Intake of a protein-enriched milk and effects on muscle mass and strength. A 12-week randomized placebo controlled trial among community-dwelling older adults

Abstract

Objectives

To investigate the effect of 20 g protein with breakfast and evening meal on muscle mass, muscle strength and functional performance in older adults.

Design

A double-blinded randomized controlled study.

Setting

Oslo and Akershus University College of Applied Sciences, Norway.

Participants

Healthy community-dwelling men and women (≥ 70 years) with reduced physical strength and/or performance.

Intervention

Subjects were randomly assigned to receive either protein-enriched milk (2 x 0.4 L/d; protein group) or an isocaloric carbohydrate drink (2 x 0.4 L/d; control group) with breakfast and evening meal for 12 weeks.

Measurements

The primary endpoints were muscle mass measured by dual X-ray absorptiometry, and tests of muscle strength (one repetition maximum test of chest press and leg press) and functional performance (handgrip strength, stair climb and repeated chair rise).

Results

In total, 438 subjects were screened, 50 subjects were randomized and 36 completed the study. Chest press improved significantly in the protein (1.3 kg (0.1-2.5), p=0.03) and the control group (1.5 kg (0.0-3.0), p=0.048), but with no difference between the groups (p=0.85). No significant change in leg press (p=0.93) or muscle mass (p=0.54) were observed between the protein and the control group. Nor did we observe any significant differences in the functional performance tests (p>0.05 for all tests) between the groups.

Conclusion

Increased protein intake (2 x 20 g/d) did not significantly improve muscle mass, muscle strength or functional performance in healthy older weight stable adults. Whether intake of > 20 g protein to each meal is necessary for preservation of muscle mass and strength in older adults should be further investigated in a larger study. This underscores the need for well-designed studies that can differentiate between the effect of protein intake and increased energy. This trial was registered at Clinicaltrials.gov (ID no. NCT02218333).

Abbreviations

EAA

essential amino acid

1RM

one repetition of maximum test

eGFR

estimated glomerular filtration rate

Introduction

Maintaining muscle mass and muscle strength in older adults is important for preserving activities of daily life and for enabling aging people to remain living at home (1). Loss of muscle mass and strength with ageing can partly be explained by reduced level of physical activity. Other factors suggested to influence this are inadequate protein intake, reduced ability to utilize available proteins and increased need of proteins (2, 3, 4). To counteract losses and maintain muscle mass and strength, the recommended daily intake of protein in healthy older adults (> 65 years) was increased to 1.1-1.3 g protein/kg body weight in the revised Nordic Nutrition Recommendation in 2012 (2). A similar recommendation for healthy older adults was launched by the Study Group on Meeting Protein Needs of Older People (PROT-AGE Study Group) in 2013 (3). The PROT-AGE study group also emphasizes that the amount of high-quality protein to each meal may have an impact on optimal maintenance and to gain muscle mass and strength in older subjects, but the evidence are not yet sufficient to support specific recommendations.

Muscle protein synthesis has been suggested to be stimulated by the circulating plasma level of essential amino acids (EAAs) (5) and especially by the leucine concentration (6). When comparing the muscle protein synthesis in younger and older adults, reduced synthesis has been demonstrated in older adults from low protein intake (< 20 g/meal) and when proteins are ingested together with carbohydrates (7, 8, 9), but high protein intake (> 20 g/meal) stimulates muscle protein similarly (10, 11, 12, 13). Thus, it has been suggested that older subjects need a higher amount of EAAs than younger subjects, for generating muscle protein response (termed anabolic resistance) (14). To prevent loss in muscle mass in older adults, intake of 20-30 g of high quality protein with each meal has been recommended (15). In a 24-hour study conducted in healthy young adults (25-55 years), the protein synthesis was more effectively stimulated by adding ~ 30 g protein to each meal (breakfast, lunch and dinner), than by an intake of 10, 15 and 65 g of protein to each meal, respectively (16). In contrast, in a more recent study in older subjects, higher protein intake to one meal, rather than the same amount of protein consumed to several meals throughout the day (1.5 g protein/kg BW), showed a higher net anabolic response (protein synthesis and breakdown) (17). Thus, recent findings from short term studies in older adults, where net anabolic response is measured, suggest that the total daily protein intake may be of greater importance than the distribution pattern (14, 18).

In a randomized controlled study in frail older adults, 15 g protein provided with breakfast and lunch improved functional performance, but not muscle mass and strength after 24 weeks intervention (19). In sarcopenic subjects, intake of 20 g whey twice a day for 13 weeks did not improve handgrip strength or functional performance in a large randomized control trial (n=380) (20). Human intervention studies with amino acid/ protein supplement in apparently healthy older subjects are relatively scarce, and also with somewhat conflicting results (21, 22, 23, 24, 25, 26).

