Growth and Final Height Among Children With Phenylketonuria
BACKGROUND AND OBJECTIVES: Growth is an important criterion to evaluate health in childhood and adolescence, especially in patients depending on special dietary treatment. Phenylketonuria (PKU) is the most common inherited disease of amino acid metabolism. Patients with PKU depend on a special phenylalanine-restricted diet, low in natural protein. The study aimed to evaluate growth, growth rate, and target height in 224 patients with PKU.
METHODS: Retrospective, longitudinal analysis of standardized, yearly measurements of height, weight, and calculated growth rate (SD score [SDS]) of patients with PKU aged 0 to 18 years were conducted by using the national computerized CrescNet database. Inclusion was restricted to patients carried to term with a confirmed diagnosis of PKU or mild hyperphenylalaninemia determined by newborn screening and early treatment initiation.
RESULTS: From birth to adulthood, patients with PKU were significantly shorter than healthy German children (height SDS at 18 years: −0.882 ± 0.108, P < .001). They missed their target height by 3 cm by adulthood (women: P = .02) and 5 cm (men: P = .01). In patients receiving casein hydrolysate during childhood, this was more pronounced compared with patients receiving amino acid mixtures (P < .001). Growth rate was significantly reduced during their first 2 years of life and in puberty (growth rate SDS: −1.1 to −0.5 m/year, P < .001 and −0.5; P < .02).
CONCLUSIONS: Early diagnosed, treated, and continuously monitored patients with PKU showed reduced height from birth onward. During the last 2 decades, this phenomenon attenuated, probably because of advances in PKU therapy related to protein supplements and special low-protein foods.
- AAM —
- amino acid mixture
- MHPA —
- mild hyperphenylalaninemia
- Phe —
- PKU —
- SDS —
- SD score
What’s Known on This Subject:
There is only limited knowledge on growth patterns in phenylketonuria (PKU). Data are heterogeneous, mostly indicating growth deficits and higher rate of overweight and obesity.
What This Study Adds:
This study is the first investigation of a large, regional cohort of patients with PKU who were diagnosed early, continuously monitored, and showed reduced height from birth, resulting in reduced final height. Patients on casein hydrolysate were more affected than those on amino acid mixtures.
Anthropometric measures are important for evaluating the somatic development and nutritional status of children and adolescents. Human linear growth can be divided into partly overlapping growth phases: fetal growth, infancy, childhood, and puberty.1 Dependent on the respective phase, growth is influenced by varying genetic factors and the endocrine system, as well as nutrition.1 General growth retardation may be caused by chronic diseases and malnutrition.2 Therefore, patients with specific nutrition requirements, markedly different from normal nutrition, represent a high-risk group for growth disturbances. This is especially true in inherited metabolic diseases requiring a strict dietary treatment. There are limited data about physical development in these patients, primarily because of the often low incidence of these diseases. Phenylketonuria (PKU; Online Mendelian Inheritance in Man database identifier 261600) is the most common inherited disease of amino acid metabolism, with an incidence of ∼1 in every 8000 across Europe.3 PKU was the first specifically treatable inherited metabolic disease.4 It was also the first metabolic disorder for which routine newborn screening was developed and introduced.5 Thus, PKU became a research model for therapeutic strategies and long-term outcomes in inherited metabolic diseases. PKU is caused by mutations in the gene encoding phenylalanine (Phe) hydroxylase (ENZYME database identifier EC 188.8.131.52), leading to deficient enzyme activity and high Phe concentrations in blood and tissues. Untreated patients suffer from irreversible psychomotor retardation. Newborn screening and prompt initiation of dietary treatment nowadays allow for normal neurocognitive development. According to their residual Phe hydroxylase activity and individual Phe tolerance, patients with PKU depend on a special Phe-restricted diet, which is low in natural protein and has substantial limitations on food choice. Hence, supplementation of amino acids and micronutrients by Phe-free amino acid mixtures (AAMs) is indispensable.6 Although well-established, the dietary treatment represents a semisynthetic diet with potentially adverse effects on patients’ development and growth. Insufficient micronutrient intake, micronutrient deficiencies, osteopenia, and osteoporosis were detected, even in patients with PKU who were diagnosed early and continuously treated.7–13 Several studies have revealed growth and weight impairments.10,14–18 Available data, however, are heterogeneous because of different study designs, patient cohorts, and treatment guidelines. With respect to height and weight development, anthropogenic and cultural influences must also be considered. So far, there are no conclusive data on longitudinal growth, growth rate, or target height in PKU.
