was measured at randomization, after 4 weeks of intervention and
after 26 weeks of intervention using a calibrated digital balance to the
nearest 0.1 kg. BMI was calculated by dividing the subject’s weight (in kg)
by the square of height (in m). Systolic and diastolic BP were measured
three times at randomization and after 26 weeks of intervention by the
same trained research staff in the morning after an overnight fast with an
automatic device after at least 5 min while resting in a supine position
according to WHO criteria. The mean value of the last two measurements
was recorded. All measurements were performed according to the same
standardized operating procedures in all participating centres. Hyperten-
sion was defined as systolic BPX140 mm Hg or diastolic BPX90 mm Hg or
use of antihypertensive agents.
Other measurements
Information on current health status, medical history, medication use and
lifestyle factors was obtained by questionnaires. Participants were
classified as current smokers, former smokers or never smokers. Alcohol
intake was assessed in grams of ethanol per day. Physical activity was
assessed with the Baecke questionnaire14 consisting of 16 items from
which three indexes were calculated: work index referring to physical
activity at work, sport index referring to sports participation during leisure
time and leisure-time index referring to physical activity during leisure time
excluding sport activities.
A 24-h urine collection was performed after 4, 14 and 26 weeks of
intervention. Completeness of urinary collection was checked by recovery
rate of para amino benzoic acid, which was determined by spectro-
photometry (Stasar, Gilford Instruments Laboratories, Oberlin, OH, USA).
Urinary nitrogen to assess adherence to the diet was determined by
Dumas combustion methodology using a VarioMax CN analyzer (Elemen-
tar, Hanau, Germany). In addition, 24-h urine samples were used for
analysis of sodium and potassium (Roche Modular, Roche, Germany).
154 assigned to
control diet of original
DiOGenes study and not
included in the present
analysis
314 assigned to
high protein diet
83 dropped out (26.4 %)
6 with missing BP data
225 adults included in the
primary analysis
& 2014 Macmillan Publishers Limited
 Dietary protein and maintaining blood pressure
MF Engberink et al
3
Statistical analysis
Table 1. Characteristics of 420 subjects from the DiOGenes study in
All analyses were performed with the SPSS software (SPSS Inc., Chicago, IL,
the high-protein diet group and the lower-protein control group after
USA). Two-sided P-values o0.05 were considered statistically significant.
initial weight loss
Values in text and tables are reported as means±s.d. or as means (95% CI),
unless stated otherwise. Diet composition was tested with an independent
sample t-test for between-group differences. Response to treatment was Lower-protein High-protein
defined as change in BP from randomization. BP values were not available for control diet diet (n ¼ 225)
subjects who dropped out during the intervention period (that is, 185 (n ¼ 195)
subjects (30%)). Between-group differences were analysed using analysis of
Age, years 42.1±5.8 42.0±6.1
variance based on the completers-only population. The analyses were
Sex, n male/female 63/132 84/141
repeated with adjustment for baseline BP. To examine a potential
intermediary role for body weight, we additionally adjusted our analyses
Study centre, n (%)
for change in body weight. In addition, we examined a potential intermediary
Denmark 38 (18.5) 44 (19.6)
role for excretion of sodium and potassium by including 24-h urinary
The Netherlands 36 (19.5) 40 (17.8)
excretion during intervention (mean of week 4, 14 and 26) as covariates.
UK 15 (7.7) 26 (11.6)
In an additional analysis, we classified subjects in two strata on the
Germany 25 (12.8) 26 (11.6)
basis of their systolic BP level (treated or untreated) X120 mm Hg or
Spain 20 (10.2) 23 (10.2)
o120 mm Hg. The analyses were also repeated in the subgroup of
Greece 10 (5.1) 19 (8.4)
participants from Denmark and The Netherlands who had followed the
Bulgaria 23 (11.7) 25 (11.1)
‘trial shop’ intervention that was more controlled, which resulted in larger
Czech Republic 28 (14.3) 22 (9.8)
contrasts in protein intake.
