Onset of puberty in boys with short stature is earlier in those born small-for-gestational age and later in those born appropriate-for-gestational age than in the general healthy population

Abstract

This study compared the age at onset of puberty among boys with short stature (prepubertal height SD score (SDS) ≤ –2.0 SD) who were small-for-gestational age (SGA) and those who were appropriate-for-gestational age (AGA), as well as healthy boys from the Ogi Growth Study. The study population consisted of boys who presented with short stature between 2008 and 2023. The study included 238 boys, including 229 whose age at puberty onset was auxologically determined and nine whose physical evaluations were performed at the onset of puberty. Of them, 40 were included in the SGA group and 198 in the AGA group. The control group consisted of 227 healthy boys from the Ogi Growth Study whose puberty was auxologically determined. Prepubertal height SDS was not significantly different between the SGA and AGA groups. Age at puberty onset was significantly earlier in the SGA group (10.87 ± 0.98 yr) than in the AGA group (11.90 ± 1.06 yr). Moreover, age at puberty onset was significantly earlier in the SGA group and significantly later in the AGA group than in the control group (11.28 ± 0.95 yr). Therefore, when following children with short stature, clinicians should consider the likelihood of early puberty if they are SGA.

Highlights

● Age at puberty onset was earlier in SGA boys than in AGA boys with short stature.

● Age at puberty onset in general healthy boys was significantly later than that in the SGA short boys and significantly earlier than that in the AGA short boys.

● Of SGA and AGA boys, 22.5% and 2.5%, respectively, met the criterion for precocious puberty treatment.

Introduction

The World Health Organization (WHO), Japan Society of Obstetrics and Gynecology, and Japan Pediatric Society define small-for-gestational age (SGA) as a condition in which height and weight at birth are below the tenth percentile according to the neonatal anthropometric charts for gestational age at birth. SGA has various causes, including abnormalities in the placenta, umbilical cord, and other organs; maternal factors such as mother’s age, nutrition, diseases, smoking, and drinking; fetus-derived factors such as chromosomal abnormalities; and multiple pregnancies (1).

Inconsistent results have been reported by studies in the context of age at the onset of puberty in children who were SGA at birth. Some reports state that it was later than that in children who were appropriate-for-gestational age (AGA) at birth, while others stated that it was earlier than that in children who were AGA at birth (2,3,4,5,6). We believe that this inconsistency can be attributed to the following factors: the definition of SGA varies among countries; the prepubertal physique of SGA boys with short stature and that of AGA boys with short stature are not consistent, even when their physiques are similar; different studies used distinct criteria for physique; and some studies included individuals treated with GH.

In Japan, the age at the onset of puberty in SGA children with short stature who received GH therapy is not different from that in healthy children (7). In addition, the age at puberty onset in children with short stature is later than that in children of standard height (8). However, the timing of puberty onset in untreated boys with short stature born SGA and AGA has not been compared.

In boys, the onset of puberty is defined as an increase in testicular volume up to 4 mL (9, 10). However, recognizing this change at home without medical supervision is impossible. In clinical practice, physicians rarely examine boys when their testicular volume reaches 4 mL. In such situations, auxological methods based on body measurements may help estimate the onset of puberty. In our previous study, we established an auxological method to determine the onset of puberty based on growth curves (11).

In this study, among untreated boys with short stature and prepubertal height ≤ −2.0 SD, age at the onset of puberty determined mainly using the auxological method was comparatively analyzed between boys with short stature born SGA and AGA and compared with general healthy boys.

Patients and Methods

Boys who visited Tanaka Growth Clinic, Kibounomori Growth and Development Clinic, or Tachibanadai Clinic with a chief complaint of short stature between 2008 and 2023 and whose prepubertal height SD score (SDS) was ≤ −2.0 SD comprised our study population. Nine boys at the onset of puberty were identified biologically. A total of 229 patients who had already reached puberty at the time of their initial visit and whose age at the onset of puberty could be auxologically determined by drawing growth curves were included in the study, as described previously (10). Of the 238 boys, 40 were in the SGA group and 198 in the AGA group. Patients with pituitary hypoplasia, those who received treatments that affected growth before the onset of puberty, including GH therapy, and those with Down’s syndrome were excluded from the study.

