Increased protein intake augments kidney volume and function in healthy infants

Transcription

1 & 2011 International Society of Nephrology original article Increased protein intake augments kidney volume and function in healthy infants Joaquin Escribano 1, Veronica Luque 1, Natalia Ferre 1, Marta Zaragoza-Jordana 1, Veit Grote 2, Berthold Koletzko 2, Dariusz Gruszfeld 3, Piotr Socha 3, Elena Dain 4, Jean-Noel Van Hees 5, Elvira Verduci 6 and Ricardo Closa-Monasterolo 7, for the CHOP Study Group 8 1 Pediatrics Unit, Universitat Rovira i Virgili, Reus, Spain; 2 Division of Nutritional Medicine and Metabolism, Dr von Hauner Childrens Hospital, Ludwig Maximilians University of Munich, Germany; 3 Clinic of Pediatrics, Children s Memorial Health Institute, Warsaw, Poland; 4 Université Libre de Bruxelles, Brussels, Belgium; 5 CHC St Vincent, Liège-Rocourt, Belgium; 6 Department of Pediatrics, University of Milan, Milano, Italy and 7 Pediatrics Unit, Universitat Rovira i Virgili, Reus, Spain Protein intake has been directly associated with kidney growth and function in animal and human observational studies. Protein supply can vary widely during the first months of life, thus promoting different kidney growth patterns and possibly affecting kidney and cardiovascular health in the long term. To explore this further, we examined 601 healthy 6-month-old formula-fed infants who had been randomly assigned within the first 8 weeks of life to a 1-year program of formula with low-protein (LP) or high-protein (HP) contents and compared them with 204 breastfed (BF) infants. At 6 months, infants receiving the HP formula had significantly higher kidney volume (determined by ultrasonography) and ratios of kidney volume to body length and kidney volume to body surface area than did infants receiving the LP formula. BF infants did not differ from those receiving the LP formula in any of these parameters. Infants receiving the HP formula had significantly higher serum urea and urea to creatinine ratios than did LP formula and BF infants. Hence, in this European multicenter clinical trial, we found that a higher protein content of the infant formula increases kidney size at 6 months of life, whereas a lower protein supply achieves kidney size indistinguishable from that of healthy BF infants. The potential long-term effects of a higher early protein intake on long-term kidney function needs to be determined. Kidney International (2011) 79, ; doi: /ki ; published online 29 December 2010 KEYWORDS: critical period; dietary protein; infant formula; metabolic programming; randomized controlled trial Correspondence: Ricardo Closa-Monasterolo, Pediatrics Unit, Faculty of Medicine, Universitat Rovira i Virgili. C/Sant Llorenc 21, Reus 43201, Spain. ricardo.closa@urv.cat 8 See Appendix. Received 9 March 2010; revised 24 September 2010; accepted 28 October 2010; published online 29 December 2010 Nutritional exposures during sensitive periods of early development may program functions and structures of the organism. 1,2 In this scenario, there has been increasing attention on protein supply during gestation and infancy, as it has been postulated as one of the main early nutritional factors that may have a lifelong effect on obesity risk, 3 6 hypertension, 7 kidney structure, and thus possibly kidney function. 8 With regard to kidney programming, prenatal animal models have shown that poor nutrition during pregnancy is related to low nephron endowment; this may be a potential driver of hypertension and renal disease later in life. 8 There is some evidence suggesting that a similar mechanism may occur in humans. Both low birth weight and prematurity have been related to smaller kidneys 9 and lower nephron endowment, which in turn have been associated with adult risk of hypertension and kidney disease These findings support the hypothesis that the kidney may be programmed during gestation in humans (while nephrogenesis takes place). Moreover, it still remains unknown whether the critical window during which the kidney could be programmed extends to early postnatal life. As the kidney increases its size and functional capacity rapidly during the first months of life, 14 it seems plausible to hypothesize that a nutritional intervention in this period could exert long-term effects on cardiovascular and kidney health. A study in rats determined the effect of combined prenatal and postnatal protein restriction on adult blood pressure and kidney function, arriving at the conclusion that the postnatal nutritional environment has a key role in determining the outcomes of developmental programming. 15 An increase in protein intake has been directly associated with increased kidney growth and function in experimental animal studies, 16,17 and with increased kidney function in healthy humans. 18 In patients with chronic kidney disease, an increased protein intake can contribute to the progression of the renal disease. 19,20 Although protein requirements in infancy have been estimated by nitrogen balance, 21 there are different levels of proteins that an infant formula can supply. 22 The median protein content of an infant formula Kidney International (2011) 79,

