Down syndrome (DS) is the most common chromosomal disorder and leading cause of intellectual disability and growth retardation. Average height is around the −2 standard deviation (SD) for most ages compared to normal population. The growth pattern is impaired from birth to adolescence and particularly during the interval of 6 months to 3 years and puberty [Cronk et al., 1988]. In addition, these patients also have reduced pubertal growth spurt, contributing to the low final height [Cronk et al., 1988; Annerén et al., 1990]. It was demonstrated that children with DS do not have growth hormone (GH) deficiency [Annerén et al., 1986], but hypothalamic dysfunction was reported by Castells et al. . A study performed by Annerén et al.  revealed that GH was given to children with DS increased growth velocity. However, Myrelid et al.  reported that GH treatment had no effect on final height.
Since children with DS have different growth patterns than other normal children, it is critically important to utilize syndrome specific growth charts. First growth charts for DS have been developed by Roche , followed by Rarick and Seefeldt . Cronk et al.  was studied 780 children with DS between 1 month and 18 years of age and established growth charts with these data in USA and also showed that moderate to severe CHD affected growth in DS children. The height of the children with DS shows variability between countries compared with the general population. Piro et al. , Fernandes et al. , Styles et al. , and Myrelid et al.  presented growth charts of DS for Sicily, Portuguese, UK, and Sweden, respectively and the studies revealed that Swedish, English, and Portuguese children were taller than corresponding US-American children and much taller than their Sicilian counterparts.
The aim of this study was to develop growth charts for Turkish children with DS between 0–18 years of age and to investigate the effects of congenital heart disease (CHD) and hypothyroidism on growth.
MATERIALS AND METHODS
A total of 1,726 DS children (736 females, 990 males) who were born in the years between 1980 and 2010 and followed up in Pediatric Genetic Department of Cerrahpasa Medical Faculty of Istanbul University were evaluated. Children with DS were followed up from birth to 18 years of age. The data were created by assessment of parameters including chromosomal analysis, TSH and T4 values, hemogram, and echocardiography which were obtained at first admission and the assessment of growth and development which were performed monthly and recorded up to the first age. For the following visits the assessment of growth and development were recorded every 2 months up to the age of 2 years, every 3 months up to the age of 3 years, every 6 months up to the age of 6 years and annually afterwards. The parameters including height, weight, and head circumference (HC) were measured by pediatricians. The parameters that were used for development of growth charts were; age at admission (months and years), height (cm), weight (gram and kg) and HC (cm). Total measurement numbers were 8,132 for height, 8,606 for weight, and 7,375 for HC (Table I). Growth curves were created from 24.113 selected measurements of 1,726 children.
Table I. The Distribution of Observation Number According to Age
The cohort of this study was evaluated in two main groups for the CHD DS children with CHD requiring surgical intervention such as atrioventricular septal defect, large ventricular septal defect or tetralogy of Fallot, and pulmonary vascular disease were categorized as severe CHD. The surgical correction of CHD was usually performed on children with DS at the age of 2–6 months in Turkey. DS children with small VSD or ASD and other anomalies not requiring surgical intervention or medication were categorized as mild group and evaluated together with DS children without CHD for the effects of CHD on growth. In addition, DS children with congenital primary hypothyroidism and compensated hypothyroidism were classified as DS with hypothyroidism.
The statistical analysis of this study was performed by utilizing a statistical software, SPSS (Statistical Package for Social Sciences) for Windows 17.0. The mean and SD scores (SDS) were calculated for each of the parameters. LMS method by Cole and Green  was carried out to fit percentile curves to the data. Reference percentile curves exhibit the distribution of a measurement since it changes according to some covariates including age. In LMS method median (M), coefficient of variation (S), and skewness (L) values are represented by three curves which show distribution changes. Skewness also defined as a Box–Cox power. Using penalized likehood, the three curves can be fitted as cubic splines by non-linear regression and the extent of smoothing required is described through smoothing parameters or equivalent degrees of freedom.
We assessed the height, weight, and HC measurements of 1,726 Turkish children with DS between 0 and18 years of age. Growth curves are presented in Figures 1–3. In addition, we also compared the height curves presented here with that for normal Turkish children [Neyzi et al., 2006, 2008] in Figure 4a–d.