Current smoking, n (%) 47 (26) 50 (23.4)
Alcohol drinking, n (%) 130 (66.7) 158 (70.2)
RESULTS Intake among users, 6.7 (3.9;15.8) 9.5 (4.5;17.8)
Baseline characteristics g per daya
Participants’ characteristics at the start of the intervention period
(that is, baseline) according to diet are shown in Table 1. Mean age Physical activity, Baecke activity scoreb
Work 2.7±0.4 2.7±0.4
of the study population was 42±6 years and mean BMI was
Sport 2.7±0.4 2.6±0.4
30.3±4.3 kg m  2. Mean BP was 118.0±13.6 mm Hg systolic and Leisure 3.0±0.6 3.0±0.6
72.9±9.6 mm Hg diastolic (including subjects using antihyperten-
sive agents), and 13.6% of the participants were hypertensive. Body mass index (BMI)
There were no significant differences between groups. There were Mean BMI, kg m  2 b 30.2±4.2 30.5±4.4
also no significant differences with subjects who dropped out BMIo25.0, n (%) 14 (7.2) 14 (6.2)
during intervention (that is, mean BP 118.2/72.0 mm Hg, mean age BMI 25.0–29.9, n (%) 92 (46.9) 111 (49.3)
40±6 years and mean BMI was 31.3±4.4 kg m  2). BMIX30.0, n (%) 89 (45.4) 100 (44.4)

Systolic BP, mm Hg 117.0±13.5 119.0±13.6

Protein intake Diastolic BP, mm Hg 72.8±9.6 73.1±9.7
Self-reported mean dietary intakes during intervention are presented
in Table 2. Mean (±s.d.) dietary protein intake was 3.2±0.5 en% BP category, n (%)c
(15.7±2.6 g per day) higher in the high-protein diet group compared Optimal 97 (49.7) 112 (49.8)
with the lower-protein control group (Po0.001), which originated Normal 42 (21.5) 50 (22.2)
mainly from animal sources. This difference in self-reported protein High normal 34 (17.4) 28 (12.4)
Hypertension 22 (11.3) 35 (15.6)
intake was confirmed by a 2.1±0.5 g higher 24-h urinary excretion of
Use of antihypertensive 12 (6.2) 16 (7.1)
nitrogen in the high-protein diet group (Po0.001). The proportion of agents, n (%)
total energy consumed from carbohydrates was 5.9±0.8% lower in
the high-protein diet group (Po0.001), whereas glycaemic index was Abbreviation: BP, blood pressure. Data are presented as mean±s.d. or as
similar for the two diets (P ¼ 0.30). n (%). aPresented as median with interquartile range because of skewed
In participants with a BPX120 mm Hg (n ¼ 191), self-reported distribution. bData available for 142 subjects (low protein) and 180 subjects
(high protein), respectively. cDefined as follows: optimal, systolic BPo120
protein intake was 2.6±0.7 en% (13.7±3.7 g per day) higher in the
mm Hg and diastolic BPo80 mm Hg; normal, 120–129 mm Hg for systolic
high-protein diet group compared with the lower-protein control BP or 80–84 for diastolic BP; high normal, 130–139 mm Hg for systolic BP or
group (Po0.001). The difference in 24-h urinary nitrogen excretion 85–89 mm Hg for diastolic BP; hypertension, systolic BPX140 mm Hg or
was 1.9±0.7 g per day (Po0.001). Detailed data on self-reported diastolic BPX90 mm Hg or use of antihypertensive agents.
dietary intake and urinary markers of this subgroup are tabulated
in the online supplement to this paper (Supplementary Table 1).
In participants from Denmark and The Netherlands (n ¼ 158) and  0.9 (  2.5, 0.6) mm Hg for diastolic BP (P ¼ 0.24). After
who had followed the ‘trial shop’ intervention that was more adjustment for baseline BP, the treatment effect for systolic BP
controlled, we observed a larger contrast in protein intake. Self- was attenuated to  1.3 (  3.8, 0.8) mm Hg (P ¼ 0.22).