The aim of this study was to conduct an evaluation of growth, including the height and weight development as well as growth rate, of a large, homogeneous group of patients with PKU. Data were analyzed by using a computerized database, CrescNet, a network for the continuous long-term monitoring of the growth and weight development of German children. We hypothesized that a PKU diet may influence growth rates and the adult heights of patients with PKU.
We performed a retrospective, longitudinal analysis of standardized, yearly measurements of the weight and height as well as calculated growth rates of patients with PKU and mild hyperphenylalaninemia (MHPA) from 2 German pediatric centers for inherited metabolic diseases (Leipzig and Magdeburg). The study was approved by the University of Leipzig’s ethics committee (registration number 440-12-17122012; International Clinical Trials Registry Platform identifier DRKS00004942).
Inclusion was restricted to patients with a diagnosis of PKU or MHPA confirmed by newborn screening and the initiation of dietary treatment (a protein- and/or Phe-restricted diet and nutrient supplementation by Phe-free protein substitute) during the neonatal period. Patients born before term (ie, at <36 weeks’ gestation), receiving sapropterin as an additional or exclusive treatment, with tetrahydrobiopterin deficiency or additional chronic diseases, long-lasting concomitant medication for other reasons than the treatment of PKU, and implementation of other diets were excluded from the analysis. Analyses were performed separately for patients with PKU and those with MHPA (dried blood Phe concentrations <600 µmol/L not undergoing any dietary therapy).
Anthropometric data (height, weight, and BMI) were obtained from the national computerized database, CrescNet.22 This Internet-based network was established in 1998 at the University Children’s Hospital in Leipzig for the early detection of growth disturbance23; it is being used by several hundred pediatricians across Germany for entering height and weight at preventive medical checkups from birth through adolescence. The database currently contains anthropometric data from ∼730 000 children and adolescents. The height and weight of patients with PKU were measured at each clinic visit and entered in the CrescNet database. All available data points were included in the analysis. Because of the variances in individual patient ages at the time of data analysis and in the individual patient frequency of follow-up appointments, the amount of data differs among patients and the different investigated time points. For the analysis of height, weight, and BMI, the measurement point nearest to the birthday (±6 months) was chosen. The individual target height was calculated by using the parental height24 when parents’ heights were available. BMI was calculated by the ratio of weight in kilograms to height in meters squared. Growth velocity was defined as the difference between 2 body height measurements divided by the time interval. This analysis only involved growth rates from annual intervals (365 ± 120 days).
To evaluate patients’ lifetime metabolic control, their individual average Phe concentration per year from birth onward to the age of 18 years was calculated retrospectively from medical records. Phe concentrations were measured in dried blood by liquid chromatography/tandem mass spectrometry or by fluorometry.27
Phe tolerance, total protein intake, and protein intake from casein hydrolysate or an AAM were gathered from medical records. Annually, the mean intake of synthetic and natural protein was calculated (intake in grams per kilogram of body weight per day).
To investigate the influence of the type of protein supplement used, patients with PKU were divided into 2 diet groups: (A) patients born before 1980 who consumed a casein hydrolysate (Berlophen) during childhood (in at least the first 10 years of life) and (B) patients born after 1989 who exclusively received an AAM.
Statistical analyses were performed by using IBM SPSS Statistics for Windows 20 (IBM Corporation). The anthropometric data, given as SDSs, were averaged and compared with reference values by using 1-sample t tests. Differences between groups were calculated by using t tests for independent samples. The comparison of patients’ final height to their target height was performed by using paired t tests.
We performed additional time series analysis of the same anthropometric variables by linear mixed model, using package lme4 of the statistical software suite R (R Foundation for Statistical Computing, Vienna, Austria; www.r-project.org). To compare PKU and MHPA, or diet groups, we treated patient identification as a random effect by assuming a random intercept model, and we treated age, sex, and diagnosis or age, sex, diet, and the interaction of age and diet as fixed effects. Comparisons of PKU, MHPA, and diet groups with the normal population were performed by testing the fixed intercept of a covariable-free random intercept model. Significance was accepted for P < .05.
