SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgments
  9. References

Objective

To examine the association of habitual animal and plant protein intake during the potentially critical period of puberty with body composition in young adulthood.

Design and Methods

Multivariable regression analyses were performed on data from 140 female and 122 male participants of the DONALD Study with ≥2 3-day weighed dietary records during puberty (girls 9-14 years; boys 10-15 years) and anthropometric measurements in young adulthood (18-25 years). Fat-free mass index (FFMI) and fat mass index (FMI) were estimated from four skinfolds.

Results

In women, a higher pubertal animal protein consumption was independently related to higher levels of FFMI (ptrend = 0.001), but not to FMI (ptrend = 0.5). Adjusted means of FFMI in energy-adjusted tertiles of animal protein intake were 15.3 (95% confidence interval: 15.0, 15.5), 15.4 (15.1, 15.7), 16.2 (15.9, 16.6) kg/m2. In men, a higher animal protein intake was related to a higher FFMI (ptrend = 0.04) and a lower FMI (ptrend = 0.001) only after adjusting FFMI for current FMI levels and vice versa. Plant protein was not associated with body composition among either sex.

Conclusions

Our results show that a higher pubertal animal protein consumption may yield a higher fat-free mass in young adulthood.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgments
  9. References

Substantial controversy exists concerning a potential effect of dietary protein intake on body mass and body composition. A beneficial effect of a higher dietary protein content has been observed in a number of weight loss- and weight control-trials [1-5]. On the other hand, some evidence from prospective observational studies points to a detrimental effect of (animal) protein intake on body weight and body mass index (BMI) in adults over the long term [6-9]. While it is plausible that these two study types—with their different designs, contexts, and times of duration—yield diverging results, the issue of how long-term protein intake relates to health remains far from being solved.

The particularities of study designs are not the only obstacle in assessing the evidence. The limited validity of body weight or BMI as proxies for body fat may be of special relevance with regard to dietary protein: In a recent randomized controlled trial (RCT) with healthy young adults [10], overeating on a low protein diet produced less weight gain than overeating on a diet with a normal or high protein content. Yet, the additional weight gained on the higher protein diets stemmed from fat-free mass only. Subsequently, a link of higher protein intakes to higher body weight may not be specific to fat mass. Puberty is a developmental phase during which major changes in body composition occur. It is possible that, because of the anabolic nature of metabolism during this phase [11], a diet that is relatively high in protein favors a body composition characterized by a higher fat-free mass.

In this study, we investigated the association of habitual protein intake during puberty with fat mass index (FMI) and fat-free mass index (FFMI) in young adulthood. A secondary aim was to consider protein intake during early childhood (age 12-24 months) and adiposity rebound (age 4-6 years) as these windows represent, similar to puberty, potentially critical periods for later obesity risk [12, 13].

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgments
  9. References

Study population

The present study was ancillary to the Dortmund Nutritional and Anthropometric Longitudinally Designed Study (DONALD Study), an ongoing, open cohort study conducted in Dortmund, Germany. Details on this study have been described elsewhere [14]. Briefly, since the recruitment began in 1985, detailed data on diet, growth, development, and metabolism between infancy and adulthood have been collected from over 1300 healthy children. The study was approved by the Ethics Committee of the University of Bonn, and all examinations are performed with parental consent.

Because of the open cohort design, many DONALD participants have not yet reached young adulthood. In total, data from 394 subjects aged > 18 years were available for this analysis. These subjects were term (37-42 week gestation) singletons with a birth weight > 2500 g and had at least one anthropometric measurement taken in young adulthood (≥ 18 and ≤ 25 years of age, mean age = 20.3 years), of which we used the last for the present analysis. Three hundred and eight of these participants had provided at least two 3-day weighed dietary records during puberty (girls 9-14 years, boys 10-15 years). Participants who consistently underreported their energy intake (i.e. who had provided more implausible [15, 16] than plausible food records) were excluded from the study (n = 23). Furthermore, participants had to have anthropometric data at puberty and information on relevant covariates such as early life and socioeconomic factors. This resulted in a final sample of 262 participants (53.6% female, 46.4% male). Overall, 1376 food records were included in the present analysis (2-7 records per participant, on average five per subject). Concerning our additional analyses on the associations of protein intake during early childhood and adiposity rebound with adult body composition, data were available for 159 participants (86 males, 73 females) and 220 participants (107 males, 113 females), respectively. As these sample sizes are quite small and as our main focus was puberty, the respective results are not presented in detail.

Anthropometric measurements

Participants are measured at each visit according to standard procedures [17], dressed in underwear only and barefoot. From the age of 2 years onward, standing height is measured to the nearest 0.1 cm using a digital stadiometer (Harpenden, Crymych, UK). Body weight is measured to the nearest 100 g using an electronic scale (Seca 753E; Seca Weighing and Measuring Systems, Hamburg, Germany). Skinfold thicknesses are measured from the age of 6 months onward at four different sites (suprailiacal, subscapular, biceps, triceps) on the right side of the body to the nearest 0.1 mm using a Holtain caliper (Holtain, Crosswell, United Kingdom). The three trained nurses performing the measurements undergo annual quality controls, conducted in six to eight healthy young adult volunteers. Average inter- and intra-individual variation coefficients obtained in the last seven years (2005-2011) were 9.1% and 12.8% for biceps, 5.0% and 5.9% for triceps, 5.1% and 7.6% for subscapular, and 8.4% and 8.5% for supra-iliacal skinfolds.