reported protein intake was 6.1±0.6 en% (23.1±3.7 g per day) Mean change in body weight was 0.02±0.4 kg in the high-
higher in the high-protein group, which was confirmed by a protein diet group versus 1.0±0.4 kg in the lower-protein control
2.9±0.6 g higher 24-h urinary nitrogen excretion (all Po0.001). group (P ¼ 0.07). The difference in systolic BP response (adjusted
Detailed data on self-reported dietary intake and urinary markers for initial systolic BP) independent of weight change was  0.9
of this subgroup are tabulated in the online supplement to this (  3.1, 1.2) mm Hg. Mean 24-h urinary excretion of sodium during
paper (Supplementary Table 2). intervention did not significantly differ between the high-protein
diet and lower-protein control diet (182.3 versus 173.7 mmol per
Effects of protein intake on BP 24 h, respectively, P ¼ 0.30, Table 2), whereas potassium excretion
BP increased in both groups during the weight maintenance was significantly higher in the high-protein group (82.3 versus
dietary intervention, but less in the high-protein diet group than in 76.5 mmol per 24 h, respectively, P ¼ 0.03). Adjusting our analysis
the lower-protein control group (Table 3). The treatment effect for sodium and potassium excretion did not change the results
was  2.2 (95% CI:  4.6, 0.2) mm Hg for systolic BP (P ¼ 0.08) (that is, BP effect:  1.1 (  3.3, 1.1) mm Hg systolic).

& 2014 Macmillan Publishers Limited Journal of Human Hypertension (2014) 1 – 6

 Dietary protein and maintaining blood pressure
MF Engberink et al
4
Table 2. Dietary intake and urinary markers during intervention in Table 4. Blood pressure during intervention in subjects from the
420 subjects of the DiOGenes study, according to high-protein and DiOGenes study, stratified by initial blood pressure levela
lower-protein control diet
Lower- High-
Lower- High-protein P-valuea protein protein
protein control diet control diet control diet
control diet
Subgroup with systolic n ¼ 105 n ¼ 124
Dietary intakeb n ¼ 169 n ¼ 203 BPo120 mm Hg
Total energy, kJ per day 6386±2151 6601±2032 0.32 Systolic BP, mm Hg
Fat, en% 28.9±8.1 31.1±6.9 o0.01 Baseline 106.8±0.8 109.3±0.6
Carbohydrates, en% 51.5±9.3 45.6±6.7 o0.001 Week 26 116.0 1.1 117.8±1.1
Glycaemic index, % 58.4±4.9 58.9±5.2 0.30 Change from baseline 9.2±1.0 8.5±1.1
Protein, en% 18.2±4.5 21.5±4.3 o0.001 Treatment effect, adjustedb 0.4 (  2.5, 3.2), P ¼ 0.79
Protein, g per day 65.2±23.5 80.8±25.3 o0.001
Animal protein, 38.7±21.0 54.7±20.5 o0.001 Diastolic BP, mm Hg
g per day Baseline 67.2±0.7 68.1±0.7
Plant protein, g per 26.5±10.1 26.1±8.8 0.68 Week 26 70.7±0.9 71.3±0.8
day Change from baseline 3.5±0.8 3.1±0.7
Fiber, g per day 21.2±9.7 19.2±7.6 0.03 Treatment effect, adjustedb  0.05 (  2.1, 2.0), P ¼ 0.96

Urinary markersc n ¼ 180 n ¼ 215 Subgroup with systolic n ¼ 90 n ¼ 101

Nitrogen, g per 24 h 12.5±4.3 14.6±4.8 o0.001 BPX120 mm Hg
Sodium, mmol per 24 h 173.6±76.1 182.3±88.9 0.30 Systolic BP, mm Hg
Potassium, mmol per 76.2±28.7 82.3±29.2 0.03 Baseline 128.8±0.8 130.9±1.0
24 h Week 26 130.3±1.3 128.1±1.3
Calcium, mmol per 24 h 4.6±2.4 4.4±2.1 0.48 Change from baseline 1.5±1.3  2.7±1.3
Magnesium, mmol 4.2±1.6 4.1±1.5 0.72 Treatment effect, adjustedb  3.4 (  6.9, 0.03),
per 24 h P ¼ 0.048

Values are presented as mean±s.d. aDifferences between groups were Diastolic BP, mm Hg
tested with an independent Student’s t-test (two-sided P-value). bBased on Baseline 79.2±0.8 79.1±0.8
two 3-day dietary records during the intervention period, that is, week 4 and Week 26 80.3±0.9 78.5±1.0
week 26. If information on one occasion was missing, data are based on one Change from baseline 1.1±0.9  0.7±0.8
measurement. cBased on three 24-h urine collections during the interven- Treatment effect, adjustedb  1.8 (  4.0, 0.9), P ¼ 0.11
tion period, that is, week 4, week 14 and week 26. If information on one or
two occasions was missing, data are based on the other measurement(s). Abbreviation: BP, blood pressure. Values are mean±s.d.; treatment effects
are presented as mean (95% CI). Differences between groups were tested
with analysis of covariance (two-sided P-values). aBoth groups (systolic
BPo120 mm Hg and systolic BPX120 mm Hg) included participants using
antihypertensive medication. bAdjusted for baseline systolic or diastolic BP,
Table 3.