A total of 233 white patients with PKU or MHPA, born between 1969 and 2014, could be identified. Of these, 224 patients (105 female patients, 119 male patients) could be included in the final analysis, 183 with PKU (82% of total) and 41 with MHPA (18% of total).
Patients with PKU born before 1990 received a casein hydrolysate supplement for several years. Since 1990, all patients with PKU had obtained AAMs from different manufacturers. In 27 patients with PKU, born between 1971 and 1981, the diet was interrupted during childhood or adolescence, according to the former guidelines (at a median age of 7.01 years [range: 6.0–15.0]; ref 28). Of these, 19 restarted therapy before adulthood (at a median age of 14.0 years [range: 12.07–18.0]), following their treating pediatricians’ advice.
Analysis of Height, Weight, BMI, and Growth Velocity According to Phenotype
From birth throughout childhood and adolescence, patients with PKU were significantly shorter than healthy German children; height was expressed as the mean height SDS (Fig 1A) and confirmed by linear mixed model analysis (Table 3). The final height SDS at the age of 18 years was significantly lower than the reference height (−0.882 ± 0.108, P < .001), corresponding to a height deficit of ∼5 to 6 cm at full growth.29,30 The mean final height of patients with PKU was 163.3 ± 5.9 cm (median: 163.4 cm, range: 152.1–175.1 cm) in women and 173.8 ± 8.9 cm (median 173.2 cm, range 158.0–189.7) in men. In comparison, healthy German women reached a mean height of 167 cm, and healthy German men reached a mean height of 180 cm.31 Individual target heights could be calculated for 23 women and 15 men; this analysis revealed a final height significantly lower than their target height24 (mean ± SD of real final height versus target height: women: 163.4 ± 6.0 vs 166.4 ± 4.1 cm, P = .02; men: 174.8 ± 8.1 vs 179.4 ± 4.9 cm, P = .01). To analyze a potential influence of maternal genotype on birth length, we compared birth length in patients with PKU from 11 mothers who were heterozygotes for R408W, a severe mutation,20 and the remaining cohort. No significant difference could be detected (P = .53).
In general, patients with PKU showed a significantly lower growth rate than the healthy population (Table 3). The initial growth rate of patients was significantly lower than the reference population’s (Fig 1B). From the age of 2 until 12 years, the growth rate normalized and reached values comparable to reference data. Afterward, it decelerated again, yielding a final negative SDS. The growth rates of ages 16, 17, and 18 could not be statistically analyzed because no reference values were available.25
The weight SDS of patients with PKU was lower than that of healthy German children (Table 3), reaching significance from the age of 2 years onward (Supplemental Table 9, Supplemental Fig 4A). The BMI SDS of patients with PKU was not significantly different from that of their healthy peers (Table 3).
Patients with MHPA also exhibited a significantly lower birth length SDS and height SDS during the first 6 years, as well as a significantly lower initial growth rate, compared with the reference (Fig 1 C–D, Table 2). However, growth rates improved after the first year of life and were not significantly different from those of the reference population. Because of the small number of patients with MHPA, an evaluation of final height is not yet possible. Regarding weight and BMI, patients with MHPA developed similarly to their healthy peers (Supplemental Table 10, Supplemental Fig 4B).
Patients with MHPA showed significantly higher weight SDS and BMI SDS than patients with PKU (Table 4).
Height and Weight of Patients With PKU According to Protein Supplementation During Childhood
The lower height SDS was observed in all patients, irrespective of diet group; it was more pronounced in patients receiving a casein hydrolysate supplement during childhood and adolescence (group A) compared with patients receiving an AAM (group B; Fig 3, Tables 5, 6, 7 and 8). The weight SDS was significantly lower in group A than in group B and in healthy peers. In addition, group B showed a higher weight SDS and BMI SDS than healthy children (Fig 3, Tables 5 through 8).
Protein Intake and Metabolic Control of Continuously Treated Patients With PKU
Total protein intake of continuously treated patients with PKU during their first 18 years of life was appropriate and greater than the recommendation for the healthy population (Supplemental Table 11). Mean dried blood Phe concentrations remained in the age-specific recommended range during the evaluation period (ref 32; Supplemental Table 12).