Anthropometric calculations

Body fat mass and fat-free body mass were calculated as “(percentage body fat (%BF) * body mass) /100” and “((100 − %BF) * body mass) /100”, respectively, and related to height to obtain the indices FMI and FFMI (kg/m2). %BF at puberty was derived using equations of Slaughter et al. for pubescent children [18], which consider triceps and subscapular skinfolds. %BF in young adulthood was estimated from skinfolds using Durnin and Womersley equations, which are based on triceps, biceps, scapular and iliacal skinfolds [19]. We chose to investigate FMI and FFMI rather than %BF as the use of this measure has recently been criticized to incorrectly reflect body-size-adjusted adiposity [20].

Nutritional assessment

Food consumption in the DONALD Study is assessed annually using 3-day weighed dietary records. All foods and beverages consumed are weighed and recorded, as well as leftovers, to the nearest 1 g over three days using electronic food scales (initially Soehnle Digita 8000; Leifheit SG, Nassau; Germany; now WEDO digi 2000; Werner Dorsch GmbH, Muenster/Dieburg, Germany). For this analysis, dietary variables were calculated as individual means of the 3-day weighed dietary records using LEBTAB, the in-house database, which is continuously updated to include all recorded food items. LEBTAB is based on the German standard food tables [21] and data obtained from commercial food products. Individual average intakes were calculated from at least two records during puberty. In a validation study conducted with data from DONALD participants, the correlation coefficient between protein intake as determined by 3-day weighed records and as determined by 24-h-urinary excretion was 0.59 among 11-13 year olds [22]. In order to create the food groups “dairy protein” and “meat protein”, foods were broken down into their components as appropriate (e.g. a pizza was broken down into dairy products, meat products, cereal products, and other product groups).

Early life and socioeconomic characteristics

On their child's admission to the study, parents are interviewed by the study paediatrician, and weighed and measured by the study nurses using the same equipment as for children from two years onward. Information on the child's birth characteristics is abstracted from the ‘Mutterpass’, a standardized document given to all pregnant women in Germany. The duration of full breastfeeding (no solid foods or liquids other than breast milk, tea or water) is inquired by dieticians at the first visits. For this analysis, the following early life and socioeconomic characteristics were considered as potentially confounding factors: breastfeeding status (fully breastfed (yes/no), defined as fully breastfed > two weeks), birth weight (two variables were tested: < 3000 g vs. ≥ 3000 g and < 3500 g vs. ≥ 3500 g (an approximate median split)) maternal overweight status (BMI ≥ 25 kg/m2) and high maternal educational status (≥ 12 years of schooling).

Statistical analysis

Participant characteristics are presented by gender and energy-adjusted tertiles of animal protein intake at puberty. Tests for differences across tertiles were performed using Kruskal-Wallis tests for continuous variables and χ2-tests for categorical variables. The association between diet during puberty and body composition in young adulthood was analyzed by multiple linear regression models. As FMI and FFMI in puberty and young adulthood were not normally distributed, they were log-transformed (FMI) or 1/x-transformed (FFMI). To account for the major gender-specific differences in the development of body composition [11], all analyses were stratified by sex. Age at the measurement in young adulthood and the respective baseline anthropometric variable (FMI, FFMI) were included in basic models (Model A). Early life and socioeconomic factors were considered separately and included if they modified the respective association substantially (that is, if their inclusion caused a change in the regression coefficient for protein intake of > 10% [23]). In a further step, we additionally adjusted for nutritional factors using the same criterion (model B). Here, we merely considered total energy intake and nutritional factors which do not provide energy (dietary glycemic index, dietary fiber, calcium), so as to avoid presenting estimates that partially reflect the substitution of protein for other macronutrients. Instead, we ran additional models that explicitly assess the effect of a substitution of animal/plant protein for carbohydrates or for fats. To simulate substitution effects, total energy and the energy-bearing nutrients to be held constant (fats and plant/animal protein or carbohydrates and plant/animal protein, respectively) were included in the models [24]. We only present substitution models for associations identified as significant in fully adjusted analyses (model B).

In order to understand more specifically how protein intake was related to fat-free mass—independently of current fat mass—and vice versa, we ran additional analyses in which we adjusted adult FFMI levels for adult FMI levels and adult FMI levels for adult FFMI levels.

All dietary variables were energy-adjusted using the residual method [24]. To account for age-dependent changes in intake levels, all variables were standardized by age group and sex. In addition to multiple linear regression analyses, which were used to obtain information on linear trends (pfor trend), we conducted analyses of covariance. Here, protein intake was included as a categorical variable (in the form of energy-adjusted tertiles) to obtain adjusted means (least-squares means) of FMI and FFMI in young adulthood by tertiles of protein intake. A P-value < 0.05 was considered statistically significant. All analyses were carried out using SAS procedures (version 9.1.3, SAS Institute, Cary, NC, USA).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgments
  9. References

In Table 1, early-life and familial characteristics and anthropometric data of the 140 female and the 122 male participants are presented by energy-adjusted tertiles of animal protein intake during puberty. Male participants in the highest tertile had a higher FFMI at baseline, but did not differ in their adult anthropometry. Females in the highest tertile had higher pubertal levels of BMI as well as higher adult levels of BMI, FMI, and FFMI and were the most likely to be overweight in young adulthood. Moreover, they were least likely to have a mother with a high educational status.