Adequate protein intake is necessary for preserving muscle mass and physical strength. Age-related loss of muscle mass and strength (sarcopenia) affects the ability to manage activities of daily life and the opportunity to live independent, and it has been associated with reduced quality of life, morbidity and even mortality (1, 27, 28). Increased life expectancy and growth in the population of older adults has been suggested to cause both economical and societal challenges. Thus, the impact of protein in prevention of sarcopenia needs to be further elucidated. In the present study we wanted to investigate whether increased daily intake of protein-enriched milk, which is a commercially available product with high quality protein, could improve muscle mass and strength among older adults. In a 12-week randomized placebo controlled trial, our primary aim was to investigate the effect of 20 g high-quality protein provided with breakfast and evening meal on muscle mass, muscle strength and functional performance in healthy community-dwelling older adults with reduced physical strength and/or performance.

Methods

Subjects

Men and women (≥ 70 years) living at home were recruited to a 12- week double-blinded, randomized controlled study conducted from August 2014 to September 2015 at Oslo and Akershus University College of Applied Sciences, Norway. Inclusion criteria were either reduced grip strength (< 20 kg in women and < 30 kg in men), gait speed < 1 m/s, timed step stair test ≥ 8.4 s or timed five times sit to stand test > 12.5 s, and willingness to keep the physical activity level stable throughout the study period. Exclusion criteria were one or more of the following: unable to perform physical tests, a Mini-Mental State Examination score < 24, a Mini Nutritional Assessment score < 17, weight change (≥ 3 kg last 3 months), allergy/intolerance to milk/dairy products, high intake of dairy products (≥ 0.4 L/day of milk, cultured milk and/or yoghurt) and alcohol consumption ≥ 40 g alcohol/d. Subjects with type I or II diabetes or HbA1c ≥ 6,5%, severe inflammation (including those using systemic glucocorticosteroids), chronic obstructive pulmonary disease, high blood pressure (> 180/105 mmHg), acute cardiovascular disease within the last six months or a history of cancer the last three years were also excluded. Furthermore, reduced kidney function (estimated glomerular filtration rate (eGFR) < 45 ml/ min), elevated CRP level (≥ 10 mg/L), and elevations (> three times the reference limit) of aspartate aminotransferase and/or alanine aminotransferase were also exclusion criteria. Subjects with thyroid-stimulating hormone outside reference range (0.2-10 mU/L) were only included if the thyroxine concentration was within the reference range (11.0-23.0 pmol/L). Thyroxine treatment, hormone therapy and use of antihypertensive drugs were allowed during the study, if a stable dosage had been maintained during the past three months. Subjects using calcium supplement could participate if the supplement was discontinued before entering the study.

Ethics

A written informed consent was given by all participants, and the study was conducted in accordance with the Declaration of Helsinki. All procedures involving human subjects were approved by the Regional Committees for Medical and Health Research Ethics, Health Region South East, Norway. Extracts from the National Population Registry was used according to, and with approval from, the Norwegian Tax Administration. The study was registered at Clinicaltrials.gov (ID no. NCT02218333).

Study design

Volunteers were recruited by postal mail. Invitation letters were sent to men and women (≥ 70 years of age) who were living in the area of Skedsmo, Norway, and listed in the National Population Register. A total of 2820 subjects were invited, 438 subjects met to the screening visit of which 388 were excluded, and 50 subjects were randomized. In total, 36 subjects completed the study. Figure 1 illustrates the flow of subjects through the trial. At baseline, subjects were stratified by gender and smoking, and randomly allocated within each stratum into two groups receiving either protein-enriched milk (2 x 0.4 L/d; 2 x 20 g protein/d) or an isocaloric carbohydrate drink (2 x 0.4 L/d) for 12 weeks. The subjects were encouraged to ingest the test drinks with breakfast and the evening meal, and to have ≤ 11 hours between the evening and the morning test drink to reduce the night fasting to ≤ 11 hours. Every one or two weeks, the subjects picked up the test drinks at the study center or the drinks were delivered to their homes by the study center staff. The subjects were encouraged to maintain their habitual diet and physical activity throughout the study period. Due to low recruitment rate (Figure 1) the study period was reduced from the originally planned 24 to 12 weeks, and the study was terminated after one year. The inclusion criteria that most subjects did not meet were low score on physical strength or functional performance, and low intake of dairy products.

Figure 1.

Flowchart of the study participants

Dietary assessment

Prior to the intervention period, two 24-h dietary recalls using a personal computer-assisted face-to-face interview, and an unannounced telephone or a face-to-face interview at the baseline visit, was performed by a dietitian. Additionally, two 24-h dietary recalls were prepared at the end of the study period, reflecting the dietary intake during the intervention period. The interviews were conducted in a three-step process as described elsewhere (29). Briefly, the interviews were performed using an in-house data program (KBS version 7.0) which was linked to the Norwegian food composition database. Dietary supplements were included in the analyses. All 24-h recall interviews were checked for errors and for plausibility, and interviews were conducted between Monday and Friday.

Blinding, side effect, adherence to study protocol

The test drinks were blinded for the study subjects and investigators by the identical color, labeling and otherwise appearance. The subjects received two cartons (0.4 L per carton) for each day, and each carton was labelled with either morning or evening drink. The subjects registered date and time for intake of each carton. Unopened containers and any remaining volume were returned to the study investigator staff. For each subject, the total volume of test drink ingested during the study period was divided by the scheduled volume for 100% compliance during the study period, to yield compliance in per cent (30). According to the study protocol, subjects with compliance ≤ 70% were excluded from the study. The mean (± SD) compliance was 97.8 ± 3.8% (n=17) in the protein group and 96.8 ± 5.7% (n =19) in the control group, and no subjects had a compliance ≤ 70%. To investigate changes in physical activity and to motivate to a stable level of physical activity level during the study period, the participants registered the daily physical activity of ≥ 30 min walking and other strength exercises such as heavy gardening and housework. The subjects completed the study within 12 ± 1 weeks.