2 original article J Escribano et al.: Protein intake and kidney size in healthy infants in Europe, while this study was being carried out, was B2.2 g/100 kcal (range ¼ g/100 kcal). 5 Therefore, the total protein supply can vary widely in infants during the first months of life, and possibly influence kidney growth and function. A study on healthy infants reported larger kidneys among those who were formula fed compared with those who were breastfed (BF). 23 This difference could be explained by the higher protein content present in infant formulas than in human milk. 24,25 These results would suggest that different protein intakes in the normal range during infancy could promote different growth patterns in the kidney; thus, if confirmed, protein intake during the first months of life could have long-term effects on kidney and cardiovascular health. Although animal models and human studies have contributed to the knowledge of the relationship between protein intake and renal growth and function, there are no published clinical trials in infants that provide further evidence for this developmental process. The aim of this study was to test whether a higher protein intake in healthy infants could modify kidney growth and function. RESULTS Study sample A total of 737 infants from 865 eligible subjects were included in the analysis; Figure 1 shows randomization, allocation, and follow-up of study participants. Diet, country, or gender had no effect on the dropout rate, and hence they were equally distributed by feeding groups. There were no differences between feeding groups in weight or length at birth. Dietary intakes The mean (s.d.) protein intake at 6 months of life was 1.94 (±0.46) g/kg per day among infants with low protein (LP) and 3.0 (±0.7) g/kg per day among infants with high protein (HP) (Po0.001). Effect of protein intake on body growth Babies in the HP group were heavier and had a higher body mass index z-score than did those in the LP and BF groups (Table 1). However, there was no difference in the length-forage z-score between the HP and the LP groups, whereas BF infants were significantly shorter. Effect of protein intake on kidney volume Absolute and relative kidney volumes were significantly larger in the HP group than in the LP group (Table 2), whereas no differences were found between the LP and the BF groups. Differences between infants with the HP and LP diets were generally greater for boys than for girls (Figure 2), which in most of the cases did not reach statistical significance. Effect of feeding on protein metabolism and kidney function The type of feeding modulated both protein catabolites and kidney function (Table 3). Infants fed the HP formula reached the highest values for urea in serum, urea/creatinine ratio (Figure 3), and urine osmolarity than did infants fed the LP formula. Infants in the HP group also presented the highest concentrations of urinary creatinine; they were significantly higher than those of BF infants, but did not reach statistical significance compared with infants in the LP group. We did not find any differences between LP and HP on estimated glomerular filtration rate (egfr), although a slight but significant correlation was found between protein intake and egfr (r ¼ 0.161, Po0.001). BF infants had significantly greater levels of egfr than did the formula-fed groups. When we stratified the BF group into two subgroups according to their current diet at 6 months (113 (60.75%), infants were still being BF and 73 (39.25%) infants had been switched to formula milk), we observed that the still-beingbreastfed group had even greater egfr levels (median ¼ (72.9, 108.9)) than did the HP (Po0.01) and LP groups (Po0.001), and infants who had been switched to formula milk (median ¼ 76.7 (62.58, ), Po0.05). The main results on differences in kidney volume, protein metabolites, and kidney function were very similar when analyzed by one-way ANOVA (analysis of variance) or Kruskal Wallis with post hoc comparisons (data not shown). Effect of growth parameters and dietary intake on kidney volume In the total sample, we found a positive correlation of kidney volume with protein intake (g/day) (r ¼ 0.213, P ¼ o0.001) and energy intake (r ¼ 0.124, Po0.01). Kidney volume also correlated with birth weight (r ¼ 0.214, Po0.001) and anthropometrical measures at 6 months, including weight (r ¼ 0.407, Po0.001), length (r ¼ 0.380, Po0.001), and body surface area (BSA) (r ¼ 0.429, Po0.001) (Figure 4). Linear regression analyses showed a significant effect of feeding a HP formula on kidney volume, even when gender, anthropometrical variables, egfr, and countries were included in the models (Table 4). Linear regression analyses showed that the country of origin influenced kidney volume, with Poland and Spain being those that had significantly higher volumes compared with Germany. The effect of country did not modify the effect of intervention. In a linear regression analysis performed using protein intakes obtained from food diaries, we estimated an increase in kidney volume for each additional gram in protein intake of cm 3 (after adjustments for recumbent length, body weight, egfr, gender, and country). Neither energy intake, breastfeeding (as compared with LP intake) nor birth weight showed any effect on kidney volume in linear regression models when the current anthropometrical variables were included. DISCUSSION Effect of protein intake on kidney size This is the first randomized controlled trial carried out in humans to demonstrate the effect of protein intake on kidney volume. Our results show that protein intake, independent of energy, affects kidney size in healthy infants. 784 Kidney International (2011) 79,