Table 1. Demographic, anthropometric, birth, and socioeconomic characteristics by energy-adjusted tertiles of animal protein intake during puberty
 Total nMales (n = 122)Females (n = 140)
Tertile 1 (n = 40)Tertile 2 (n = 40)Tertile 3 (n = 41)PaTertile 1 (n = 46)Tertile 2 (n = 47)Tertile 3 (n = 47)Pa
  1. DONALD Study, Germany

  2. Abbreviations: BMI, body mass index; FMI, fat mass index; FFMI, fat-free mass index.

  3. Values are medians (25th percentile, 75th percentile) for continuous variables and n (%) for categorical variables.

  4. a

    Differences between the tertiles were tested using a Kruskal-Wallis test for continuous variables and χ2-test for categorical variables. P-values ≥ 0.2 with one decimal.

  5. b

    18–25 years; one measurement available per participant.

  6. c

    BMI ≥ 25 kg/m2.

  7. d

    Girls 9–14 years, boys 10–15 years; we used the first measurement available.

  8. FMI and FFMI were derived on the basis of skinfold thicknesses. Height, weight, and skinfold measurements were conducted by trained nurses.

Young adulthood characteristicsb         
Age (years)26218.2 (18.0, 22.1)22.0 (18.1, 22.7)21.2 (18.0, 22.1)0.219.1 (18.3, 22.4)19.0 (18.1, 21.9)19.0 (19.0, 22.5)0.5
Overweight, n (%)c26213 (32.5)15 (36.6)9 (22.0)0.35 (10.9)2 (4.3)10 (21.3)0.04
BMI (kg/m2)26223.1 (20.4, 25.4)23.4 (20.7, 26.1)23.2 (21.1, 25.0)0.821.5 (19.4, 22.9)20.7 (19.7, 23.4)22.4 (21.4, 24.8)0.001
FMI (kg/m2)2624.1 (2.7, 5.5)4.0 (2.9, 5.6)3.9 (3.0, 4.8)0.76.2 (5.0, 7.3)6.0 (4.8, 7.2)6.7 (6.1, 7.8)0.02
FFMI (kg/m2)26218.7 (17.3, 19.8)19.4 (18.3, 20.2)19.2 (18.3, 20.3)0.1815.3 (14.2, 16.1)15.1 (14.4, 16.2)15.9 (15.3, 17.3)0.001
Pubertal characteristicsd         
Age (years)26210.0 (10.0, 10.0)10.0 (10.0, 10.1)10.0 (10.0, 10.1)0.49.0 (9.0, 9.1)9.0 (9.0, 9.1)9.0 (9.0, 9.1)0.4
BMI (kg/m2)26216.2 (16.3, 18.3)17.4 (15.9, 18.2)17.7 (16.2, 18.8)0.1216.0 (15.1, 17.1)16.3 (14.9, 17.4)17.3 (15.6, 19.0)0.049
FMI (kg/m2)2622.1 (1.6, 3.7)2.6 (1.8, 3.4)2.5 (1.9, 3.5)0.32.7 (2.3, 3.3)2.7 (2.1, 3.6)3.0 (2.3, 4.4)0.3
FFMI (kg/m2)26213.9 (13.4, 14.6)14.5 (13.5, 15.1)15.0 (13.9, 15.7)0.0213.3 (12.6, 14.0)13.4 (12.8, 13.8)13.8 (13.0, 14.6)0.06
Early life characteristics         
Birth weight ≥ 3500g, n (%)26226 (65.0)23 (56.1)18 (43.9)0.1622 (47.8)17 (36.2)17 (36.2)0.4
Fully breastfed > 2 weeks, n (%)26230 (75.0)25 (61.0)32 (78.1)0.1932 (69.6)37 (78.7)29 (61.7)0.2
Family characteristics         
Maternal overweight, n (%)c2629 (22.5)14 (34.2)14 (34.2)0.415 (32.6)13 (27.7)19 (40.4)0.4
Smoking in the household, n (%)2428 (21.6)18 (48.7)11 (28.2)0.0414 (33.3)18 (41.9)13 (29.6)0.5
Mother ≥ 12 y schooling, n (%)26223 (57.5)10 (24.4)23 (56.1)0.00322 (47.8)25 (53.2)12 (25.5)0.02

Nutritional data at baseline by tertiles of animal protein intake are presented in Table 2. As expected, intakes of protein and calcium differed notably between tertiles in both genders. Similarly, carbohydrate nutrition was associated with animal protein, except for fiber intake levels. Of note, higher animal protein intakes were related to higher total fat and saturated fat intakes in females only.

Table 2. Nutritional data by energy-adjusted tertiles of animal protein intake during puberty
 Total nMales (n = 122)Females (n = 140)
Tertile 1 (n = 40)Tertile 2 (n = 40)Tertile 3 (n = 41)PaTertile 1 (n = 46)Tertile 2 (n = 47)Tertile 3 (n = 47)Pa
  1. DONALD Study, Germany

  2. Abbreviations: GI, glycemic index; GL, glycemic load. Values are medians (25th percentile, 75th percentile).

  3. a

    Differences between the tertiles were tested using a Kruskal-Wallis test for continuous variables and χ2-test for categorical variables. P-values ≥ 0.2 with one decimal.

  4. Intakes of meat protein and milk protein do not add up to total animal protein consumption as fish- and egg protein are missing.