The mean birth length was reduced by −0.5 SD for Turkish standards in both boys and girls. The mean height SDS of boys and girls at 6 months of age were −0.7 for healthy counterparts, corresponding to −1.9 and −2.2 at 3 years of age and at 8 years of age, respectively. The mean final height values were 160 and 144 cm for boys and girls, respectively, and −2.56 and −3.06 SD for Turkish standards.
The mean birth weight SDS of DS children were −0.8 for healthy counterparts. The SDS of weight were −1, −0.7, and − 0.5 at 6 months, 3 years, and 5 years of age, respectively. The final weight SDS were −0.3 (boys) and +0.5 (girls) for Turkish standards.
Compared to healthy counterparts, the mean HC values of DS children were corresponding to −0.9 SD at birth, however after 6-month, were below −2 SD. The final HC values were 54 cm (−1.02 SD) in boys and 51 cm (−2.21 SD) in girls for Turkish Standards.
Evaluation of the effects of CHD on growth is presented in Table II. In the first 3 months of life weight was decreased (P < 0.0001) in severe CHD group and this condition were observed until 4 years of age. However, there is no statistical difference in regard to weight of DS children with and without severe CHD at 5 years of age. We observed that treated hypothyroidism had no effect on growth of DS children.
Table II. Effect of Congenital Heart Diseases on Growth
Severe CHD (−)
Severe CHD (+)
Bolded numbers represent significant P values.
We present the assessment of height, weight, and HC measurements of children with DS in order to establish growth curves for Turkish population. We have compared the mean height of children with DS with the mean height of normal population. We have observed that the birth length of children with DS was −0.5 SD of the mean of the normal population. When the children with DS reaches to 3 years of age the SDS was below to −1.9 and throughout the life span the SD value continues to 2 SD less than in the normal Turkish population [Neyzi et al., 2006, 2008]. It is known that growth velocity of children with DS decelerates markedly between 6 months to 3 years and at puberty [Cronk et al., 1988; Myrelid et al., 2002]. In our study, the boys with DS who were reached to the pubertal growth spurt between 11 and 16 years of age, achieved the annual growth rate of 8.15 ± 4.13 cm. Their final height was accomplished at the age of 17 with the height of 160 ± 7.3 cm. The girls with DS reached puberty spurt between 9 and 12 years, and achieved the annual growth rate of 5.9 ± 1.64 cm; then growth slowed and reached final height at 14 years. Their final height was 144 ± 6.01 cm. The study performed by Bundak et al.  showed that healthy Turkish boys have their pubertal onset at 11.6 ± 1.2 years of age, and they achieve the annual growth rate of 10.1 ± 1.6 cm. The duration of puberty is 4.9 ± 0.6 years and their final height is 176 ± 6 cm. The same study performed by Bundak et al.  demonstrated that healthy Turkish girls have their pubertal onset at 10.1 ± 1 years of age by achieving the annual growth rate of 8.5 ± 1 cm. Their puberty duration is 4.9 ± 1.2 years and final height is 163.7 ± 6 cm. Although the age at onset of pubertal growth spurt and puberty duration of the boys with DS were similar to normal population, their pubertal annual growth rate was inadequate; thus the final height was −2.56 SD for Turkish standards. However, the pubertal growth spurt of the girls with Down syndrome started 1 year earlier, their puberty duration was short and their annual peak height velocity was about 2–2.5 cm lower than that for normal children and as a result, the final height was −3.06 SD (Fig. 4a–d). Since Turkish girls with DS reached puberty earlier, their final height was shorter.
The mean birth weight of our patients with DS was slightly lower than normal population (−0.8 SD). Final weight values also were similar to Turkish standards (boys −0.4 SD, girls +0.5 SD).
When we compared the values of our patients with growth charts for DS ages between 0 and 4 years by Fernandes et al.  which have been developed for Portuguese, we found that weight and height percentiles of the boys were similar (P > 0.05). However, at the age of 4 years, our girls were shorter than Portuguese counterparts (P < 0.05). The mean values of height of Turkish children with DS were also similar to those in Egyptian counterparts in the 0–3 years group [Meguid et al., 2004], but also higher than Saudi children with DS (0–5 years) [Al Husain, 2003]. Turkish children with DS were taller and heavier than those (0–14 years) in Sicilian counterparts [Piro et al., 1990]. We have used growth charts by Cronk et al.  for US-American children with DS for DS follow up in our department for 20 years. The comparison of the height growth charts between 1 month to 3 years showed that Turkish children with DS and US-American counterparts were similar. However, after the age of 3 years, our boys also were taller than US-American boys. These differences may be the result of time delay between these studies. When comparing final heights of Turkish DS children with growth charts for English and Irish children with DS (0–19 years) [Styles et al., 2002] and for Swedish children with DS (0–18 years) [Myrelid et al., 2002] Swedish boys and girls and English girls were taller than our patients; English boys were shorter than our boys with DS.