Study products

The protein-enriched milk and the control drink were both produced and provided by TINE SA, Oslo, Norway. The protein-enriched milk is available for commercial sale in food stores in Norway, but was not enriched with vitamin D when used in our study. The protein-enriched milk provided on average 5.1% protein, 4.9% carbohydrates, < 0.1% fat and approximately 174 kJ (41 kcal)/100 g. Analysis showed that ~80% of the milk protein was casein and that the remaining ~20% was whey protein (Core Facility for Proteomics and Mass Spectrometry of Oslo University Hospital-Rikshospitalet, Oslo, Norway). An isocaloric, non-nitrogenous control drink was prepared from carbohydrates (sugar, xantan gum and maltosweetTM®). Calcium was added to the control drink to match the content in the protein-enriched milk of approximately 178 mg/100 g. Titadioksid (E171) (0.1%) was added to give the control drink a milky appearance. The composition of the test drinks was analyzed in different batches at an accredited laboratory (Eurofins Food & Feed Testing, Moss, Norway). The macro- and micronutrient composition in the test drinks are shown in Table 1A, and the amino acid content in the proteinenriched milk are shown in Table 1B.

Table 1A.

Composition of the protein-enriched milk and the control drink

Energy and nutrients	Protein-enriched milk	Control drink
	Per 100 g
Energy (kJ)	174	160
Fat (g)	0.1	0.1
Carbohydrates (g)	4.9	9.1
Protein (g)	5.1	< 0.1
Vitamin A (µg)	< 10	< 10
Vitamin E (mg)	< 0.08	< 0.08
Vitamin B1 (mg)	0.06	0.03
Vitamin B6 (mg)	0.04	0.01
Vitamin B12 (µg)	0.66	0.01
Calcium (mg)	1567	1667
Phosphorus (mg)	1200	< 50
Iodine (mg)	0.19	0.01
Riboflavin (µg)	0.17	< 0.01
Folate (µg)	14.3	14.6
Potassium (mg)	1500	< 50
Magnesium (µg)	130	< 50
Sodium (µg)	450	< 50
Copper (µg)	< 1.0	< 1.0
Selenium (µg)	0.02	< 0.01
Zink (µg)	5.8	< 0.1
Niacin (µg)	0.13	< 0.1

Table 1B.

Composition of amino acids in protein-enriched milk

Amino acids	g/100g	%	g/d
Leucine	0.51	9.4	4.1
Isoleucine	0.26	4.8	2.1
Valine	0.33	6.0	2.6
Alanine	0.17	3.1	1.4
Arginine	0.18	3.3	1.4
Aspartate	0.41	7.5	3.3
Cystein	0.04	0.7	0.3
Glutamic acid	1.13	20.6	9.0
Glycine	0.10	1.8	0.8
Histidine	0.14	2.6	1.1
Lysine	0.44	8.1	3.5
Methionine	0.14	2.5	1.1
Phenylalanine	0.26	4.7	2.0
Proline	0.50	9.2	4.0
Serine	0.30	5.5	2.3
Threonine	0.23	4.3	1.9
Tryptophan	0.07	1.4	0.6
Tyrosine	0.24	4.5	2.0
Essential amino acids	2.40	43.9	19.1

Each participants consumed 0.8 L protein-enriched milk/d.

Body composition

Dual energy X-ray absorptiometry (DXA) measurements were performed at the Norwegian School of Sport Sciences, Oslo, Norway. DXA (Lunar iDXA, GE Healthcare, Buckinghamshire, United Kingdom) and enCORE Software (version 14.10.022, GE Lunar) were used to estimate total and regional distribution of lean body mass and fat mass. Muscle mass was defined as the sum of lean body mass of the four limbs. To reduce the possibility of measurement errors due to biological variation in hydration status and food consumption, participants were scanned in the morning after an overnight fast (≥ 12 h).

Muscle strength and functional performance tests

Strength tests, stair climb test and repeated chair raise test were performed at the Norwegian School of Sport Sciences, Oslo, Norway. Before the physical tests were performed, the subjects consumed a standardized test meal consisting of yoghurt with wholegrain cereals (160 g) and juice (0.25 L), containing a total of 803 kJ (equivalent to 30 g carbohydrates, 15 g proteins and 1 g fat). Strength tests were performed in chest press and leg press machines (Technogym, Selection Line, Gambettola, Italy) using a one repetition maximum test (1RM). Before each test, the participants performed a standardized warm up protocol consisting of three sets with gradually increasing load and decreasing number of repetitions (10, 5 and 3 repetitions). The first attempt of 1RM was performed with an approximate 95% load of predicted 1RM, and after each successful attempt, the load was increased by 2-5% until the participant was unable to perform the required movement. The last acceptable attempt was determined as 1RM. Rest between each attempt was approximately three minutes. Verbal encouragement was used to maximize the efforts in each 1RM attempt. Repeated tests showed a coefficient of variation (CV) for these strength tests of < 2%.