Anthropometric parameters are important in evaluating the state of health in childhood and adolescence, especially in patients who depend on special dietary treatment. In this longitudinal study, we investigated the growth development of patients with PKU and MHPA during their first 18 years of life. Patients with PKU showed reduced height and growth rates, compared with the reference group, which were apparently modified by the kind of protein supplement. There is no evidence for either severe malnourishment or a higher percentage of overweight or obesity in patients with PKU.
The study, through its longitudinal design and the large regional cohort, was ideal for analyzing growth development. In this regard, the computerized CrescNet database has proved to be a valuable tool for population-based studies.23,33 The data were compared by using the Kromeyer-Hauschild et al26 reference values for height, weight, and BMI, which continue to be the current common reference values in Germany. One might put forward the criticism that these reference values were obsolete and imprecise because they referred to pooled data from 17 studies with different methods of data collection performed between 1989 and 1999. However, the differences with recent reference data for height and weight, collected by standardized measurement methods, were only marginal.34
Human growth is divided into 4 growth phases corresponding to fetal growth, infancy, childhood, and puberty, which are variously related to genetic, endocrine, and nutritional influences.1 At birth, patients with PKU and MHPA already showed a reduced body length compared with the national reference data, suggesting intrauterine growth retardation. Importantly, none of the patients was a premature infant. Our data confirm the findings from previous studies in smaller cohorts.15,35,36 In the current study, patients with PKU remained shorter throughout childhood and adolescence up to the age of 18 years, compared with their healthy peers, confirming the findings from earlier studies detecting growth retardation in patients with PKU who were diagnosed early and continuously monitored.10,14–18 However, the authors of most of these studies have reported far smaller cohorts or cross-sectional data. Aldámiz-Echevarría et al18 report on a similar study design; however, the cohort was heterogeneous, deriving from 19 different treatment centers throughout Spain, the Baleares, and the Canary Islands, and there was some discrepancy between height measurements and longitudinal growth rates.
In a subgroup of patients, we could confirm our data by revealing a significantly shorter final height compared with their expected target height according to the formula by Hermanussen and Cole.24 A direct comparison with healthy siblings was not possible because of the limited availability of their growth data. However, the parents showed no sign of reduced final height, which underlines the idea that the growth deficit in patients with PKU is disease- and/or therapy-related (patients’ fathers: 179.4 ± 7.9 cm, healthy German men: 174.8 ± 8.1, patients’ mothers: 166.0 ± 6.2 cm, healthy German women: 166.4 ± 4.1; ref 31). None of the previously published articles revealed such homogenous results.
Intrauterine growth retardation is 1 symptom of the maternal PKU syndrome resulting from high maternal Phe concentrations during pregnancy.37 Even heterozygotes for PKU showed a reduced ability to metabolize Phe compared with mothers without any mutation in the PAH gene.38,39 In this regard, heterozygotes with more severe mutations display a more pronounced decrease of Phe hydroxylation and, therefore, lower plasma tyrosine concentrations compared with heterozygotes with milder mutations.38,40 These effects presumably influence intrauterine growth, depending on the maternal genotype and the related maternal Phe hydroxylation capacity.35 In this cohort, an influence of maternal genotype on intrauterine growth could neither be confirmed nor disproved because of the limited data on maternal genotypes. However, an analysis of a small subgroup revealed no significant differences regarding birth length between patients with PKU from mothers heterozygous for R408W and the remaining cohort.