Blood pressure during intervention in 420 subjects from the respectively.
DiOGenes study, according to high-protein and lower-protein control
diet
to  3.4 (  6.9,  0.03) mm Hg (P ¼ 0.048). Further adjustment for
Lower-protein High-protein
control diet control diet change in body weight (that is, 0.7±5.1 kg for the lower-protein
(n ¼ 195) (n ¼ 225) group and  1.1±7.1 kg for the high-protein group, P ¼ 0.045)
resulted in a BP response of  2.7 (  6.1, 0.6) mm Hg (P ¼ 0.11),
Systolic BP, mm Hg whereas adjustment for sodium and potassium excretion (in
Baseline 117.0±0.9 119.0±1.0 addition to baseline BP) resulted in a BP response of  3.1 (  6.5,
Week 26 122.6±1.0 122.4±0.9 0.4) mm Hg systolic (P ¼ 0.08). We observed no significant effect of
Change from baseline 5.7±0.8 3.4±0.9 protein intake on BP in subjects with systolic BPo120 mm Hg
Treatment effect, crude  2.2 (  4.6, 0.2), P ¼ 0.08 (Table 4). Diastolic BP showed similar results (Table 4).
Treatment effect, adjusteda  1.3 (  3.5, 0.8), P ¼ 0.22
In participants from Denmark and The Netherlands (n ¼ 158),
Diastolic BP, mm Hg the treatment effect was  3.3 (  7.0, 0.4) mm Hg for systolic BP
Baseline 72.8±0.7 73.1±0.6 (P ¼ 0.08). After adjustment for initial BP, the treatment effect was
Week 26 75.1±0.7 74.5±0.7 attenuated to  2.5 (  6.0, 0.9) mm Hg (P ¼ 0.15). Adjustment for
Change from baseline 2.4±0.6 1.4±0.5 both systolic BP and weight change (that is, 2.5±3.7 kg for the
Treatment effect, crude  0.9 (  2.5, 0.6), P ¼ 0.24 lower-protein group and 0.7±5.2 kg for the high-protein group,
Treatment effect, adjusteda  0.9 (  2.3, 0.6), P ¼ 0.26 Po0.05) yielded a systolic BP effect of  1.6 mm Hg (P ¼ 0.40),
Abbreviation: BP, blood pressure. Values are mean±s.d.; treatment effects whereas adjustment for sodium and potassium excretion did not
are presented as mean (95% CI). Differences between groups were tested change the results, that is, treatment effect  2.5 (  6.0, 1.0)
with AN(C)OVA (2-sided P-values). aAdjusted for baseline systolic or mm Hg systolic. Diastolic BP showed similar results (data not
diastolic BP, respectively. shown).

Subgroup analysis
In subjects with a systolic BPX120 mm Hg (n ¼ 191), the treatment
effect was  4.2 (  7.7,  0.7) mm Hg systolic (P ¼ 0.02, Table 4).
After adjustment for initial BP, the treatment effect was attenuated
DISCUSSION
In the present study, based on data from the DiOGenes study, we
examined the effect of protein intake on maintaining a reduced
BP after 8 weeks of energy restriction. During the 26 weeks of
intervention (aiming at maintaining body weight), BP increased in

Journal of Human Hypertension (2014) 1 – 6 & 2014 Macmillan Publishers Limited


most participants, but this increase was 2.2 mm Hg less in
(pre)hypertensive participants who increased protein intake to
B22 en% at the expense of carbohydrates. The effect of protein
on BP was attenuated and was no longer statistically significant
after adjustment for weight change, suggesting that the effect
was partly mediated by body weight.
Strengths of the DiOGenes study include its prospective large-
scale, randomized controlled design. BP was measured according
to strict and standardized protocols in all participating centers.