The stair climbing test (20 steps, 0.16 m height) was performed under two conditions: without load and with 10 kg weight vest. Times were recorded using photo cells (Speedtrap 2, Brower Timing Systems, Utah, USA) positioned 0.85 m above the floor at the bottom of the stairs and 0.70 m beyond the last step. Each participant was given two attempts under each condition, and the best performances were registered. The CV was 1.3% and 2.1% for unloaded and loaded stair climb, respectively.

The repeated chair rise test involved five consecutive rises from a 0.46 m chair. The participants had to touch the backrest each time they sat down and straighten their knees and hip in the upright position (CV=5.2%).

Handgrip strength of the dominant and non-dominant hand was measured using a digital handheld dynamometer (KE-MAP80K1, Kern MAP, Elstra, Germany). The maximum grip strength of three measurements for each hand was registered.

Blood samples and routine laboratory analyses

Venous blood samples were drawn in the morning after an overnight fast (≥ 12 h), at the approximate same time at baseline and after 12 weeks, and the samples were prepared according to relevant protocols for analysis at an accredited medical laboratory (Fürst Medical Laboratory, Oslo, Norway). Serum was obtained from silica gel tubes (Becton Dickenson Vacutainer Systems, Plymouth, UK), kept at room temperature for at least 30 min until centrifugation (1500 g, 15 min). Whole blood samples were collected in EDTA tubes (Becton Dickenson Vacutainer Systems, Plymouth, UK).

Statistical analysis

The study subjects were stratified by gender and smoking status (current smoker or non-smoker) prior to randomization. The 1:1 block-randomization was performed by a statistician. The randomization code was concealed from the study investigators until the statistical analyses were completed. Assuming a clinically relevant difference of 0.5 (SD 1.0) kg in muscle mass between the study groups after 12 weeks, and with a 20 % drop-out rate during the study period, recruitment of 75 participants in each study arm (a total of 150) was required to detect a difference (power of 80% and significance level of 5%). Changes from baseline within each study group were performed using the One-sample t-test, and differences between the study-groups at baseline and after 12 weeks were performed by the Independent samples t-test. Transformation was prepared in not normally distributed variables (circulating level of hs-CRP and C-peptide), which was not found normally distributed after transformation. Thus, non-parametric tests, Wilcoxon signed-rank test and Wilcoxon-Mann-Whitney, were used to analyze differences within and between study groups at baseline and after 12 weeks, respectively. Correlation analyses were performed by calculating Pearson correlation coefficients. Significance was defined as p < 0.05, and all tests were twosided. All analyses were performed using SPSS for Windows (version 22.0; SPSS, Inc., Chicago, IL, USA).

Results

Characteristics of the subjects

In total, 36 subjects (24 women and 12 men) completed the study. Baseline characteristics and dietary intakes are presented in Tables 2 and 3. No significant baseline differences between the protein group and the control group were observed, except for differences in the dietary intake.

Table 2.

Baseline characteristics

	Protein group	Control group
Gender (n men/ n women)	5/12	7/12
Age (y)	76.8 ± 6.2	77.1 ± 4.7
Body mass index (kg/m2)	27.6 ± 4.2	25.9 ± 4.9
Systolic blood pressure (mmHg)	141 ± 20	145 ± 22
Diastolic blood pressure (mmHg)	79 ± 11	77 ± 11
Mini-Mental State Examination score	27.9 ± 2.0	27.6 ± 2.3
Mini Nutritional Assessment score	27.3 ± 2.2	27.7 ± 1.7
Total SPPB1 score	11.1 ± 1.2	11.0 ± 1.6

Data are presented as mean ± SD. Differences between groups were analyzed using Independent samples t-test. No significant differences between groups were observed (P > 0.05). 1. SPPB, short physical performance battery.

Table 3.

Dietary intake from 2 x 24 t recall performed prior and during the study period

	Protein group		Control group
	0 wk	12 wk	0 wk	12 wk
Energy and nutrients	Mean ± SD	Mean ± SD	Mean ± SD	Mean ± SD	P1	P2
Energy (MJ)	6.9 ± 2.1	7.8 ± 1.6	7.4 ± 2.6	7.7 ± 1.9	0.61	0.89
Protein (E %)	18.8 ± 4.3	23.9 ± 4.6	16.9 ± 3.6	15.0 ± 4.6	0.15	< 0.0001
Protein (g)	77.5 ± 20.8	103.6 ± 16.1	72.3 ± 18.2	66.9 ± 24.1	0.42	< 0.0001
Protein (g/kg body weight)	1.0 ± 0.3	1.4 ± 0.5	1.0 ± 0.3	0.9 ± 0.4	0.91	0.003
Total fat (E %)	34.6 ± 5.8	33.0 ± 7.5	40.5 ± 7.3	30.6 ± 6.3	0.012	0.31
Saturated fat (E %)	13.2 ± 3.0	12.6 ± 3.3	16.9 ± 3.5	12.6 ± 3.3	0.002	0.96
Polyunsaturated fat (E %)	6.2 ± 2.2	5.9 ± 2.4	6.4 ± 2.6	4.7 ± 1.9	0.78	0.09
Monounsaturated fat (E %)	11.6 ± 1.5	11.2 ± 3.4	13.4 ± 3.5	10.6 ± 2.6	0.06	0.54
Carbohydrates (E %)	43.4 ± 6.2	33.1 ± 7.5	38.5 ± 6.7	42.2 ± 6.6	0.030	0.001
Added sugar (E %)	7.8 ± 4.3	5.9 ± 3.5	6.0 ± 3.7	5.2 ± 3.3	0.24	0.55
Fiber (E %)	2.4 ± 0.8	1.7 ± 0.5	2.0 ± 0.5	1.7 ± 0.5	0.049	0.84
Alcohol (E %)	0.0 (0.2)	0.0 (0.0)	0.8 (3.9)	0.0 (0.0)	0.028	0.99