3 J Escribano et al.: Protein intake and kidney size in healthy infants original article Recruited at birth, n = 1263 Randomization Control group Lower protein Higher protein Breastfeeding (N = 419) (N = 430) (N = 414) Lost to follow-up, n = 43 Exclusion, n = 48 (Lack of compliance, illness/medication) Lost to follow-up, n = 36 Exclusion, n = 70 (Lack of compliance, illness/medication) Lost to follow-up, n = 102 Exclusion, n = 76 (Lack of compliance, breastfeeding dropout, illness/medication) Still in the study at 6 months, n = 865 Lower protein (N = 328) GE: 69, BE: 55, PO: 80, SP: 124 Higher protein (N = 324) GE: 72, BE: 46, PO: 79, SP: 127 Breastfeeding (N = 236) GE: 62, BE: 58, PO: 48, SP: 68 US not done, n = 29 US not done, n = 22 US not done, n = 32 Infants who did the kidney ultrasound, n = 805 Lower protein (N = 299) GE: 58, BE: 43, PO: 80, SP: 118 Higher protein (N = 302) GE: 60, BE: 42, PO: 79, SP: 121 Breastfeeding (N = 204) GE: 53, BE: 43, PO: 48, SP: 60 Excluded for kidney anomaly or measurement error, n = 23 Excluded for kidney anomaly or measurement error, n = 27 Excluded for kidney anomaly or measurement error, n = 18 Analyzed, n = 737 Lower protein (N = 276) (132 males) GE: 55, BE: 37, PO: 68, SP: 116 Higher protein (N = 275) (129 males) GE: 54, BE: 41, PO: 65, SP: 115 Breastfeeding (N = 186) (82 males) GE: 47, BE: 42, PO: 39, SP: 58 Figure 1 Randomization, allocation, and follow-up of study participants. BE, Belgium; GE, Germany; PO, Poland; SP, Spain; US, ultrasound. Increased protein intake has been previously related to increased kidney growth in animals and humans. 17,26 28 In infants, an observational study found that those who were BF had smaller kidneys than did those babies fed exclusively with an infant formula. 23 The mean difference in kidney volume between HP and the other groups (namely the BF and LP groups) was 10%. Differences by gender in kidney growth have been discussed in many previous publications, with contradictory results. 29,30 In our population, boys had larger kidneys than did girls in all of the feeding groups. However, linear regression analyses attributed a marginal effect of gender on kidney volume when anthropometrical measures and feeding variables were included in the models. On other hand, this study confirms a gender-dependent response of kidney growth to protein intake, as the effect of a HP diet is more pronounced in boys than in girls. This gender-dependent effect has been described in animal studies, 27 which have shown responses of lower intensity in kidney growth among females to different stimuli (such as a HP diet). 23,27 This effect has been previously suggested for 3-month-old infants, 23 and our data corroborate it at 6 months of life. Schmidt et al. 23 proposed a hormonal mechanism based on the coincidence with mini-puberty timing, suggesting the influence of sex steroids that could explain these findings. Kidney International (2011) 79,

4 original article J Escribano et al.: Protein intake and kidney size in healthy infants Table 1 Anthropometric measurements at 6 months High protein Low protein Breastfed Mean (s.d.) Mean (s.d.) Mean (s.d.) (n=275) (n=275) (n=185) Weight (kg) 7.84 (0.87) 7.67 (0.86)* 7.44 (0.88) z,a,b Length (cm) (2.25) (2.08) (2.59) w,a Body surface area (m 2 ) (0.026) (0.025)* (0.027) z,b Weight-for-age z-score 0.23 (0.88) 0.03 (0.92)* 0.23 (0.97) z,b Length-for-age z-score 0.39 (0.96) 0.33 (0.90) 0.12 (1.12) z,b BMI-for-age z-score 0.01 (0.95) 0.22 (1.01) w 0.39 (1.03) z Abbreviation: BMI, body mass index. P-values: *Po0.05, w Po0.01, and z Po0.001 vs high protein; a Po0.05, b Po0.01 vs low protein. Post hoc comparisons revealed similar differences between feeding groups, except for length-for-age z-score (did not differ between low-protein and breastfed infants) and body surface area (did not differ between high and low protein). Table 2 Kidney size parameters by gender and feeding type High protein Low protein Breastfed Mean (s.d.) Mean (s.d.) Mean (s.d.) (n=271) (n=272) (n=185) Right kidney volume (cm 3 ) 21.4 (5.2) 19.6 z (4.4) 19.0 z (4.5) Left kidney volume (cm 3 ) 21.2 (5.0) 19.9 w (4.7) 19.1 z (4.5) Total kidney volume (cm 3 ) 42.6 (9.6) 39.6 z (8.4) 38.0 z (8.4) Kidney volume/cm 0.63 (0.14) 0.59 z (0.12) 0.57 z (0.12) Kidney volume/kg 5.45 (1.12) 5.17 w (1.07) 5.12 w (1.01) Kidney volume/m (22.7) z (20.9) z (20.1) P-values: w Po0.01, and z Po0.001 vs high protein. Post hoc comparisons revealed similar differences between feeding groups. Kidney volume/body length (cm 3 /cm) b High protein Low protein Breastfed Figure 2 Effect of feeding type on kidney volume. Black symbols represent boys, white symbols represent girls. w Po0.01, z Po0.001 vs high protein; b P ¼ 0.01, c Po0.001 vs boys. Relationship between anthropometrical parameters and kidney size In this study, we demonstrated a direct effect of protein intake on body growth parameters, which in turn, have also been related to kidney size. As other studies have previously reported, 29,31 length, weight, and BSA have been directly correlated with kidney