Data from 227 healthy boys from the Ogi Growth Study were used as controls. The participants in the Ogi Growth Study were born in Saga Prefecture between 1963 and 1978 and participated in the Ogi Growth Study between 1979 and 1992. They were measured once a year and had seven or more longitudinal height and weight measurement records and bone age measurement records. The age of the participants at the onset of puberty was determined using an auxological method with growth curves, as reported previously (10).

The testicular volume reaching 4 mL or LH level being ≥ 0.3 mIU/mL with 3 mL of testicular volume was defined as the biological onset of puberty. If the gestational age was unknown, it was assumed to be 3 d. If whether the boy’s mother was primiparous or multiparous at birth was unknown, she was assumed to be primiparous.

To calculate the height SDS, degree of obesity, and BMI SDS, we used the calculation file for the physical index (taikakushisu_v3.3.xlsx) published by the Japanese Society for Pediatric Endocrinology. This calculation file is based on body measurement data published by the Ministry of Health, Labour and Welfare and the Ministry of Education, Culture, Sports, Science and Technology in fiscal year 2000 (12). The Ito method was used to determine the degree of obesity.

EZR software (13) was used for all statistical analyses. The Mann-Whitney U test was used to analyze differences between the SGA and AGA groups. When significant differences were determined using the Kruskal-Wallis test for clinical factors among the three groups, clinical factors that showed significant differences were evaluated using Bonferroni’s multiple comparison test among the three groups. The log-rank test was used to determine significant differences between the Kaplan-Meier curves. Statistical significance was set at P < 0.05.

This study was approved by the Japan Medical Association Ethics Review Committee (R4–14) and conducted in accordance with the tenets of the Declaration of Helsinki.

Results

In the context of clinical factors at birth (Table 1), gestational age was significantly earlier in SGA boys with short stature than in the AGA boys with short stature (p < 0.001). Birth weight and height were significantly lower in SGA boys with short stature than in AGA boys with short stature (p < 0.0001). No significant difference was observed in the height of the parents between AGA boys with short stature and SGA boys with short stature.

Table 1. Clinical characteristics at birth.

No significant difference was observed in the prepubertal height SDS between AGA boys with short stature and SGA boys with short stature. However, the height SDSs in SGA and AGA short boys were significantly lower than those in boys of the general population. Prepubertal auxological data were obtained from a health examination conducted at an elementary school. Weight and BMI SDS were significantly lower in SGA boys with short stature than in AGA boys with short stature and were lower than those in the general population (Table 2). The BMI SDS was < +1 SD in all SGA boys with short stature; therefore, none of them were obese. Of the AGA boys with short stature, six (3.0%) had a BMI SDS of > +1 SD. Among them, one (0.5%) had a BMI SDS > +2 SD, more specifically +2.59 SD.

Table 2. Clinical characteristics at prepuberty.

The ages at the onset of puberty among the three groups were significantly different (Table 3). Age at the onset of puberty in SGA boys with short stature (10.87 ± 0.98 yr) was significantly earlier than that in boys of the general population (11.29 ± 0.95 yr) (p < 0.05) and in the AGA boys with short stature (11.90 ± 1.06 yr) (p < 0.0001). Height and height SDS at the onset of puberty were significantly lower in SGA boys with short stature.

Table 3. Clinical characteristics at the onset of puberty .

A histogram of the age at puberty onset in SGA and AGA boys with short stature is shown in Fig. 1. The age at the onset of puberty in SGA boys with short stature ranged from 9 to < 13 yr, with a peak at 10 yr. None of them attained puberty when they were < 9 yr or ≥ 13 yr of age. AGA boys with short stature demonstrated a peak at 11 yr of age, which was 1 yr later than that of SGA boys with short stature. Moreover, AGA boys had a wider range of puberty onset, ranging from 9 to < 15 yr of age.