Data are presented as mean ± SD or median (IQR). Content of the test drinks (protein-enriched milk and the carbohydrate content of the control drink) are included in the dietary intake after 12 weeks. Between group effects (protein group versus control group) were analyzed with Independent samples t-test or Wilcoxon-Mann-Whitney test at ¹ baseline and ² on changes from baseline. Two subjects were missing after 12 weeks (one in each group).

Body composition and body weight

The subjects in the two study groups were similar (p > 0.05) at baseline in body composition (all variables) and body weight (Table 4). Following 12 weeks of intervention, no significant increase in muscle mass was observed in the protein group (0.1 kg (-0.1 to 0.4), p=0.38), in the control group (0.2 kg (0.0 to 0.4), p=0.05) or between the two groups (p= 0.54). Nor were there any significant differences in change in total lean body mass, fat mass, body fat percentage or body weight between the two groups (Table 4).

Table 4.

Effect of protein-enriched milk and isocaloric control drink on body composition and body weight

	Protein group		Control group		Between groups
	Mean ± SD	Δ Mean (95% CI)	Mean ± SD	Δ Mean (95% CI)	P
Muscle mass (kg)	19.7 ± 4.0	0.1 (-0.1 to 0.4)	19.8 ± 4.8	0.2 (0.0 to 0.4)	0.54
Total lean body mass (kg)	44.1 ± 7.2	0.4 (0.0 to 0.8)	44.3 ± 8.9	0.4 (0.0 to 0.9)	0.85
Trunk lean body mass (kg)	21.2 ± 3.3	0.2 (0.0 to 0.5)	21.4 ± 3.9	0.0 (-0.3 to 0.4)	0.33
Fat mass (kg)	29.2 ± 9.2	0.0 (-0.5 to 0.5)	24.6 ± 9.6	0.3 (-0.4 to 1.0)	0.48
Body fat (%)	39.2 ± 7.3	-0.2 (-0.7 to 0.3)	35.0 ± 7.8	0.0 (-0.7 to 0.7)	0.64
Body weight (kg)	75.3 ± 14.4	0.7 (-0.1 to 1.5)	71.4 ± 16.1	0.6 (-0.0 to 1.3)	0.95

Data at baseline for each group are presented as mean ± SD, and mean changes from baseline are presented with 95% confidence interval (CI). Between group effects were analyzed with Independent samples t-test on changes from baseline. Muscle mass; appendicular lean body mass.

Muscle strength and functional performance

The protein group and the control group were similar (p > 0.05) at baseline in terms of all physical strength- and functional performance tests (Table 5). After 12 weeks, chest press improved significantly in the protein (1.3 kg (0.1 to 2.5), p=0.03) and the control group (1.5 kg (0.0 to 3.0), p=0.048) with no significant difference between the groups (p=0.85). No significant difference in leg press was observed within or between the two groups (p=0.93) (Table 5). The stair climb test (without load) improved significantly in the protein group (-0.4s (-0.8 to 0.1), p=0.03), but not in the control group (0.0s (-0.7 to 0.7), p=0.96), but the change in the protein group was not significantly different from the control group (p=0.22). Nor did we observe any significant differences between the two groups in the changes in the functional performance tests: repeated chair rise, stair climb test (with 10 kg load), and hand grip strength (Table 5).

Table 5.

Effect of protein-enriched milk and isocaloric control drink on physical strength and performance

	Protein group		Control group		Between groups
	Mean ± SD	Δ Mean (95% CI)	Mean ± SD	Δ Mean (95% CI)	P
Leg press (kg)	144.0 ± 30.9	5.7 (-1.7 to 13.1)3	146.4 ± 43.23	6.2 (-2.2 to 14.6)4	0.93
Chest press (kg)	32.5 ± 9.1	1.3 (0.1 to 2.5)	33.3 ± 13.9	1.5 (0.0 to 3.0)3	0.85
Repeated chair raise (s)	9.0 ± 1.6	-0.2 (-1.2 to 0.7)3	8.9 ± 2.53	-0.3 (-1.2 to 5.4)4	0.83
Stair climb (s)	9.9 ± 1.4	-0.4 (-0.8 to - 0.1)3	9.7 ± 1.84	0.0 (-0.7 to 0.7)5	0.22
Stair climb 10 kg (s)	10.3 ± 1.7	-0.2 (-0.7 to 0.3)3	10.5 ± 2.64	-0.2 (-1.0 to 0.6)5	0.94
Handgrip dominant (kg)	24.5 ± 6.9	0.1 (-0.7 to 1.0)	25.3 ± 10.13	- 0.5 (-1.5 to 0.5)3	0.27
Handgrip nondominant (kg)	22.1 ± 5.6	0.5 (-0.5 to 1.5)	22.8 ± 8.1	0.5 (-0.4 to 1.4)	0.99