size. A child s length has been used to define normal kidney size ranges, 30,32,33 and with our linear regression models, we have found a similar effect of length and weight on kidney volume. Kidney growth has been directly correlated with birth weight in small-for-gestational age infants and animals Although we found a correlation between birth weight and kidney volume,c in our study, it could be due to a global effect of growth rather than a specific effect directly on the kidney. In fact, the effect of birth weight on kidney size disappears in linear regression models, including the current anthropometrical variables. Although body growth is closely related to kidney size, protein intake seems to induce a direct effect on kidney volume, independent of body growth. This increased proteininduced kidney growth could be mediated by insulin-like growth factor-1. Protein intakes have been directly correlated in animals 38 and human observational studies with insulinlike growth factor-1 levels, 39,40 which in turn, has shown to be related to kidney growth and function. 41 Effect of feeding on protein metabolism and kidney function The effect of a HP intake on kidney size has been attributed to increased renal workload, secondary to increased protein metabolite filtration (mainly urea), inducing hyperplasia, both in the glomeruli and in the renal tubules. 42 In animal and human studies, an increased function has been described as secondary to increased protein intake, which may be due to increased glomerular plasma flow and transglomerular pressure, produced by a functional overload. 42 However, this effect has not been clearly demonstrated in healthy subjects. In a study of healthy babies, modifications in kidney size by feeding type were identified, but it was not possible to demonstrate changes in plasma creatinine levels or the egfr. 23 In our series, we were unable to demonstrate significant differences between intervention groups for urinary creatinine or egfr. The use of isolated creatinine values, which are partially influenced by lean body mass, which is in turn modified by protein intake, could reduce precision in filtration rate estimates and therefore mask the possible effect of protein intake. 42 However, we have observed significantly increased renal workloads (as shown by the serum urea/creatinine ratios) among infants fed the HP content formula as compared with those fed the LP content formula. BF infants had the lowest renal overload (as shown by urine urea and urine osmolarity) but the highest egfr, which may indicate the action of other physiological mechanisms promoted by different compounds of human milk or a higher bioavailability of human milk proteins. Even so, egfr is directly correlated with kidney volume, suggesting a connection between all of these variables. The highest urea values in the plasma were found among infants fed a higher protein content formula; this may be due to the renal workload instead of a diminished kidney function, as shown by the urea/creatinine quotient and the increased urinary osmolarity in this group. Protein-induced kidney growth: possible mechanisms to program kidney Therefore, the effect of protein intake on kidney size may be produced by two mechanisms: (1) Kidney growth increases as a compensatory mechanism to allow excretion of a larger renal molar load. (2) Higher protein intakes increase global body growth (probably through insulin-like growth factor-1), which may in turn be accompanied by an increased renal size. 786 Kidney International (2011) 79,

5 J Escribano et al.: Protein intake and kidney size in healthy infants original article Table 3 Effect of feeding type on protein catabolites and kidney function Variables High protein Median (IQR) Low protein Median (IQR) Breastfed Median (IQR) Serum parameters n=252 n=247 n=162 Serum urea (mg/dl) 29 (20, 36) 18 (13, 21) z 11 (8, 16)) z,a Serum creatinine (mg/dl) 0.40 (0.30, 0.47) 0.40 (0.30, 0.49) 0.36 (0.28, 0.45)*,b egfr (ml/min per 1.73 m 2 ) 76.0 (64.3, 103.5) 74.6 (62.0, 101.1) 83.0 (67.4, 109.4) b Serum urea/creatinine 73.2 (50.0, 109.3) 42.6 (31.4, 60.0) z 32.6 (22.2, 50.0) z,a Urinary parameters n=184 n=195 n=110 Urinary osmolarity (mmol/l) 361 (244, 631) 261 (173, 406) z 125 (88, 206) z,a Urinary creatinine (mg/dl) 18.3 (11.1, 36.5) 18.6 (11, 30.3) 12.4 (8.65, 21.1) z,b Abbreviations: egfr, estimated glomerular filtration rate; IQR: interquartile ranges. Mann Whitney U-test comparisons: *Po0.05 and z Po0.001 vs high protein; a Po0.001 and b Po0.01 vs low protein. Post hoc comparisons revealed similar significant differences, except for serum creatinine, which did not differ between high-protein and breastfed infants. 200 Serum urea/creatinine ratio (mg/mg) High protein Low protein Breastfed Figure 3 Effect of feeding type on renal workload. z Po0.001 vs high protein, c Po0.001 vs low protein. We hypothesize that kidney growth as a compensatory mechanism could disappear after ceasing the increased protein intake (as suggested by studies on the progression of renal disease linked to protein intake, see Lentine and Wrone 43 ). Schmidt et al. 44 found in an observational study that increased kidney volume by feeding infant formula was a reversible process. They found an increased kidney size among 3-month-old formula-fed infants, compared with BF infants, which was not accompanied by differences in weight or body mass index. This different compensatory kidney growth was a reversible effect, as they did not find any difference in kidney size when infants were 18 months old. However, increased body growth induced by protein intake has been shown to have a possible programming effect, as higher protein content of the infant formula during the first year of life was associated with higher weight and body mass index at 2 years, by this randomized clinical trial. 