Compliance to the protein manipulation was confirmed by urinary
excretion of nitrogen. Body weight, an important BP determinant,
was carefully monitored because it was the primary outcome of
the DiOGenes study. We were able to examine the effect of
protein intake on the BP response independent of changes in
body weight, in contrast to other studies of protein intake and BP
after weight loss.6,11 Moreover, we examined a potential
intermediary role for salt intake and performed stratification by
(pre)hypertensive status.
The present study also had limitations. First, because of missing
BP values, we excluded subjects who dropped out during the
intervention (30%). However, baseline characteristics were similar
to those subjects included in the analysis, making it unlikely that
this non-differential drop-out has influenced our findings. Second,
the difference in protein intake was lower (B3.3 en%) than
targeted (12.0 en%11), which may partly explain why the observed
overall BP response was relatively small and non-significant. We
observed larger BP effects in subgroups with a larger contrast in
protein intake, that is,  3.3 mm Hg systolic in participants who
followed the ‘trial shop’ intervention (protein difference B6 en%).
Third, initial BP (B118 mm Hg systolic) may have been too low to
find a significant effect. The effect was larger and statistically
significant in participants with BPX120 mm Hg. Finally,
participants in the high-protein group increased their intake of
protein-rich foods and reduced their intake of carbohydrate-rich
foods. Because dietary intakes were not fully controlled, other
nutrients that could influence BP may also have changed,
including fiber, fatty acids and polyphenols. We adjusted our
analyses for sodium and potassium excretion during intervention,
but we cannot exclude residual confounding by unmeasured
nutrients (for example, polyphenols).
Previous BP trials in which protein intake was increased (at
the expense of carbohydrates) during weight loss or weight
maintenance showed conflicting results.5–9 Our results in
(pre)hypertensive participants were in line with those from
Delbridge et al.6 who showed that a high-protein diet was more
successful compared with a carbohydrate-rich diet in maintaining
BP in participants with systolic BPX130 mm Hg (that is,
 6.6 mm Hg, Po0.05). Larssen et al.7 reported a 4.3 mm Hg
lower systolic BP for participants on a high-protein diet, which was
borderline significant (P ¼ 0.05). In both studies, despite
randomization, baseline BP was 4–5 mm Hg higher in the
intervention group than in the control group.6,7 This difference
was not accounted for in the analyses and, as a result, BP effects
may have been overestimated. Other weight loss8,9 or weight
maintenance5 studies found no significant effect for BP in favour
of a high-protein diet. In these trials,5,8,9 BP was not the primary
outcome, sample size was relatively small (no20 per intervention
group)5,9 or baseline BP was low (B110 mm Hg systolic),8 which
may (partly) explain the absence of an effect on BP. Finally, the BP
effect in these trials was not adjusted for weight change.
The BP effect of protein without (initial) weight loss has also
been investigated in observational studies and several trials,
suggesting a small beneficial effect of protein on BP.15 In the
Omniheart trial, the modest reduction in BP that was observed
for the high-protein diet compared with the diet high in
carbohydrates was more pronounced in hypertensives
compared with prehypertensives (  3.5 versus  0.9 mm Hg for
systolic BP),4 which is comparable to our findings. Our results are
& 2014 Macmillan Publishers Limited
Dietary protein and maintaining blood pressure
MF Engberink et al
5
also in agreement with findings from two recently published
isoenergetic trials on the effect of protein supplementation on
BP.16,17 In a 4-week randomized, double-blind study in 94 Dutch
adults with elevated BP (mean systolic BPB149 mm Hg), systolic
BP was 4.9 mm Hg (Po0.01) lower after supplementation with 60 g
per day protein (20% pea, 20% soy, 30% egg and 30% milk-protein
isolate) compared with 60 g per day maltodextrin.17 Another
large randomized, double-blind cross-over trial in 352 adults
with prehypertension or stage 1 hypertension (mean systolic
BPB126 mm Hg) that was recently published indicated that 8
weeks of supplementation with 40 g per day soy protein or milk
protein reduced systolic BP (  2.0 mm Hg for soy protein and
 2.3 mm Hg for milk protein) compared with carbohydrates.16
The mechanism through which protein might reduce BP
remains unclear. Several hypotheses have been put forward. First,
protein content of the diet has been suggested to favourably
influence body weight (maintenance).18,19 Body weight is a well-
established BP determinant, that is, 1 kg reduction in body weight
results in a 1 mm Hg lower systolic BP.20 When we adjusted our
analysis for change in body weight, the BP effect was attenuated
(with B0.6 mm Hg) and no longer statistically significant. This
suggests that the effect of protein on BP may (at least partly) be
explained by its effect on body weight. Second, dietary protein
has been proposed to increase renal sodium excretion.21 Cirillo
et al.21 examined urea excretion in an overnight urine sample in
3705 Italians, and found an inverse association with BP only in
subjects with a high sodium excretion. They hypothesized that a
high-protein intake could counteract the sodium-dependent BP
rise via stimulation of renal sodium excretion. In the present study,
protein intake did not affect urinary sodium excretion during
intervention. This makes a potential intermediary role for renal
sodium excretion unlikely, although the sodium content of the
diet was not available to confirm this finding. Finally, it cannot be
excluded that a reduced intake of carbohydrates instead, rather
than a higher intake of protein, is responsible for a beneficial
effect on BP.