Total energy (kcal)2622139 (1838, 2396)2086 (1945, 2343)2039 (1920, 2304)0.91738 (1565, 1887)1703 (1568, 1890)1711 (1584, 1902)0.95
Fat (% of energy)26235.7 (32.5, 38.0)36.2 (33.9, 38.4)35.9 (33.2, 38.1)0.634.7 (32.1, 37.0)35.4 (33.1, 37.1)37.6 (35.7, 39.9)0.0005
Saturated fatty acids (% of energy)26215.0 (13.1, 16.7)15.9 (14.9, 16.8)15.5 (14.2, 17.0)<.000115.6 (13.9, 16.6)15.4 (14.2, 16.3)16.4 (15.3, 17.8)0.01
Protein (% of energy)26212.0 (11.3, 12.6)13.3 (12.7, 13.7)14.6 (14.0, 15.4)<.000111.2 (10.9, 11.8)12.8 (12.2, 13.4)14.4 (13.4, 15.2)<.0001
Animal protein (% of energy)2627.0 (6.6, 7.5)8.4 (8.1, 8.8)10.0 (9.5, 10.6)<.00016.4 (5.9, 6.6)7.9 (7.7, 8.2)9.6 (9.0, 10.4)<.0001
Meat protein (% of energy)2622.6 (2.0, 3.3)3.3 (2.8, 3.9)4.3 (3.1, 5.5)<.00012.0 (1.3, 2.5)3.0 (2.3, 4.0)4.2 (3.0, 4.9)<.0001
Dairy protein (% of energy)2623.4 (2.5, 4.3)4.2 (3.7, 4.7)4.8 (3.8, 5.8)0.00013.4 (3.0, 3.9)3.8 (3.1, 4.5)4.6 (3.6, 5.3)<.0001
Vegetable protein (% of energy)2625.1 (4.6, 5.5)4.8 (4.2, 5.2)4.6 (4.1, 4.9)0.0095.0 (4.5, 5.5)4.8 (4.2, 5.3)4.6 (4.2, 5.1)0.03
Carbohydrate (% of energy)26253.3 (50.0, 55.5)50.0 (48.7, 52.9)49.2 (47.5, 52.7)0.00153.6 (51.9, 56.3)52.2 (49.7, 53.6)48.5 (45.4, 50.0)<.0001
Added sugar (% of energy)26214.9 (12.7, 19.9)14.4 (11.9, 17.6)12.3 (10.2, 15.3)0.0315.6 (12.4, 19.6)14.7 (11.2, 18.7)11.8 (10.0, 16.2)0.007
Dietary GI26257.0 (55.3, 58.6)56.5 (55.0, 57.7)55.0 (52.3, 56.9)0.00156.1 (55.4, 57.6)56.3 (54.3, 57.9)55.3 (53.8, 56.4)0.01
Dietary GL (g/ 1000 kcal26274.8 (72.1, 77.7)72.0 (67.7, 74.8)67.8 (65.1, 71.9)<.000175.3 (72.0, 79.9)73.9 (67.1, 77.1)66.5 (63.2, 70.9)<.0001
Dietary fiber (g/ 1000 kcal)26210.6 (9.1, 11.9)9.7 (8.6, 11.0)9.6 (8.1, 10.5)0.00710.9 (9.7, 12.1)10.2 (9.0, 12.3)10.1 (9.0, 11.6)0.15
Calcium (mg/ 1000 kcal)262420 (348, 488)487 (429, 524)526 (420, 591)0.0009415 (377, 466)472 (380, 521)501 (428, 575)<.0001

Pubertal protein intake and adult body composition

Among females, neither intake of animal protein nor intake of plant protein during puberty was related to FMI in young adulthood in our main multivariable analyses (Table 3. However, a higher intake of animal protein during puberty was associated with a higher FFMI in young adulthood (Table 4, pfor trend = 0.001). Substitution models revealed that consuming more energy from animal protein while consuming less energy from carbohydrates was significantly related to a higher FFMI among females (pfor trend = 0.01; adjustment for age, baseline FFMI, breastfeeding, maternal education status, calcium, fats, plant protein, and energy; data not shown). Concerning a substitution of animal protein for fats, we observed only a trend (pfor trend = 0.08; adjustment for age, baseline FFMI, breastfeeding, maternal education status, glycemic index, calcium, carbohydrates and plant protein; data not shown). In males, animal or plant protein intake during puberty was neither related to FMI nor to FFMI (Table 3. Further adjustment for a variable describing the presence/ absence of smokers in the family household (yes/ no) did not change our main results (within the subset of participants for which the variable was available, (n = 242)).

Table 3. Relation of dietary protein intake during puberty to fat mass index and fat-free mass index in young adulthood
 Fat mass index, FMI (kg/m2)Fat-free mass index, FFMI (kg/m2)
Tertile 1Tertile 2Tertile 3pfor trendTertile 1Tertile 2Tertile 3p for trend
  1. DONALD Study, Germany

  2. Values are least squares means and 95% confidence intervals.

  3. Tertiles: energy-adjusted tertiles of animal protein intake during puberty (baseline).

  4. p for trend: p for linear trend, calculated in multiple linear regression analyses.

  5. Model A: adjusted for the respective baseline value (FMI in FMI-models and FFMI in FFMI-models) and age in young adulthood.

  6. Model B: Model A + adjustment for early life factors (breast feeding), socioeconomic factors (maternal education status) and nutritional factors (dietary glycemic index, intakes of calcium and energy). Models for FMI were further adjusted for birth weight and maternal overweight, as well as for dietary fiber—except when considering plant protein in order to avoid multicollinearity.

  7. a

    Median intake levels of animal protein, as % of energy.