5 Therefore, there may be a permanent effect of protein intake on kidney growth, accompanied by an increase in weight, which may be mediated by the insulin-like growth factor-1 axis. Thus, protein intake would stimulate insulin-like growth factor-1 secretion, which would induce an increased body weight that may, in turn, be accompanied by an increase in the size of its organs, such as the kidney.,c Total kidney volume (cm 3 ) Body surface area (m 2 ) Figure 4 Relationship between body surface area and kidney volume. Pearson s correlation (r ¼ 0.429, Po0.001). Regression adjusted by country. It has been suggested that kidney size and function may be programmed in humans only during the second half of gestation, while nephrogenesis takes place. 8 Our study would provide a new insight into a possible programming effect during postnatal kidney maturation that animal models suggest. 45 Protein supply in early infancy and later consequences It should be stressed that protein supply by infant formulas within the current normatives (and similar to those formulas that were on the market while the study took place) can visibly modify kidney structure. Such an effect in this sensitive life period may have long-term health consequences. The total protein intake in our study population provided by these formulas was higher than the requirements published by the WHO (World Health Organization). 21 This suggests a need to revise the current normative for infant formulas, which could have implications for public health, given that protein supply during the first months of life has been related to the risk of obesity, poor cardiovascular health, and possibly kidney function later in life. Kidney International (2011) 79,

6 original article J Escribano et al.: Protein intake and kidney size in healthy infants Table 4 Effect of different variables on kidney volume at 6 months of age Variables with effect on kidney volume (cm 3 ) n Estimate (Upper, lower) 95% CI P-value R 2 Formula (high protein (0.90, 3.68) o vs low protein) Recumbent length (cm) 0.78 (0.34, 1.21) o0.001 Body weight (500 g) 1.48 (0.94, 2.01) o0.001 egfr (ml/min per 1.73 m 2 ) 0.05 (0.02, 0.07) o0.05 Gender 0.29 ( 1.79, 1.21) NS Abbreviations: CI, confidence interval; egfr, estimated glomerular filtration rate. The long-term consequences that increased kidney growth could have on kidney function and blood pressure still remain unknown. In small-for-gestational age infants, a larger kidney size predicts lower cardiovascular risk As breastfeeding is considered the gold standard, and has been identified as a major long-term protective factor, 46,47 we cannot speculate whether a HP formula (and therefore a large kidney size) may lead to a lower cardiovascular risk in healthy-term neonates because other nutritional and metabolic factors are involved. Further research is required to elucidate long-term health outcomes of early kidney development in healthy infants. The long-term follow-up of these infants may provide some solutions to these unanswered questions. In conclusion, a HP content in infant formula induces larger renal volume than in BF infants and in infants fed a formula with LP content. The long-term consequences of this altered renal growth remains to be elucidated. MATERIALS AND METHODS Design This double-blind, randomized intervention trial compared infants fed during the first year of life with a HP content formula (infant formula: 2.05 g/100 ml; follow-on formula: 3.2 g/100 ml) and infants fed a LP content formula (infant formula: 1.25 g/100 ml; follow-on formula: 1.6 g/100 ml). Detailed information about nutrient composition of study formulas was published elsewhere. 5 Families changed from infant formula (formula from birth to 4 6 months) to follow-on formula (formula from 4 6 months to 12 months) according to their pediatricians advice, and did not receive any further nutritional intervention from the study team. The composition of study formulas complied with the 1991 EU Directive on Infant and Follow-on Formulae, 48 and the energy contents of HP and LP formulas were indistinct. As a baseline reference, an observational control group of BF infants (BF at least during the first 3 months of life) was additionally studied in parallel for comparison. The results of this BF infants group are shown as a gold standard reference. This trial was registered as the European Childhood Obesity Project. Study sample The EU Childhood Obesity Project was conducted in five European countries: Germany, Belgium, Italy, Poland, and Spain, but the Italian team was not able to carry out the kidney study at their study center. Figure 1 shows the recruitment, randomization, and follow-up procedures for study participants included in the evaluation of kidney size. Our target population was a sample of month-old babies (±6 days) from 4 European countries that had been recruited at birth to participate in the EU Childhood Obesity Project. Those infants were randomly allocated to two different groups of formula milk (HP or LP) or to the BF observational control group. Main outcome measures Anthropometry. The nude weight and length of the infant were determined using a SECA 336 baby scale (SECA, Hamburg, Germany) (precision: ±10 g) and a SECA 232 stadiometer (precision: ±1 mm), respectively, at inclusion (median age 14 days) and at 6 months. We calculated body mass index (body mass index ¼ weight (kg)/length (m 2 )) and BSA (BSA (m 2 ) ¼ ((weight (kg) length (cm))/3600)) 2 (ref. 49). Weight, length, and body mass index are expressed as z- scores relative to the WHO for BF babies. 