In conclusion, the present analyses within the multicentre
DiOGenes study shows that reduced BP after initial weight loss
may be better maintained by increasing the protein content of the
diet at the expense of carbohydrates, especially in (pre)hyperten-
sive subjects. This effect may be partly mediated via body weight.
What is known about this topic
 Dietary prevention strategies are important for maintaining a healthy
blood pressure (BP).
 Weight reduction is, amongst others, an established measure to lower
BP of overweight and obese individuals.
 In addition, diet composition and macronutrient intake may have a
role in BP control.
What this study adds
 The present analysis within the pan-European Diet and Obesity, and
Genes (DiOGenes) study, including 420 overweight adults, showed
that a BP reduction after weight loss is better maintained when intake
of protein is increased at the expense of carbohydrates, especially in
hypertensive individuals.
 This effect is only partly mediated by body weight.
CONFLICT OF INTEREST
The authors declare no conflict of interest.
ACKNOWLEDGEMENTS
The DiOGenes project (http://www.diogenes-eu.org) is funded by a grant from the
European Union Food Quality and Safety Priority of the Sixth Framework Program
Journal of Human Hypertension (2014) 1 – 6
 Dietary protein and maintaining blood pressure
MF Engberink et al
6
(contract no. FP6-2005-513946). DiOGenes is supported by the European Community
(contract no. FOOD-CT-2005-513946). The writing of this report and the determina-
tion of 24-h urinary sodium and potassium excretion was funded by Top Institute (TI)
Food and Nutrition (project number A-1003), Wageningen, The Netherlands. TI Food
and Nutrition is a public/private partnership that generates vision on scientific
breakthroughs in food and nutrition, resulting in the development of innovative
products and technologies (www.tifn.nl). Partners are major food companies and
Dutch research organizations.
REFERENCES
1 Ezzati M, Lopez AD, Rodgers A, Vander Hoorn S, Murray CJ. Comparative Risk
Assessment Collaborating G. Selected major risk factors and global and regional
burden of disease. Lancet 2002; 360: 1347–1360.
2 Appel LJ, Brands MW, Daniels SR, Karanja N, Elmer PJ, Sacks FM. Dietary
approaches to prevent and treat hypertension: a scientific statement from the
American Heart Association. Hypertension 2006; 47: 296–308.
3 Whelton PK, He J, Appel LJ, Cutler JA, Havas S, Kotchen TA et al. National High
Blood Pressure EducationProgram Coordinating Committee. Primary prevention
of hypertension: clinical and public health advisory from The National High Blood
Pressure Education Program. JAMA 2002; 288: 1882–1888.
4 Appel LJ, Sacks FM, Carey VJ, Obarzanek E, Swain JF, Miller 3rd ER et al. OmniHeart
Collaborative Research Group. Effects of protein, monounsaturated fat, and car-
bohydrate intake on blood pressure and serum lipids: results of the OmniHeart
randomized trial. JAMA 2005; 294: 2455–2464.
5 Brinkworth GD, Noakes M, Keogh JB, Luscombe ND, Wittert GA, Clifton PM. Long-
term effects of a high-protein, low-carbohydrate diet on weight control and
cardiovascular risk markers in obese hyperinsulinemic subjects. Int J Obes Relat
Metab Disord 2004; 28: 661–670.