Animal protein        
Females (n = 140)6.4%a7.9%9.6% 6.4%7.9%9.6% 
Model A6.2 (5.8–6.7)6.1 (5.7–6.5)6.9 (6.4–7.4)0.415.3 (15.0–15.6)15.3 (15.0–15.6)15.9 (15.6–16.2)0.02
Model B6.3 (5.9–6.9)6.4 (5.9–6.9)7.2 (6.6–7.8)0.515.3 (15.0–15.5)15.4 (15.1–15.7)16.2 (15.9–16.6)0.001
Males (n = 120)7.0%8.4%10.0% 7.0%8.4%10.0% 
Model A4.2 (3.7–4.7)3.7 (3.3–4.2)3.6 (3.2–4.1)0.418.9 (18.5–19.3)19.1 (18.7–19.5)18.9 (18.5–19.3)0.2
Model B4.3 (3.7–5.0)3.7 (3.2–4.2)3.6 (3.1–4.1)0.1418.9 (18.4–19.3)19.0 (18.6–19.4)18.9 (18.5–19.4)0.3
Plant protein        
Females (n = 140)4.1%4.8%5.5% 4.1%4.8%5.5% 
Model A6.3 (5.8–6.7)6.5 (6.0–6.9)6.4 (6.0–6.9)0.415.6 (15.3–15.9)15.7 (15.4–16.0)15.2 (14.9–15.5)0.13
Model B6.4 (6.0–6.9)6.8 (6.3–7.3)6.7 (6.2–7.2)0.315.7 (15.3–16.0)15.8 (15.5–16.1)15.4 (15.1–15.7)0.3
Males (n = 122)4.1%4.8%5.5% 4.1%4.8%5.5% 
Model A3.9 (3.4–4.5)3.8 (3.4–4.4)3.7 (3.2–4.2)0.9919.0 (18.6–19.4)19.0 (18.6–19.4)18.9 (18.5–19.3)0.9
Model B3.8 (3.3–4.3)4.1 (3.6–4.7)3.6 (3.1–4.1)0.718.8 (18.4–19.2)19.2 (18.7–19.6)18.8 (18.4–19.2)0.97
Table 4. Relation of pubertal protein intake to young adult levels of FMI at fixed levels of FFMI, and vice versa
 Fat mass index, FMI (kg/m2)Fat-free mass index, FFMI (kg/m2)
Tertile 1Tertile 2Tertile 3pfor trendTertile 1Tertile 2Tertile 3pfor trend
  1. DONALD Study, Germany

  2. Values are least squares means and 95% confidence intervals.

  3. Tertiles: energy-adjusted tertiles of animal protein intake during puberty (baseline).

  4. p for trend: p for linear trend, calculated in multiple linear regression analyses.

  5. FMI-models: adjusted for FFMI in young adulthood, baseline- fat mass index, age in young adulthood, breast feeding, birth weight, maternal overweight, maternal education status, glycemic index, intakes of fiber, calcium and energy. When considering plant protein, there was no adjustment for fiber in order to avoid multicollinearity.

  6. FFMI-models: adjusted for FMI in young adulthood, baseline- fat-free mass index, age in young adulthood, breast feeding, maternal education status, glycemic index, intakes of calcium and energy.

  7. a

    Median intake levels of animal protein, as % of energy.

Animal protein        
Females (n = 140)6.4%a7.9%9.6% 6.4%7.9%9.6% 
 6.5 (6.1–7.0)6.5 (6.0–6.9)6.7 (6.2–7.2)0.315.3 (15.0–15.5)15.4 (15.2–15.7)16.0 (15.7–16.3)0.001
Males (n = 120)7.0%8.4%10.0% 7.0%8.4%10.0% 
 4.5 (4.0–5.1)3.6 (3.2–4.1)3.4 (3.0–3.8)0.00118.7 (18.3–19.0)19.0 (18.7–19.4)19.0 (18.7–19.4)0.04
Plant protein        
Females (n = 140)4.1%4.8%5.5% 4.1%4.8%5.5% 
 6.5 (6.1–7.0)6.5 (6.0–6.9)6.6 (6.2–7.1)0.215.6 (15.3–15.9)15.8 (15.5–16.0)15.3 (15.1–15.6)0.09
Males (n = 122)4.1%4.8%5.5% 4.1%4.8%5.5% 
 3.8 (3.4–4.3)4.0 (3.5–4.5)3.6 (3.2–4.1)0.918.9 (18.5–19.2)19.0 (18.6–19.4)18.9 (18.5–19.2)0.7

To acquire additional information on the relevant protein source driving the association between animal protein intake and FFMI in females, we investigated two major sources of animal protein: meat and dairy foods. A higher intake of protein from meat during puberty was related to a higher FFMI in young adulthood (adjusted means of FFMI within tertiles of meat protein intake: 15.4 (15.1-15.7), 15.5 (15.2-15.8), 16.0 (15.7-16.4) kg/m2; pfor trend = 0.002). There was no significant relation of dairy protein intake during puberty and FFMI in young adulthood (adjusted means of FFMI within tertiles of milk protein intake: 15.7 (15.4-16.0), 15.6 (15.3-16.0), 15.5 (15.2-15.8) kg/m2; pfor trend = 0.17).

Pubertal protein intake and FMI at fixed levels of FFMI and vice versa

Additional analyses in which we adjusted FMI for current FFMI and vice versa (Table 4 yielded similar results for the association between animal protein and FMI or FFMI among females. In men, on the other hand, a higher consumption of animal protein during puberty was now significantly related to a lower FMI (pfor trend = 0.001) and to a slightly higher FFMI in young adulthood (pfor trend =0.04). Plant protein remained unrelated to adult body composition among both sexes.