50 Blood and urine sampling and analysis. At 6 months of age, a venous blood sample was drawn from a vein access and collected in a K 2 EDTA-containing tube. A spontaneous urine sample was collected using an adhesive urine collection bag. For ethical reasons, blood and urine were not collected under fasted conditions. Serum creatinine (mg/dl) and serum urea nitrogen were measured locally with the Jaffe method, 51,52 using a kinetic colorimetric assay and a kinetic ultraviolet assay, respectively. 53 Urine and serum samples were stored at 701C and transported in dry ice to a central laboratory (Diagnostic Laboratory, The Children s Memorial Health Institute, Warsaw, Poland) for analysis. Urinary osmolarity (mmol/l) was measured using an osmometer. Urinary creatinine (mg/dl) was analyzed using the kinetic Jaffe s reaction in an automated ADVIA 1650/Mega Bayer R (Leverkusen, Germany). The egfr (ml/min per 1.73 m 2 ) was calculated using the Schwartz equation (egfr ¼ (0.45 length (cm))/(serum creatinine) (mg/dl)). 54 We also calculated the ratio: serum urea/serum creatinine. Ultrasonography. Ultrasonographic kidney measurements were carried out by 17 trained and blinded examiners using a linear or sector ultrasound (5 7.5 MHz) with a posterior approach in the prone position or a lateral approach in the prone position to measure the longest length possible. 55 Any infant with a kidney development anomaly or a urinary tract disease (such as congenital kidney disease, hydronephrosis, or asymmetry defined as 415% length difference between the kidneys) was excluded from this part of the study, because such disorders may affect kidney size. Length, width, maximum depth in the longitudinal section (D1), and maximum depth in the transverse section (perpendicular to the hilar region) (D2) of both the kidneys were measured (cm). Kidney volume (thereafter also kidney size) 788 Kidney International (2011) 79,

7 J Escribano et al.: Protein intake and kidney size in healthy infants original article was assessed by the equation of an ellipsoid (kidney volume (cm 3 ) ¼ length width 0.5 (D1 þ D2) 0.523) (ref. 30). Total kidney volume (cm 3 ) was calculated as the sum of the right and left kidney volumes. The relative kidney volume was calculated as kidney volume/ body weight (kidney volume/kg) (cm 3 /kg), kidney volume/bsa (kidney volume/m 2 ) (cm 3 /m 2 ) or kidney volume/body length (kidney volume/cm) (cm 3 /cm). Dietary intake. Protein (g/day per kg body weight) and energy (kcal/day) intakes were assessed by 3-day food diaries for formulafed babies. Human milk from BF babies was not analyzed. Standard operating procedures were developed to carefully collect infant formula preparation and intake, as well as weaning foods. All the diaries were checked and entered into the databases by trained nutritionists. The food compositions of all of the different test countries were introduced into the food database. Dietary intakes were calculated using dedicated software. 56 Statistics This study followed the recommendations made in the CONsolidated Standards of Reporting Trials guidelines. 57 Data management and statistical analyses were carried out using the software packages SAS version 9.2 (SAS Institute, Cary, NC) and Stata version 9.2 (StataCorp LP, College Station, TX). Z-scores for anthropometry variables were calculated using WHO Anthro software for personal computer. Descriptive results were expressed as mean (±s.d) or median and interquartile ranges after visually assessing the normal distribution of the studied variables. T-tests and Mann Whitney U-tests have been applied for statistical comparisons between feeding groups and genders. To take into account the probability of obtaining significant differences between feeding groups for repetition, we have performed one-way ANOVA and Kruskal Wallis tests with post hoc comparisons. Pearson s correlations were used to test for linear associations among continuous variables. Linear regression analyses were applied to assess the effect of feeding type (formula HP vs LP), current anthropometrical parameters, and renal workload on kidney volume, adjusting by potential confounders, such as gender and country. Models that quantify the effect of protein intake (g/day) on kidney volume were also applied, adjusting by the same parameters. Statistical significance was accepted at Po0.05. The study was performed according to the Declaration of Helsinki II Principles and approved by the local ethics committees. Written consent was obtained from all