6 Delbridge EA, Prendergast LA, Pritchard JE, Proietto J. One-year weight main-
tenance after significant weight loss in healthy overweight and obese subjects:
does diet composition matter? Am J Clin Nutr 2009; 90: 1203–1214.
7 Larsen RN, Mann NJ, Maclean E, Shaw JE. The effect of high-protein, low-carbo-
hydrate diets in the treatment of type 2 diabetes: a 12 month randomised
controlled trial. Diabetologia 2011; 54: 731–740.
8 Leidy HJ, Carnell NS, Mattes RD, Campbell WW. Higher protein intake preserves
lean mass and satiety with weight loss in pre-obese and obese women. Obesity
(Silver Spring) 2007; 15: 421–429.
9 Meckling KA, Sherfey R. A randomized trial of a hypocaloric high-protein diet, with
and without exercise, on weight loss, fitness, and markers of the metabolic
syndrome in overweight and obese women. Appl Physiol Nutr Metab 2007; 32:
743–752.
Supplementary Information accompanies this paper on the Journal of Human Hypertension website (http://www.nature.com/jhh)
Journal of Human Hypertension (2014) 1 – 6
10 Larsen TM, Dalskov S, van Baak M, Jebb S, Kafatos A, Pfeiffer A et al. The
diet, obesity and genes (Diogenes) dietary study in eight European
countries—a comprehensive design for long-term intervention. Obes Rev 2010;
11: 76–91.
11 Larsen TM, Dalskov SM, van Baak M, Jebb SA, Papadaki A, Pfeiffer AF et al.
Diet, Obesity, and Genes (Diogenes) Project. Diets with high or low protein
content and glycemic index for weight-loss maintenance. New Engl J Med 2010;
363: 2102–2113.
12 Gogebakan O, Kohl A, Osterhoff MA, van Baak MA, Jebb SA, Papadaki A et al.
DiOGenes. Effects of weight loss and long-term weight maintenance with diets
varying in protein and glycemic index on cardiovascular risk factors: the diet,
obesity, and genes (DiOGenes) study: a randomized, controlled trial. Circulation
2011; 124: 2829–2838.
13 Moore CS, Lindroos AK, Kreutzer M, Larsen TM, Astrup A, van Baak MA et al.
DiOGenes. Dietary strategy to manipulate ad libitum macronutrient intake, and
glycaemic index, across eight European countries in the Diogenes Study. Obes Rev
2010; 11: 67–75.
14 Baecke JA, Burema J, Frijters JE. A short questionnaire for the measurement of
habitual physical activity in epidemiological studies. Am J Clin Nutr 1982; 36:
936–942.
15 Altorf-van der Kuil W, Engberink MF, Brink EJ, van Baak MA, Bakker SJ,
Navis G et al. Dietary protein and blood pressure: a systematic review. PLoS ONE
2010; 5: e12102.
16 He J, Wofford MR, Reynolds K, Chen J, Chen CS, Myers L et al. Effect of dietary
protein supplementation on blood pressure: a randomized, controlled trial.
Circulation 2011; 124: 589–595.
17 Teunissen-Beekman KF, Dopheide J, Geleijnse JM, Bakker SJ, Brink EJ,
de Leeuw PW et al. Protein supplementation lowers blood pressure in overweight
adults: effect of dietary proteins on blood pressure (PROPRES), a randomized trial.
Am J Clin Nutr 2012; 96: 966–971.
18 Claessens M, van Baak MA, Monsheimer S, Saris WH. The effect of a low-fat, high-
protein or high-carbohydrate ad libitum diet on weight loss maintenance and
metabolic risk factors. Int J Obes (Lond) 2009; 33: 296–304.
19 Westerterp-Plantenga MS, Lejeune MP, Nijs I, van Ooijen M, Kovacs EM. High
protein intake sustains weight maintenance after body weight loss in humans.
Int J Obes Relat Metab Disord 2004; 28: 57–64.
20 Neter JE, Stam BE, Kok FJ, Grobbee DE, Geleijnse JM. Influence of weight
reduction on blood pressure: a meta-analysis of randomized controlled trials.
Hypertension 2003; 42: 878–884.
21 Cirillo M, Lombardi C, Laurenzi M, De Santo NG. Gubbio Study Research G.
Relation of urinary urea to blood pressure: interaction with urinary sodium. J Hum
Hypertens 2002; 16: 205–212.
& 2014 Macmillan Publishers Limited