Data at baseline for each group are presented as mean ± SD, and mean changes from baseline are presented with 95% confidence interval (CI). Between group effects were analyzed with Independent samples t-test on changes from baseline. ³ One subject missing. ⁴ Two subjects missing. ⁵ Four subjects missing.

Correlation analysis

At baseline, a strong relationship between muscle mass and muscle strength (sum of chest press (1 RM) and leg press (1 RM)) was observed for the total study sample, r=0.62 (p<0.0001) (n=35). However, the change in muscle mass in the protein group was not positively related to change in muscle strength (sum of chest press (1 RM) and leg press (1 RM)) following 12 weeks intake of protein-enrich milk r= -0.4 (p=0.08) (n=17). Nor was there any relationship between muscle mass and the relative muscle strength (sum of chest press (1 RM) and leg press (1 RM)/body weight (kg)), at baseline for the total study sample r= -0.10 (p=0.57) (n=35).

Dietary intake

Dietary intake of total fat, saturated fat, carbohydrates and fiber were significantly different between the two study groups at baseline, while the protein intake was similar in both groups (1.0±0.3 g/kg body weight/d) (Table 3). However, during the study period, the amount of calories and the macro nutrient composition in the background diet were not significantly different between the study groups. When we added the test drinks to the background diet, the only significant differences between the study groups was a significantly higher intake of protein in the protein group, when compared to the control group (1.4 ± 0.5 versus 0.9 ± 0.4 g /kg body weight/d, p=0.003). Additionally, a significant lower intake of carbohydrates in the protein versus control group was observed (33.1 ± 7.5 E% versus 42.2 ± 6.6 E%, p=0.001).

Measurements of renal and lipid biomarkers

Blood urea nitrogen levels increased significantly after intake of protein-enriched milk and serum creatinine levels increased significantly after intake of carbohydrates. Significant differences in changes of urea nitrogen and creatinine levels were thus observed between the two groups (p < 0.0001 and p=0.04, respectively) (Table 6). Following 12 weeks of intervention, the proportion of subjects with reduced eGFR (< 60 ml/min/1.73m2) increased from one to four subjects after intake of carbohydrates, while no subjects had reduced eGFR after intake of protein-enriched milk. The change in eGFR was not significantly different between the study groups (p=0.09).

Table 6.

Effect of protein-enriched milk and isocaloric carbohydrate drink on blood values

	Protein group		Control group		Between groups
	Mean ± SD	Δ Mean (95% CI)	Mean ± SD	Δ Mean (95% CI)	P
S-Total cholesterol (mmol/L)	5.4 ± 0.8	-0.5 (-0.8 to -0.2)3	5.3 ± 0.8	-0.1 (-0.4 to 0.2)3	0.06
S-LDL-cholesterol (mmol/L)	3.4 ± 1.0	-0.3 (-0.5 to -0.1)3	3.2 ± 0.7	-0.1 (-0.3 to 0.1)3	0.25
S-HDL-cholesterol (mmol/L)	1.5 ± 0.4	0.0 (-0.1 to 0.1)	1.7 ± 0.6	0.0 (-0.0 to 0.1)3	0.41
S-Triglycerides (mmol/L)	1.3 ± 0.4	-0.1 (-0.3 to 0.0)3	1.2 ± 0.53	0.1 (- 0.2 to 0.2)3	0.05
S-F-Glucose (mmol/L)	5.3 ± 0.5	-0.1 (-0.3 to 0.2)	5.3 ± 0.8	0.1 (-0.1 to 0.3)3	0.36
S-C-peptid (pmol/L)1	578 ± 240	-47 (-101 to 7)	604 ± 313	-28 (-121 to 65)	0.96
S-hsCRP (mg/L)1	4.3 ± 5.7	1.5 (0.0 to 2.6)	2.5 ± 3.6	0.0 (-0.2 to 0.2)3	0.07
S-Urea (mmol/L)	6.8 ± 1.3	1.8 (1.2 to 2.5)	6.2 ± 1.3	-0.1 (-0.1 to 0.5)3	< 0.0001
S-Creatinine (µmol/L)	73.2 ± 14.5	-0.1 (-2.8 to 2.7)	75.6 ±16.0	6.0 (1.0 to 11.0)3	0.04
eGFR (ml/min/1.73m2)	75 ± 13	0.29 (-3.1 to 2.5)	74 ± 12	-4.4 (-8.6 to -0.3)3	0.09

hsCRP, high-sensitive C-reactive protein; eGFR, estimated glomerular filtration rate. Data at baseline for each group are presented as mean ± SD, and mean changes from baseline are presented with 95% confidence interval (CI). Between group effects were analyzed with Independent samples t-test or Wilcoxon-Mann-Whitney test on changes from baseline. ¹ Not normally distributed ³ One subject missing. ⁴Two subjects missing.