families. DISCLOSURE All the authors declared no competing interests. ACKNOWLEDGMENTS We gratefully acknowledge the families taking part in this study and the physicians and nurses who cooperated in the recruitment, blood sampling, and ultrasonography of participants. We thank the CHOP study group for their effort and dedication to this work and Fabio Confalionieri for his support with the food data entry. The studies reported herein have been carried out with partial financial support from the Commission of the European Communities, specific RTD Programme Quality of Life and Management of Living Resources, within the 5th Framework Programme, research grant nos QLRT and QLK1-CT , and the 6th Framework Programme, contract no This manuscript does not necessarily reflect the views of the Commission and in no way anticipates the future policy in this area. REFERENCES 1. Lucas A. Programming by early nutrition: an experimental approach. J Nutr 1998; 128(Suppl 2): S401 S Koletzko B. Early nutrition and its later consequences: new opportunities. Adv Exp Med Biol 2005; 569: Metges CC. Does dietary protein in early life affect the development of adiposity in mammals? J Nutr 2001; 131: Koletzko B, Broekaert I, Demmelmair H et al. Protein intake in the first year of life: a risk factor for later obesity? The E.U. childhood obesity Project. Adv Exp Med Biol 2005; 569: Koletzko B, von Kries R, Closa R et al. Lower protein in infant formula is associated with lower weight up to age 2 y: a randomized clinical trial. Am J Clin Nutr 2009; 89: Metges CC. Long-term effects of pre- and postnatal exposure to low and high dietary protein levels. Evidence from epidemiological studies and controlled animal experiments. Adv Exp Med Biol 2005; 569: Bagby SP. Maternal nutrition, low nephron number, and hypertension in later life: pathways of nutritional programming. J Nutr 2007; 137: Langley-Evans SC. Nutritional programming of disease: unravelling the mechanism. J Anat 2009; 215: Drougia A, Giapros V, Hotoura E et al. The effects of gestational age and growth restriction on compensatory kidney growth. Nephrol Dial Transplant 2009; 24: Alexander BT. Fetal programming of hypertension. Am J Physiol Regul Integr Comp Physiol 2006; 290: R1 R Zandi-Nejad K, Luyckx VA, Brenner BM. Adult hypertension and kidney disease: the role of fetal programming. Hypertension 2006; 47: Luyckx VA, Brenner BM. Low birth weight, nephron number, and kidney disease. Kidney Int Suppl 2005; 97: S68 S Moritz KM, Singh RR, Probyn ME et al. Developmental programming of a reduced nephron endowment: more than just a baby 0 s birth weight. Am J Physiol Renal 2009; 296: F1 F Rhodin MM, Anderson BJ, Peters AM et al. Human renal function maturation: a quantitative description using weight and postmenstrual age. Pediatr Nephrol 2009; 24: Hoppe CC, Evans RG, Moritz KM et al. Combined prenatal and postnatal protein restriction influences adult kidney structure, function, and arterial pressure. Am J Physiol Regul Integr Comp Physiol 2007; 292: R462 R Jakobsson B, Celsi G, Lindblad BS et al. A. Influence of different protein intake on renal growth in young rats. Acta Paediatr Scand 1987; 76: Hammond KA, Janes DN. The effects of increased protein intake on kidney size and function. J Exp Biol 1998; 201: Chan AY, Cheng ML, Keil LC et al. Functional response of healthy and diseased glomeruli to a large, protein rich meal. J Clin Invest 1988; 81: Brenner BM, Meyer TW, Hostetter TH. Dietary protein intake and the progressive nature of kidney disease: the role of hemodynamically mediated glomerular injury in the pathogenesis of progressive glomerular sclerosis in aging, renal ablation, and intrinsic renal disease. N Engl J Med 1982; 307: Klahr S. Effects of protein intake on the progression of renal disease. Annu Rev Nutr 1989; 9: Report of a JointWHO/FAO/UNU Expert Consultation. Protein and Amino Acid Requirements in Human Nutrition. World Health Organ Tech. Rep. Ser. [935]. WHO Technical Report Series Koletzko B, Baker S, Cleghorn G et al. Global standard for the composition of infant formula: recommendations of an ESPGHAN coordinated international expert group. J Pediatr Gastroenterol Nutr 2005; 41: Schmidt IM, Damgaard IN, Boisen KA et al. Increased kidney growth in formula-fed versus breast-fed healthy infants. Pediatr Nephrol 2004; 19: Martin Martinez B. Estudio comparativo de la leche de la leche de mujer con las leches artificiales [Comparison study between human milk and infant formulas]. An Pediatr Monogr 2005; 3: Heinig MJ, Nommsen LA, Peerson JM et al. Energy and protein intakes of breast-fed and formula-fed infants during the first year of life and their association with growth velocity: the DARLING Study. Am J Clin Nutr 1993; 58: Murray BM, Campos SP, Schoenl M et al. Effect of dietary-protein intake on renal growth possible role of insulin-like growth factor-i. J Lab Clin Med 1993; 122: Kidney International (2011) 79,