In a recent study from Netherlands that has been published during the preparation of this manuscript, total of 1,596 children with DS were enrolled for the assessment of their height [Van Gameren-Oosterom et al., 2012]. In their study the mean final heights were 163.4 cm and 151.8 in boys and girls, respectively which were higher than those reported in the present study and other studies [Styles et al., 2002; Myrelid et al., 2002]. The reason for this differences in the final heights of the children with DS might be due to the ethnic background of the studied populations. On the other hand the observation of early puberty and shortening of puberty duration in DS, affecting the final height was similar to our observation.
Final weight of Turkish girls with DS was similar to that reported in other studies [Myrelid et al., 2002; Styles et al., 2002]. However weight of Turkish boys was greater than Swedish, English, and Irish children with DS [Myrelid et al., 2002; Styles et al., 2002]. Recently, Rosenbloom et al.  reported weight growth charts for DS children in USA; final weight values of children with DS were greater in other studies [Myrelid et al., 2002; Styles et al., 2002]. In the last decades, obesity has become an important health issue for US-American as well as Turkish DS children.
At birth, HC of DS children was −0.9 SD for healthy Turkish children; however after 6 months, it was below −2 SD. The final HC SDS also were −1.02 (boys) and −2.2 (girls) for Turkish standards [Örmeci et al., 1997]. Comparison of our HC data of the DS patients with other studies revealed that at 0–5 years of age English, Swedish, Saudi, and Portuguese and at 15-months of age Dutch counterparts had greater HCs than Turkish DS children [Fernandes et al., 2001; Myrelid et al., 2002; Styles et al., 2002; Al Husain, 2003; Van Gameren-Oosterom et al., 2012]. On the other hand, the values of our DS girls were similar to Egyptian and USA counterparts [Palmer et al., 1992; Meguid et al., 2004]. There are no data regarding HCs of DS children greater than 5 years of age in the literature.
Cronk et al.  studied on growth in DS and showed that moderate to severe CHD affected growth and indicated that this difference predominated in infancy and diminished through puberty. Meguid et al.  also found that the weight retardation was greater in DS children with CHD than in DS children without CHD between 1 and 36 months. We found that CHD affected weight between 3 months and 4 years in CHD group whereas weight were similar at the age of 5 in patients with or without CHD group. However, there is no statistically significant difference in values of height and HC between patients with or without severe CHD group similar to study of Meguid et al. . We found that treated hypothyroidism had no influence on growth to any significant extent similar to the previous study by Myrelid et al. .
In conclusion, the height and HC of Turkish children with DS were lower compared to normal population and growth velocity was reduced especially between 6 months and 3 years and at puberty. Compared to general Turkish standards, final height and HC SDS of Turkish girls with DS were lower than Turkish boys with DS whereas final weight SDS were mildly increased in Turkish girls with DS. The final height of Turkish children with DS was shorter than Swedish and Dutch boys and girls and English girls. The HCs in Turkish DS was also less than HCs of Swedish and English children with DS between 0 and 5 years. Syndrome specific growth curves are important for general health assessment and early diagnosis of accompanying illnesses. We suggested the use of more accurate syndrome-specific as well as country specific growth curves and thought that these percentile curves would be reference charts for physicians from Turkey.
We would like to thank Prof. Dr. Ahmet Dirican from Istanbul University, Medical Faculty, Department of Statistics and Prof. Dr. Andrzej Furman from Bogazici University, Institute of Environmental Sciences for contribution of statistical analysis, Pediatric residents and students for helping the measuring of the growth parameters and Adem Kabak for helping the drawing of the graphies. We are also greatful to Erdal Özkan who is chairman of “Dost Yasam” Turkish Down syndrome foundation and our families of DS patients for their collaboration.