Protein intake during earlier phases of life and body composition in young adulthood

We observed no associations between protein intake during early childhood and body composition in young adulthood (all pfor trend - values ≥ 0.3). By contrast, dietary animal protein during the period of adiposity rebound tended to be related to FFMI in young adulthood among males (adjusted means of FFMI within tertiles of animal protein intake: 18.8 (18.1-19.5), 19.7 (19.0-20.4), 19.7 (19.0-20.5) kg/m2; pfor trend = 0.05), but not in females [adjusted means: 15.5 (14.9-16.1), 16.1 (15.6-16.6), 16.2 (15.6-16.8) kg/m2; pfor trend = 0.13]. Animal or plant protein in the period of adiposity rebound was not related to adult FMI.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgments
  9. References

Our data suggest a link between a higher pubertal animal protein consumption and a higher adult FFM, primarily among women. In terms of the driving force of this association, a substitution of animal protein for carbohydrates seemed to be slightly more relevant than a substitution for fats. Moreover, it seemed to be protein from meat sources that was responsible for the results observed concerning animal protein in women. However, as the variation of dairy protein intake in our female study sample was substantially smaller than that of meat protein intake, the interpretability of this finding is limited.

Young adult FMI levels were not related to pubertal plant or animal protein consumption in females. However, the difference in mean FMI between the extreme tertiles of animal protein intake was essentially the same as for FFMI (both 0.9 kg/m2; see Table 5). The width of the confidence intervals, on the other hand, was much bigger for FMI than for FFMI. Thus, insufficient statistical power resulting from the relatively high variance of FMI [34] and from our relatively small sample sizes must be considered as one cause for the absence of significant results concerning FMI.

In men, we only observed relations between animal protein intake and adult FFMI or FMI when holding levels of the respective complementary component of body composition (adult FMI, FFMI) constant. The emergence of significant associations after vice versa adjustment is probably because of the opposing associations (i.e. the fact that higher pubertal animal protein intakes tended to be related to lower adult FMI and higher adult FFMI levels) as well as the marked correlation of adult FMI and FFMI levels (r = 0.61 among males and r = 0.55 in females). The public health relevance of these findings is not straightforward, as holding one part of body mass constant makes it difficult to consider body composition as the entity that it is in reality. However, the results are of interest from a mechanistic point of view. In men, the difference in FMI between low and high pubertal animal protein consumption (absolute difference between the first and the third tertile, Table 4: 1.1 kg/ m2) was notably larger than the respective difference in FFMI (absolute difference, Table 4: 0.3 kg/ m2), indicating that among males variations in pubertal animal consumption may primarily affect adult fat mass. Of note, this contrasts to our findings for females, in whom a higher pubertal animal consumption appeared to primarily affect adult fat-free mass. It is imaginable that, in boys, hormonal influences on the construction of muscle mass are much more important than variations in protein consumption within the range of usual intakes.

We conducted additional analyses that revealed that besides puberty, the period of adiposity rebound, but not early childhood, may be relevant for the long-term development of body composition: Among males, we observed a positive relation between animal protein intake during the period of adiposity rebound and FFMI in young adulthood with borderline significance (p = 0.05). However, our analyses conducted for the early life and the adiposity rebound period are limited by small sample sizes and must hence be interpreted with caution.

One physiological mechanism by which higher intakes of animal protein could increase FFMI is an anabolic effect of essential amino acids (EAA) on muscle mass, which has been observed in small experimental studies among younger and older adults [25]. A suggested biochemical pathway for the stimulation of muscle protein synthesis by the EAA leucine is the activation of the protein kinase mammalian target of rapamycin (mTOR) and its downstream effectors eukaryotic initiation factor 4E (EIF4E) and ribosomal S-6 kinase (S6K1) [26]. Given the generally lower EAA content of plant protein compared to animal protein [27], our finding that plant protein intake was not associated with FFMI, neither in males nor in females, is quite plausible. Still, in order to verify that such a relation was not only masked because of a weak inverse correlation of plant protein and animal protein consumption in our sample (r = 0.27), we ran additional analyses in which we adjusted models examining plant protein intake for animal protein (data not shown). However, such an adjustment did not notably change our results.

RCTs conducted among a pubertal or young adult population provide evidence for a role of animal protein from different sources as stimulants of fat-free mass increase: An RCT with 6-14 year old Kenyan children showed an effect of a meat supplementation on mid-upper-arm muscle area, but not mid-upper-arm fat area [28]. In an RCT with 98 eight to ten year old Chilean girls, replacement of sugar-sweetened beverages for milk for 16 weeks yielded an additional gain in fat-free mass, but not fat mass, in comparison with the control group [29]. In three other intervention studies with pubertal or young adult subjects, a supplementation with milk products had no significant effect on body composition [30-32]; however, these studies were not specifically designed to investigate changes in body composition and had smaller sample sizes. Hence, our main finding of a prospective relation between animal protein during puberty and FFMI in young adulthood among women is generally consistent with evidence from RCTs.

The findings from other epidemiologic studies with pubertal or young adult study populations concerning a link between protein intake and fat mass are mixed. In a Danish cohort study with 350 participants, higher protein intakes in puberty were related to a higher percentage body fat in young adulthood among women [35], and in an American cohort study with 2909 participants, higher protein intakes in young adulthood were prospectively related to a slightly higher waist-to-hip-ratio among white participants [7]. On the other hand, in a Dutch epidemiologic study with 364 subjects, higher intakes of protein were prospectively associated with a lower FMI among leaner girls, and with a higher FFMI among girls in the 5th BMI-quintile [36].