8 original article J Escribano et al.: Protein intake and kidney size in healthy infants 27. Murray BM, Brown GP, Schoenl M. Interaction of gender and dietary protein on renal growth and the renal growth hormone-insulin-like growth factor axis. J Lab Clin Med 1998; 131: Skov AR, Toubro S, Bulow J et al. Changes in renal function during weight loss induced by high vs low-protein low-fat diets in overweight subjects. Int J Obes Relat Metab Disord 1999; 23: Schmidt IM, Molgaard C, Main KM et al. Effect of gender and lean body mass on kidney size in healthy 10-year-old children. Pediatr Nephrol 2001; 16: Dinkel E, Ertel M, Dittrich M et al. Kidney size in childhood. Sonographical growth charts for kidney length and volume. Pediatr Radiol 1985; 15: Lane PH, Belsha CW, Plummer J et al. Relationship of renal size, body size, and blood pressure in children. Pediatr Nephrol 1998; 12: Gavela T, Sanchez BM, Gomez MG et al. [Ecographic study of kidney size in children]. Nefrologia 2006; 26: Peters H, Weitzel D, Humburg C et al. Sonographic determination of the normal kidney volume in newborn infants and infants. Ultraschall Med 1986; 7: Rakow A, Johansson S, Legnevall L et al. Renal volume and function in school-age children born preterm or small for gestational age. Pediatr Nephrol 2008; 23: Brion MJ, Leary SD, Lawlor DA et al. Modifiable maternal exposures and offspring blood pressure: a review of epidemiological studies of maternal age, diet, and smoking. Pediatr Res 2008; 63: Matsuoka OT, Shibao S, Leone CR. Blood pressure and kidney size in term newborns with intrauterine growth restriction. Sao Paulo Med J 2007; 125: Schmidt IM, Chellakooty M, Boisen KA et al. Impaired kidney growth in low-birth-weight children: distinct effects of maturity and weight for gestational age. Kidney Int 2005; 68: Dardevet D, Manin M, Balage M et al. Influence of low- and high-protein diets on insulin and insulin-like growth factor-1 binding to skeletal muscle and liver in the growing rat. Br J Nutr 1991; 65: Smith WJ, Underwood LE, Clemmons DR. Effects of caloric or protein restriction on insulin-like growth factor- I (IGF-I) and IGF-binding proteins in children and adults. J Clin Endocrinol Metab 1995; 80: Norat T, Dossus L, Rinaldi S et al. Diet, serum insulin-like growth factor-i and IGF-binding protein-3 in European women. Eur J Clin Nutr 2007; 61: Rabkin R, Schaefer F. New concepts: growth hormone, insulin-like growth factor-i and the kidney. Growth Horm IGF Res 2004; 14: King AJ, Levey AS. Dietary protein and renal function. J Am Soc Nephrol 1993; 3: Lentine K, Wrone EM. New insights into protein intake and progression of renal disease. Curr Opin Nephrol Hypertens 2004; 13: Schmidt IM, Main KM, Damgaard IN et al. Kidney growth in 717 healthy children aged 0-18 months: a longitudinal cohort study. Pediatr Nephrol 2004; 19: Boubred F, Buffat C, Feuerstein JM et al. Effects of early postnatal hypernutrition on nephron number and long-term renal function and structure in rats. Am J Physiol Renal Physiol 2007; 293: F1944 F Singhal A, Cole TJ, Lucas A. Early nutrition in preterm infants and later blood pressure: two cohorts after randomised trials. Lancet 2001; 357: Taittonen L, Nuutinen M, Turtinen J et al. Prenatal and postnatal factors in predicting later blood pressure among children: cardiovascular risk in young Finns. Pediatr Res 1996; 40: European Commission. Commission Directive 91/321/EEC of 14 May 1991 on infant and follow-on formulae. Official Journal of the European Commission 1991; 175: Mosteller RD. Simplified calculation of body-surface area. N Engl J Med 1987; 317: WHO Multicentre Growth Reference Study Group. WHO Child Growth Standards based on length/height, weight and age. Acta Paediatr Suppl 2006; 450: S76 S Jaffe MZ. [About the precipitation caused by pikrinic acid in normal urine and about a new reaction of creatinine]. Physiol Chem 1886; 10: Bartels H, Cikes M. [Chromogens in the creatinine determination of Jaffe]. Clin Chim Acta 1969; 26: Whelton A. Nitrogen metabolites and renal function. In: Burtis CA, Ashwood ER (eds). Tietz Textbook of Clinical Chemistry. 2nd edn. WB Saunders Company: Philadelphia, Schwartz GJ, Brion LP, Spitzer A. The use of plasma creatinine concentration for estimating glomerular filtration rate in infants, children, and adolescents. Pediatr Clin North Am 1987; 34: Nakamura M, Itoh K. Renal length changes with position. Pediatr Radiol 1998; 28: Schiess S, Grote V, Scaglioni S et al. Introduction of complementary feeding in 5 European countries. J Pediatr Gastroenterol Nutr 2010; 50: Moher D, Schulz KF, Altman D. The CONSORT Statement: revised recommendations for improving the quality of reports of parallel-group randomized trials Explore (NY) 2005; 1: Appendix CHOP Study Group: Georgina Mendez (Pediatrics Unit, Universitat Rovira i Virgili, Reus, Spain), Jeannette Beyer, Hans Demmelmair, Michaeal Fritsch, Gudrun Haile, Uschi Handel, Iris Hannibal, Susanne Kreichauf, Ingrid Pawellek, Sonia Schiess, Sabine Verwied-Jorky, Rüdiger von Kries (Division of Pediatric Epidemiology, Institute of Social Pediatrics and Adolescent Medicine and Division of Nutritional Medicine and Metabolism, Dr von Hauner Children s Hospital, Ludwig Maximilians University of Munich), Anna Dobrzañska, Roman Janas, Agnieszka Kowalik, Ewa Pietraszek, Anna Stolarczyk, Jerzy Socha (Children s Memorial Health Institute, Warsaw, Poland), Clotilde Carlier, Philippe Goyens, Joana Hoyos, Pascale Poncelet (ULB Bruxelles), Jean- Paul Langhendries, Franc oise Martin, Annick Xhonneux (CHC St Vincent Liege, Belgium), Carlo Agostoni, Marcello Giovannini, Silvia Scaglioni, Sabrina Tedeschi, Fiammetta Vecchi (University of Milan). 790 Kidney International (2011) 79,