Fig. 1.

Histogram of age at the onset of puberty in SGA and AGA boys with short stature.

Fig. 2 shows the Kaplan–Meier curves. Including boys from the general healthy population of the Ogi Growth Study (11) (prepubertal height SDS, −0.26 ± 0.86 SD; age at onset of puberty, 11.28 ± 0.95 yr), three groups were compared using the log-rank test, and a significant difference was observed (p < 0.0001). Moreover, comparisons between SGA boys with short stature and boys of the general population and between AGA boys with short stature and boys of the general population were also performed. Significant differences were observed in both comparisons (p < 0.05 and p < 0.001, respectively).

Fig. 2.

Survival curves of age at puberty onset in SGA boys with short stature, AGA boys with short stature, and boys of the general population (Ogi Growth Study).

In this study, none of the participants reached puberty before the age of 9 yr, which is the diagnostic criterion for precocious puberty established by the Hypothalamo-Pituitary Disorder Study Group. However, it is allowed in the guideline to treat boys < 10 yr of age, which is a year older than the criterion for precocious puberty, when his height SDS is ≤ −1 SD. In the context of this criterion, precocious puberty was observed in 22.5% of SGA boys with short stature and in 2.5% of AGA boys with short stature. In contrast, when the criterion for delayed puberty is ≥ 14 yr of age, none of the SGA boys with short stature and 3.5% of the AGA boys with short stature met this criterion.

Discussion

A meta-analysis by Deng et al. that compared AGA and SGA children (14) showed no difference in age at the onset of puberty in boys born SGA. However, when individual studies were analyzed (Table 4), multiple conflicting results were observed. Among the children with short stature, SGA children were faster than, not different from, and slower than AGA children. This was due to differences in the method of determining age at the onset of puberty, criteria for identifying SGA at birth, criteria for determining the physique of the AGA group used as a control, and the presence or absence of GH treatment.

Table 4. Previous studies on age at onset of puberty in SGA and AGA children.

Therefore, more appropriate criteria are required for comparison. The age at the onset of puberty can be determined using two methods: defining the onset of puberty as the start of an upward rise in the growth curve from its prepubertal slope and methods determining Tanner stages based on physical examination. When Tanner staging is performed at intervals of 3–12 mo in the prepuberty stage, with the longest interval being 1 yr, it may create a discrepancy between the determined age at the onset of puberty and the actual age at the onset of puberty. To determine the age at puberty onset accurately, we believe that our retrospective method using a growth curve would be effective. A previous study has reported that the age at the onset of puberty determined by this growth curve method is similar to that of physiological examinations (11).

Stepwise multiple regression analysis performed in the Ogi Growth Study showed that the prepubertal height SDS affected age at the onset of puberty (15). Therefore, it is necessary to perform a comparative analysis in a population of individuals with a similar prepubertal height SDS. Only a few studies have performed comparative analyses of participants with similar prepubertal height SDS. These include studies by Lazar et al. (2) and Ueta et al. (5) (Table 4). Participants in the study by Lazer et al. (2) were “untreated,” and the only used criterion at birth was “birth weight < −2 SD.” Although no criteria were used for birth height, all their participants had “a prepubertal height SDS < −1.7 SD,” and these were like our criteria with a slight difference. Similar to our study, their results showed that the age at puberty onset was earlier in SGA children than in AGA children. In Japan, Ueta et al. (5) followed “untreated” children with short stature with “prepubertal height < −2 SD.” The criteria for physique at birth and prepubertal height SDS were identical to those used in the present study. However, their results differed from those of the present study in that no significant difference was observed in the age at the onset of puberty. Their study differed from ours in the following aspects. First, they investigated age at the onset of puberty as the SDS of age at the onset of puberty in a population comprising boys and girls. Second, the criterion for the onset of puberty in boys was a testicular volume of 3 mL. Finally, of 38 participants, 20 were boys, only three of whom were SGA. We believe that these differences contributed to the differences in the results.