In terms of public health relevance, our study does not suggest a strong unfavorable effect of relatively high animal protein intake levels on body composition. Even under the assumption that a larger sample size would render the results for FMI seen in women significant, the effect sizes seen in this study do not imply a disproportionally higher increase in FMI than in FFMI. On the other hand, even in the highest energy-adjusted tertile of animal protein intake, protein accounted for less than 15% of energy intake (see Table 2. Still higher intake levels could affect body composition in a different manner.

Limitations of our study include, first of all, its observational design. We cannot exclude that our results might be biased by residual confounding. Additionally, as observational studies need energy-adjustment to limit the impact of confounding and underreporting, potential satiety-mediated effects of protein [38] can be shown less well than in intervention studies under ad-libitum conditions.

Second, we determined FMI and FFMI on the basis of skinfold thickness measurements, which have a higher susceptibility to measurement error than specialized research methods such as hydrodensitometry, magnetic resonance imaging, and BodPod. Yet, the skinfold equations of Durnin and Womersley [19] agree, on average, very well with the results from hydrodensitometry [39]. In addition, measurements were performed by trained and quality-monitored personnel, which has been shown to notably reduce intra- and interobserver variability [40]. It would have been interesting to specifically consider abdominal fat mass as a body fat distribution characterized by high intra-abdominal fat is known to be particularly detrimental. However, in our sample, there was not enough data available to conduct such analyses. Third, DONALD participants are characterized by a relatively high socio-economic status [14] and only 20.6% of the participants study sample were overweight in young adulthood. It is possible that the relative homogeneity of our sample means that extremes of diet and behavior are not represented. Forth, we merely disposed a crude measure of physical activity at the age of five years derived from parental questionnaires, available for 198 participants of our study sample. When we repeated our analyses with further adjustment for physical activity in this subgroup, we obtained essentially the same results as in our main analyses. The main strengths of our study are its prospective nature and the carefully collected, repeated data on growth and diet, covering the entire time of childhood and adolescence. The availability of data on several potential confounders, such as parental characteristics, is an additional strength of our analysis.

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgments
  9. References

In conclusion, our results indicate that, particularly among females, a habitually higher consumption of animal protein during puberty yields a higher FFMI in young adulthood. This argues against a selective increase in fat mass as a consequence of relatively high intake levels of animal protein.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgments
  9. References