Final height depends on several variables; 1 of them is birth length,30 whereas postnatal growth is influenced by genetic and endocrine factors as well as nutrient supply.2 The importance of nutrition in regard to physical development could be impressively observed during the 19th and 20th centuries, when the population’s height generally increased, a phenomenon referred to as “secular trend.” This was caused by improved health care and sanitary conditions, as well as by better food availability and, therefore, improved nutrient supply.41
All patients of our cohort were born in the eastern part of Germany (formerly the German Democratic Republic). The products used for protein substitution changed with the German reunification: although a casein hydrolysate supplement (Berlophen) was used until 1989, AAMs from several manufacturers became available afterward. The exact amino acid composition of Berlophen was unknown, varied between batches, and was not standardized, and it did not contain any additional micronutrients. Additional supplements were therefore required.42,43 A suboptimal nutrient supply for these patients during childhood may thus have had a detrimental influence on their growth. This is supported by the observation that younger patients with PKU, born after 1989 and exclusively receiving an AAM, exhibited better growth. A former cross-sectional study had already revealed more stunted growth in older patients with PKU, compared with later birth cohorts.44 The nutrient supplementation by protein supplements is indispensable for patients with PKU because their natural food choices are limited. Approximately 50% to 90% of the recommended daily protein and micronutrient requirements are met by nutrient supplements.45,46 Over the last decades, there have been continuous advancements in these supplements, improving their composition toward becoming an adequate nutrient supply.47
In patients with PKU born before 1980, dietary therapy was interrupted during childhood, according to the former guidelines.28 Despite the chance of increased intake of natural protein, this did not improve growth rate. Apparently, these children did not change their basic eating habits sufficiently to achieve catch-up growth. Eating habits are established during early childhood, and a fundamental change in eating patterns is almost impossible beyond the first years of life.48,49
There was no evidence for a higher rate of overweight or obesity in the entire cohort, which is in contrast to the findings of recent studies from various countries.50–52 These heterogeneous results might be explained by different culturally conditioned eating habits as 1 important factor influencing weight. Interestingly, our younger patients, born after 1989, showed an increase of weight up to the average values of the reference population. Both improvements of the dietary treatment and the secular trend may explain this finding.42
Patients with PKU who are diagnosed early, treated, and continuously monitored show reduced height and growth from birth onward, leaving them ∼6 cm shorter than their healthy peers. During the last 2 decades, this phenomenon attenuated, probably because of the use of modern AAMs and the broad palette of special low-protein foods. However, patients with PKU remain shorter, suggesting that this phenomenon cannot be explained by nutrition alone. Extended longitudinal studies should include other potential influencing factors: for example, tyrosine supply and genotypes.
We thank all participating patients for providing additional data, especially for adding data on their parents’ height; Beate Peinel, who performed some of the regular measurements of weight and height of the patients; Mandy Vogel (former staff member of CrescNet) for valuable discussions and help with the statistical analysis; and Drs Markus Ott, Oec. Troph., Andrea Sülzle, Troph., and Ms Janina Lahl, Dipl. Troph. from Nutricia GmbH, Nutricia Metabolics, for providing us with information about composition of former amino acid supplements.
- Accepted August 11, 2017.
- Address correspondence to Skadi Beblo, MD, Hospital for Children and Adolescents, Center for Pediatric Research Leipzig (CPL), Department of Women and Child Health, University Hospitals, University of Leipzig, Liebigstraße 20 a, D-04103 Leipzig, Germany. E-mail:
This trial has been registered with the International Clinical Trials Registry Platform (identifier DRKS00004942).
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
FUNDING: Supported by an unrestricted research grant from Nutricia Metabolics, Nutricia GmbH Germany and Merck Serono GmbH to Dr Beblo and Ms Thiele through the University of Leipzig.
POTENTIAL CONFLICT OF INTEREST: This investigator-initiated study was partly supported by an unrestricted research grant from Nutricia Metabolics, Nutricia GmbH, Germany and Merck Serono GmbH to Dr Beblo and Ms Thiele through their employer the University of Leipzig; the other authors have indicated they have no potential conflicts of interest to disclose.
- Rogol AD,
- Clark PA,
- Roemmich JN
- Guthrie R,
- Susi A
- Scriver CR,
- Kaufmann S,
- Eisensmith RC,
- Woo SL
- Verkerk PH,
- van Spronsen FJ,
- Smit GP,
- Sengers RC; The National PKU Steering Committee
- Allen JR,
- Humphries IR,
- Waters DL, et al
- Danecka MK,
- Woidy M,
- Zschocke J,
- Feillet F,
- Muntau AC,
- Gersting SW
- DaBelle Laboratory
- University Children’s Hospital, University of Leipzig
- Kromeyer-Hauschild K,
- Wabitsch M,
- Geller F, et al
- Destatis Statistisches Bundesamt
- Bremer HJ,
- Bührdel P,
- Burgard P, et al
- Neuhauser H,
- Schienkiewitz A,
- Rosario AS,
- Dortschy R,
- Kurth BM
- Koch R,
- Hanley W,
- Levy H, et al
- Theile H
- Mennella JA,
- Griffin CE,
- Beauchamp GK
- Copyright © 2017 by the American Academy of Pediatrics