Serum total- and LDL-cholesterol were significantly decreased in the protein group after 12 weeks (p=0.002 and p=0.046, respectively). However, when compared to the control group, changes in total- and LDL-cholesterol and triglycerides were not significantly different (p=0.06, p=0.24 and p=0.05 respectively) between the two groups.

Discussion

To preserve muscle mass and strength with aging, short term studies have suggested that intake of > 20 g protein to each meal is required in older adults to achieve maximal muscle protein synthesis (5, 9, 12, 15, 31). In the present study, we were not able to demonstrate that intake of 20 g protein to breakfast and evening meal in older adults for 12 weeks changed muscle mass, muscle strength or physical performance differently, when compared to an isocaloric carbohydrate drink.

Previous amino acid/protein supplemental studies in healthy older adults, where physical activity was not a part of the intervention program, are relatively scarce and with somewhat conflicting results (21, 22, 23, 24, 25, 26). In accordance with our results, Verhoeven and coworkers observed no changes in muscle mass or strength in healthy men (~ 71 years, n=30) after intake of leucine (3 x 2.5 g/d) or placebo (wheat flour) for three months (22). Similarly, essential amino acids (0.21 g /kg body weight/d, three months, n=25) did not lead to increased lean tissue mass in healthy men (65-75 years) when compared to an isocaloric control (lactose) (21). However, functional performance tests (30 s chair-stand test and 6 min walking test) improved after three months. In contrast, total body- or appendicular lean mass increased in healthy older adults (> 60 years) after adding a protein-rich cheese (18 protein g/d) to the habitual diet (control: habitual diet), capsules with essential amino acid (15 g/d) (control: unspecified content) or extra proteins (0.165 g protein/kg body weight/d) for breakfast and lunch (control: isocaloric carbohydrate drink) (23, 24, 25). Muscle protein synthesis has been shown to be stimulated by EAAs in a dose-dependent manner (9), reaching a plateau at 15 g EAAs per meal (32). More recent findings suggest a linear relationship between EAAs and the net anabolic response (14, 18), which is in accordance with previous nitrogen balance studies used to define the protein requirement (2). Whether the amount of EAAs provided to each meal in the present study (~9.5 g/meal), and in some of the studies discussed above, can explain the lack of improvement on muscle mass and strength is uncertain, but not unlikely. However, the non-significant findings are in agreement with and supported by the conclusion in a recent meta-analysis where no significant effect on muscle mass and physical strength from amino acid/protein supplements were observed in older patients (33). Additionally, increased muscle mass from intake of amino acids/ protein supplement have not consistently been associated with improved physical strength in sarcopenic subjects (33, 34). Interestingly, adding 30 g whey protein/d to the daily diet for two years in older sarcopenic women (~74 years), did not improve maintenance of muscle mass or hand grip strength when compared to a control drink (2.1 g protein/d) (26).

At present, providing a higher daily intake of proteins to overcome the anabolic resistance has been suggested to be of importance in older subjects (14, 18). However, the hypothesis that a high protein intake to each meal has a more beneficial effect when compared to a skewed protein intake has recently been questioned from studies where the net anabolic response has been measured (14, 18). A benefit from a high protein intake to each meal for maximal muscle protein synthesis in order to preserve muscle mass and strength seem difficult to confirm in long-term studies. Thus, an additional effect beyond the daily recommended intake of 1.0-1.3 g protein/kg body weight in order to preserve muscle mass and strength therefore remains uncertain. The adequate total daily intake of protein among the older subjects recruited in the present study (1.0 ± 0.3 g per kg body weight) could, at least partly, explain the lack of effects on the primary outcomes. Another possible explanation could be the apparently healthy older subjects included. In studies where EAAs or protein supplements have been provided to either older subject with disease-related malnutrition, during bed rest, to frail or sarcopenic older subjects, improved muscle strength and physical performance have been observed (19, 34, 35, 36).

In the present study, the average increase in chest press strength was approximately 5% in both study groups. In order to remain home living, preservation of muscle strength is considered more important than maintenance of muscle mass. Thus, in the perspective of clinical relevance, the improvement in muscle strength within both groups must be considered important. Combining the two study groups, a significant improvement of chest press (1RM), and leg press (1RM) and muscle mass were observed (data not shown), suggesting a positive effect mediated by increased energy intake independent of macronutrient source.