We thank the staff of the Research Institute of Child Nutrition for carrying out the anthropometric measurements and for collecting and coding the dietary records and all participants of the DONALD Study. The DONALD study is supported by the Ministry of Science and Research of North Rhine Westphalia, Germany, and this analysis was partially funded by the World Cancer Research Fund Netherlands (grant no. 2010/248).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgments
  9. References
  • 1
    Clifton PM, Keogh JB, Noakes M. Long-term effects of a high-protein weight-loss diet. Am J Clin Nutr 2008;87:2329.
  • 2
    Due A, Toubro S, Skov AR, Astrup A. Effect of normal-fat diets, either medium or high in protein, on body weight in overweight subjects: A randomised 1-year trial. Int J Obes 2004;28:12831290.
  • 3
    Lejeune MPGM, Kovacs EMR, Westerterp-Plantenga MS. Additional protein intake limits weight regain after weight loss in humans. Br J Nutr 2005;93:281289.
  • 4
    Larsen TM, Dalskov SM, van Baak M, et al. Diets with High or Low Protein Content and Glycemic Index for Weight-Loss Maintenance. N Engl J Med 2010;363:21022113.
  • 5
    Westerterp-Plantenga MS, Lejeune MPGM, Nijs I, van Ooijen M, Kovacs EMR. High protein intake sustains weight maintenance after body weight loss in humans. Int J Obes Relat Metab Disord 2004;28:5764.
  • 6
    Vergnaud A-C, Norat T, Romaguera D, et al. Meat consumption and prospective weight change in participants of the EPIC-PANACEA study. Am J Clin Nutr 2010;92:398407.
  • 7
    Ludwig DS, Pereira MA, Kroenke CH, et al. Dietary fiber, weight gain, and cardiovascular disease risk factors in young adults. JAMA 1999;282:15391546.
  • 8
    Bujnowski D, Xun P, Daviglus ML, Van Horn L, He K, Stamler J. Longitudinal association between animal and vegetable protein intake and obesity among men in the United States: The Chicago Western Electric Study. J Am Diet Assoc 2011;111:11501155.e1151.
  • 9
    Halkjaer J, Olsen A, Overvad K, et al. Intake of total, animal and plant protein and subsequent changes in weight or waist circumference in European men and women: The Diogenes project. Int J Obes 2011;35:11041113.
  • 10
    Bray GA, Smith SR, de Jonge L, et al. Effect of dietary protein content on weight gain, energy expenditure, and body composition during overeating a randomized controlled trial. JAMA 2012;307:4755.
  • 11
    Rogol AD, Roemmich JN, Clark PA. Growth at puberty. J Adolesc Health 2002;31:192200.
  • 12
    Gunther ALB, Remer T, Kroke A, Buyken AE. Early protein intake and later obesity risk: Which protein sources at which time points throughout infancy and childhood are important for body mass index and body fat percentage at 7 y of age? Am J Clin Nutr 2007;86:17651772.
  • 13
    Dietz WH. Critical periods in childhood for the development of obesity. Am J Clin Nutr 1994;59:955959.
  • 14
    Kroke A, Manz F, Kersting M, et al. The DONALD Study. History, current status and future perspectives. Eur J Nutr 2004;43:4554.
  • 15
    Sichert-Hellert W, Kersting M, Schoch G. Underreporting of energy intake in 1 to 18 year old German children and adolescents. Z Ernahrungswiss 1998;37:242251.
  • 16
    Goldberg GR, Black AE, Jebb SA, et al. Critical evaluation of energy intake data using fundamental principles of energy physiology: 1. Derivation of cut-off limits to identify under-recording. Eur J Clin Nutr 1991;45:569581.
  • 17
    World Health Organization. Measuring obesity-classification and description of anthropometric data. Report on a WHO Consultation of the Epidemiology of Obesity. Nutrition Unit Document, EUR/ICP/NUT 123. WHO: Copenhagen, 1989.
  • 18
    Slaughter MH, Lohman TG, Boileau RA, et al. Skinfold equations for estimation of body fatness in children and youth. Hum Biol 1988;60:709723.
  • 19
    Durnin JV, Womersley J. Body fat assessed from total body density and its estimation from skinfold thickness: Measurements on 481 men and women aged from 16 to 72 years. Br J Nutr 1974;32:7797.
  • 20
    Cole TJ, Fewtrell MS, Prentice A. The fallacy of using percentage body fat as a measure of adiposity. Am J Clin Nutr 2008;87:19591959.
  • 21
    Souci S, Fachmann W, Kraut H. Food Composition and Nutrition Tables, 7th edn. Medpharm Scientific Publishers: Stuttgart, Germany, 2008.
  • 22
    Bokhof B, Gunther ALB, Berg-Beckhoff G, Kroke A, Buyken AE. Validation of protein intake assessed from weighed dietary records against protein estimated from 24 h urine samples in children, adolescents and young adults participating in the Dortmund Nutritional and Longitudinally Designed (DONALD) Study. Public Health Nutr 2010;13:826834.
  • 23
    Maldonado G, Greenland S. Simulation study of confounder-selection strategies. Am J Epidemiol 1993;138:923936.
  • 24
    Hu FB, Stampfer MJ, Rimm E, et al. Dietary fat and coronary heart disease: A comparison of approaches for adjusting for total energy intake and modeling repeated dietary measurements. Am J Epidemiol 1999;149:531540.
  • 25
    Paddon-Jones D, Sheffield-Moore M, Zhang XJ, et al. Amino acid ingestion improves muscle protein synthesis in the young and elderly. Am J Physiol Endocrinol Metab 2004;286:E321E328.
  • 26
    Anthony JC, Anthony TG, Kimball SR, Jefferson LS. Signaling pathways involved in translational control of protein synthesis in skeletal muscle by leucine. J Nutr 2001;131:856S860S.
  • 27
    Michaelsen KF, Hoppe C, Roos N, et al. Choice of foods and ingredients for moderately malnourished children 6 months to 5 years of age. Food Nutr Bull 2009;30:S343404.
  • 28
    Neumann CG, Jiang L, Weiss RE, et al. Meat supplementation increases arm muscle area in Kenyan schoolchildren. Br J Nutr 2013;109:12301240.
  • 29
    Albala C, Ebbeling CB, Cifuentes M, Lera L, Bustos N, Ludwig DS. Effects of replacing the habitual consumption of sugar-sweetened beverages with milk in Chilean children. Am J Clin Nutr 2008;88:605611.
  • 30
    Merrilees MJ, Smart EJ, Gilchrist NL, et al. Effects of diary food supplements on bone mineral density in teenage girls. Eur J Nutr 2000;39:256262.
  • 31
    Chan GM, Hoffman K, McMurry M. Effects of dairy products on bone and body composition in pubertal girls. J Pediatr 1995;126:551556.
  • 32
    Cadogan J, Eastell R, Jones N, Barker ME. Milk intake and bone mineral acquisition in adolescent girls: Randomised, controlled intervention trial. BMJ 1997;315:12551260.
  • 33
    Rolland-Cachera MF, Deheeger M, Maillot M, Bellisle F. Early adiposity rebound: Causes and consequences for obesity in children and adults. Int J Obes (Lond) 2006;30(Suppl 4):S11S17.
  • 34
    Schutz Y, Kyle UU, Pichard C. Fat-free mass index and fat mass index percentiles in Caucasians aged 18–98 y. Int J Obes Relat Metab Disord 2002;26:953960.
  • 35
    Koppes LLJ, Boon N, Nooyens ACJ, van Mechelen W, Saris WHM. Macronutrient distribution over a period of 23 years in relation to energy intake and body fatness. Br J Nutr 2009;101:108115.
  • 36
    van Vught AJAH, Heitmann BL, Nieuwenhuizen AG, Veldhorst MAB, Brummer R-JM, Westerterp-Plantenga MS. Association between dietary protein and change in body composition among children (EYHS). Clin Nutr 2009;28:684688.
  • 37
    Sluijs I, Beulens JWJ, van der A DL, Spijkerman AMW, Grobbee DE, van der Schouw YT. Dietary Intake of Total, Animal, and Vegetable Protein and Risk of Type 2 Diabetes in the European Prospective Investigation into Cancer and Nutrition (EPIC)-NL Study. Diabetes Care 2010;33:4348.
  • 38
    Westerterp-Plantenga MS. The significance of protein in food intake and body weight regulation. Curr Opin Clin Nutr Metab Care 2003;6:635638.
  • 39
    Fogelholm M, Lichtenbelt WV. Comparison of body composition methods: A literature analysis. Eur J Clin Nutr 1997;51:495503.
  • 40
    Moreno LA, Joyanes M, Mesana MI, et al. Harmonization of anthropometric measurements for a multicenter nutrition survey in Spanish adolescents. Nutrition 2003;19:481486.