Komatsu et al. (8) reported that age at the start of the growth spurt was significantly later in children with short stature with prepubertal height being ≤ −2 SD compared with a group of standard height children, wherein age at the start of the spurt was defined as age at the nadir of growth velocity curve after the growth curve was interpolated by smoothing spline function. The finding of our study that age at the onset of puberty was significantly later in AGA boys with short stature with prepubertal height SDS ≤ −2 SD than in boys from the general population supported their report.

Hormonal puberty begins with the disinhibition and reactivation of the LHRH pulse generator. The LHRH pulse generator develops and functions in the fetus, continuously functions in early infancy, gradually decreases in activity during late infancy, and remains quiescent during childhood. Boys born SGA have higher serum LH and FSH levels during early infancy (16, 17). Boys born SGA may have a higher sensitivity to LHRH and respond earlier to reactivated LHRH pulses than boys born AGA.

None of the boys met the diagnostic criterion for precocious puberty in Japan, which is the onset of puberty at < 9 yr of age. However, the criterion for treating children with short stature of ≤ −1 SD is 1 yr older, that is, younger than 10 yr of age. According to this criterion, the proportion of boys with precocious puberty was low (2.5%) among AGA boys with short stature and high (22.5%) among SGA boys with short stature. Therefore, the onset of puberty in children with short stature with height ≤ −2.0 SD is rarely early if they were AGA at birth and can be early in approximately one-fourth of children who were SGA at birth.

Delayed puberty in boys is defined as not attaining puberty by the age of 14 yr (11, 18, 19). In the present study, all seven boys who attained puberty at ≥ 14 yr were AGA boys with short stature (3.5% of all AGA boys). They had been < −2.0 SD in height SDS before the onset of late puberty.

In clinical practice, it is advisable to observe children with short stature, considering the presence of syndromes and the possibility of early puberty if they are SGA. The factor that most affects adult height is height at puberty onset (15). Therefore, SGA children with short stature who do not catch up and have an early onset of puberty are expected to have a lower height at the onset of puberty and lower adult height.

Children born SGA comprise a heterogeneous group. Although we did not investigate their etiologies, the fact that they tended to enter puberty earlier suggests that SGA itself influenced early puberty regardless of etiology.

The average number of weeks of gestation for SGA infants is 36 wk, including many preterm births. However, preterm birth does not necessarily lead to earlier puberty, and SGA is thought to have a stronger effect on age at puberty onset (20).

One of the limitations of this study is that both biological and auxological definitions of puberty onset were used, although auxological puberty onset is close to biological puberty onset (11), and only nine boys were biologically determined. The second limitation is that the comparison between the participants of the Ogi Growth Study and the patients of clinics is not appropriate, as the participants of the Ogi Growth Study were born between 1963 and 1978, and there is a secular trend of pubertal maturation. However, this comparison can provide a clue for understanding the difference in pubertal onset between short and normal-height children. The third limitation was the use of bone age, which is useful for evaluating puberty. As most patients visited our clinic after entering puberty, we could not obtain bone age before puberty.

Conclusion

The age at puberty onset was significantly earlier in SGA boys with short stature than in AGA boys with short stature. Moreover, the age at puberty onset was significantly earlier in SGA boys with short stature and later in AGA boys with short stature than in healthy boys.

The proportion of SGA boys with short stature who met the treatment criteria for precocious puberty was approximately nine times higher than that of AGA boys with short stature. None of the SGA boys with short statures met the criteria for delayed puberty.

In clinical practice, during follow-up of children with short stature, the possibility of early puberty if they are SGA should be considered.

Conflict of interests

The authors declare no conflict of interest.