Interestingly, Arnarson and coworkers demonstrated that healthy older subjects (n=161) improved lean body mass, physical strength and functional performance similarly after 12 weeks from intake of either whey protein (20 g/d) or an isocaloric carbohydrate drink provided immediately after resistance exercise. No between-group effects were observed (37). Similar findings were found in a smaller study in healthy older women (n=12), where a within-subject study design was used, which allowed a direct comparison from intake of whey protein (40 g) versus isocaloric carbohydrate drink. Increased muscle strength and muscle thickness were observed from resistance training twice a week for ten weeks, but with no difference between intake of whey protein and the control drink (38). Similarly, in a larger supplement study (n=380) where sarcopenic subjects received either whey (2x20 g/d) or an isocaloric carbohydrate drink for 13 weeks, significantly improved handgrip strength and short physical performance battery score (primary endpoints) were observed in both study groups and no between-group differences were shown (20). These studies are in accordance with the present study, showing no effect of protein when compared to isocaloric intake of carbohydrates. The non-significant results observed in the amino acid/protein supplemental studies may have several explanations, including insufficient content of essential amino acids, ingestion of protein together with carbohydrates, and the potential beneficial effect achieved by increased energy intake. Combining amino acid/protein supplements with physical activity improved muscle mass or strength in some studies (34, 39, 40), while in others not (37, 41, 42). Increased physical activity (both endurance- and resistance-type exercise) is most likely the most effective treatment of low of muscle mass and strength (34, 40, 43, 44).

Frequently, concern is raised regarding the impact of high protein intake on kidney function in older adults (2, 3, 45). In the present study, the serum creatinine level only increased after intake of carbohydrates, thus, the estimated glomerular filtration rate was only reduced in the control group. However, hyperfiltration of blood is an adaptive mechanism of increased protein intake (46), and may explain why the serum creatinine level only increased in the control group. Further, whether protein-enriched milk has a beneficial effect on blood lipids, should be further investigated in larger studies in older adults.

A major limitation of the present study was the lower number of subjects randomized than estimated by the power calculation and the high number (33%) of participants discontinuing after randomization. The discontinuation was mainly related to discomfort from intake of the test drinks, which in part was due to the large amounts (0.4L twice a day) of test drink to be consumed with breakfast and evening meal. We also experienced that some of the participants had dayto- day variation in health and physical pain (e.g. back pain), and to avoid that physical testing caused any physical harm, not all tests were performed on all subjects. The low number of subjects affects the external and the statistical conclusion validity of the present study. The observed differences are considerably smaller than those assumed in power calculations. This may imply a substantially larger number of participants were needed than it was estimated by a priori power calculations. The small treatment effect might indicate that the effect of intake of protein-enriched milk (2 x 20 g high quality protein/d) versus an isocaloric carbohydrate drink may be of little clinical relevance. However, we cannot exclude the possibility that long term intake of protein-enriched milk could lead to increased maintenance or improved muscle mass and strength.

A major strength of the present study was the randomized and double-blind study design and the administration of an isocaloric control drink similar in appearance as the proteinenriched milk. Additionally, the baseline characteristics were similar with regard to primary outcomes, the self-reported compliance to both test drinks was high (97.3 ± 4.9%), and the compliance in the protein-enriched milk group was supported by the significant increase in blood urea level (Table 6). Finally, the present study identifies potential practical problems such as recruitment of home-dwelling healthy older adults with reduced physical strength or functional performance, and it reveals insights for other researchers planning future intervention studies in this study population.

Conclusion

In the present study, we were not able to demonstrate that intake of additional 20 g protein with breakfast and evening meal was more effective compared to intake of an isocaloric drink with respect to gain or maintain muscle mass or strength in older adults with reduced physical strength and/or physical performance. Whether a high intake of protein to each meal is necessary for preservation of muscle mass and strength in older adults thus needs to be further investigated in a larger study. The significant improvement in muscle mass and strength observed after the intervention from combining the two study groups, underscores the need for well-designed studies that can differentiate between the effect of protein intake and increased energy.

Acknowledgments: The authors gratefully acknowledge the participants who volunteered to this study, and we thank Ellen Raael, Navida-Akhter Sheikh, Marit Sandvik, Linn Øyri, Kristin S. Sandvei, Kristin Torvik and Grete Skjegstad for valuable assistance in this project. I.O., L.F.A., A.B., A.S.B., K.R., P.O.I., T.R., S.M.U. and K.B.H. designed the study; I.O., A.T.L., H.H. G.O.G and A.S.B. conducted the research; I.O. and J.B-S performed statistical analyses; I.O., T.R., S.M.U. and K.B.H. wrote the paper. All authors read and approved the final manuscript.

Competing interest: The present study was supported by the Research Council of Norway (grant number 225258/E.40). The funding source had no role in the data collection, analysis or interpretation of the data.

Conflict of Interest: I Ottestad, A T Løvstad, H Hamarsland, J Šaltytė Benth, L F Andersen, A Bye, P O Iversen and T Raastad have no conflicts of interest. K. Retterstøl reports grants from Oslo Economics, personal fees from Amgen, Mills DA, Sanofi, The Norwegian Medical Association, and The Norwegian Directorate for Health; none of which are related to the contents of this manuscript. S M Ulven has received research grants from Mills DA and Olympic Seafood; none of which are related to the content of this manuscript. K B Holven reports grants from Mills DA, TINE DA, Olympic Seafood, Sanofi, Pronova and personal fees from Amgen; none of which are related to the contents of this manuscript. The protein-enriched milk and the isocaloric carbohydrate drink were provided by TINE SA, Oslo, Norway, where G O Gjevestad and A S Biong are researchers employed. They have no financial interest to declare.

Ethical Standards: The study complies with the current laws in Norway.