|Home | About | Journals | Submit | Contact Us | Français|
Studies have postulated that hypospadias, prematurity, and low birth weight are linked by defects in androgen signaling. To determine whether premature, hypospadiac boys are small and remain so, we compared their size at birth and at hypospadias repair to premature boys who underwent post-neonatal circumcision.
We identified premature boys admitted to Texas Children's Hospital who underwent either hypospadias repair or circumcision after 4 months of age. Age, weight, and height at birth and surgery were recorded.
Fifty-four boys had hypospadias and 34 did not. For hypospadiac boys, the mean birth weight and age, height, and weight at surgery were lower than for boys without hypospadias. More importantly, length-for-age and weight-for-age percentiles were also lower for hypospadiac boys. When subset analysis was performed on boys younger than 2 years at surgery, however, there were no significant differences in height or weight between hypospadiac and nonhypospadiac boys.
Our series suggests that premature, hypospadiac boys are born smaller than age-matched, non-hypospadiac controls. However, there were no age-corrected size differences between hypospadiac and non-hypospadiac boys at surgery. This implies that hypospadiac boys exhibit postneonatal ‘rebound’ growth. Global growth deficits, if any, do not persist in hypospadiac boys.
Multiple groups have reported associations between prematurity, low birth weight, and hypospadias [1-5]. Placental insufficiency, which is linked to intrauterine growth restriction, low birth weight, and prematurity, has been hypothesized to lead to low levels of placental human chorionic gonadotrophin and fetal androgens. In turn, this may result in hypospadias [1-5]. An alternative but non-exclusive theory is that a global, persistent deficit in androgen signaling may be responsible for hypospadias . The latter possibility would result in impaired somatic growth of hypospadiac boys during childhood. We sought to test this hypothesis by comparing the birth and post-neonatal weights of premature hypospadiac boys to those of premature boys without hypospadias.
Institutional review board approval was obtained for this study.
Boys who underwent hypospadias repair or circumcision at our institution between 29 October 1999 and 7 April 2009, inclusive, were identified through the operative case logs of our institution's pediatric urology service. The medical records of these boys were screened to determine which patients were former premature infants (gestational age <37 weeks) admitted to our institution's neonatal intensive care unit. Boys with genetic syndromes known to affect growth (i.e. Prader-Willi) were excluded. Patients who underwent circumcision and featured cryptorchidism, ventral skin deficiency of the penis, and/or penile torsion, webbing or curvature were also excluded. Boys with hypospadias and either normally descended or undescended testes were included in the hypospadias cohort as long as they did not meet any of the exclusion criteria.
Data were collected on urologic diagnoses, degree of hypospadias, birth weight and height, gestational age at birth, age at surgery, and weight and height at surgery. Body mass index (BMI) was calculated for each patient, and BMI-for-age, weight-for-age, height-for-age, and weight-for-height percentiles were determined using the definitions of the Centers for Disease Control and Prevention (CDC), Division of Nutrition, Physical Activity and Overweight and Obesity, and National Center for Chronic Disease Prevention and Health Promotion. These definitions are based on reference data from the MO/WI Natality, NHES II and III, and NHANES I, II, and III studies [6-7].
Microsoft Excel 2007 was used to calculate two-tailed Student t-tests assuming equal variances between comparison groups (homoscedastic).
We identified 54 boys with hypospadias and 34 without hypospadias who met all inclusion and exclusion criteria. Mean gestational age at birth was 31.3 and 32.6 weeks for hypospadiac and non-hypospadiac boys, respectively (Table 1). The mean birth weight of hypospadiac boys was lower than that of boys without hypospadias (1541 vs 1875 g, P = 0.59).
Mean age at surgery was 1.2 and 3.2 years for hypospadiac and non-hypospadiac boys, respectively (Table 2, P < 0.00004). Accordingly, mean height and weight at surgery were lower for boys with hypospadias (P <0.00003 and <0.000007). More importantly, length-for-age and height-for-age percentiles were also lower for hypospadiac boys (P = 0.01 and = 0.02).
To better adjust for age, we performed subset analysis on boys younger than 2 years at surgery (whether hypospadias repair or circumcision). In this analysis, there were no significant differences between hypospadiac and non-hypospadiac boys in absolute or age-adjusted height or weight (Table 3, all P > 0.5).
Hypospadias may have an endocrinologic basis. Mouse models have shown that hypospadias and thinning of the periurethral corpora spongiosum can be seen in the male offspring of dams administered estrogens [8-9]. Sons born to mothers exposed in utero to the estrogenic compound diethylstilbestrol are at increased risk of hypospadias . In contrast, prenatal exposure to exogenous androgens may thicken the periurethral spongiosum . Boys born as a result of in vitro fertilization feature a higher likelihood of having hypospadias . This enhanced risk may be related to maternal progesterone administration, or other maternal or fetal endocrine abnormalities that may or may not be related to infertility and its treatment.
Holmes et al. have reported that serum levels of adrenocorticotropic hormone, androgen precursors and metabolites do not differ between hypospadiac and nonhypospadiac boys . This observation does not preclude the possibility that some cases of hypospadias are caused by defects in other aspects of androgen metabolism such as low follicle-stimulating hormone  and placental insufficiency. Yet another potential explanation is maternal androgen insufficiency; in one study, maternal testosterone levels at 6 14 weeks’ gestation were significantly decreased in pregnancies that resulted in growth restriction and male genital abnormalities .
An endocrinologic aspect to hypospadias may affect more than urethral seam formation. In utero exposure of male mammals to androgens is thought to result in greater birth weight . When sheep are exposed to prenatal, exogenous androgens, female offspring are larger, and other morphometric indices are increased in both sexes . In toto, these studies suggest that, in mammals, larger male offspring are a result of differences in androgen concentrations between male and female embryonic milieus. Conversely, inhibiting male hormone activity, either through direct antagonism or by increasing exposure to ‘female’ hormones, may result in smaller male offspring. Indeed, this has been the case in rodent exposures to very high doses of estrogens . Among humans, males have greater birth weights than females, presumably due to differences in fetal androgen levels [15, 18].
A large body of literature supports links between prematurity, low birth weight, and hypospadias [1-5, 19-25]. Hypospadias is more common in low-birth-weight boys, and this association is independent of gestational age . In monozygotic twins discordant for hypospadias, the boy with the lower birth weight has been reported to be more likely to have hypospadias . Furthermore, hypospadiac boys have been shown to have lower birth weights than their nonhypospadiac brothers . At 1.5 and 3 years of life, hypospadiac boys still tend to be smaller than average.
It is unclear which of these factors, if any, are causes and which are downstream effects. For instance, placental insufficiency, which is associated with intrauterine growth restriction, low birth weight and prematurity, has been proposed to lead to low levels of placental human chorionic gonadotrophin and fetal androgens ; this may lead to hypospadias [1-5]. Another hypothesis is that a generalized, persistent defect in androgen pathways may be responsible for hypospadias . The placental insufficiency and global androgen defect hypotheses are not mutually exclusive. Global androgen signaling deficits could conceivably hamper somatic growth of hypospadiac boys during childhood. We attempted to examine this theory by evaluating the birth and postneonatal height and weight of premature hypospadiac and non-hypospadiac boys.
Analysis of our raw data (non-age-matched cohorts) suggested that premature boys with hypospadias trended towards being lighter at birth as compared to premature boys without hypospadias. This is in agreement with prior findings [1-5]. This initial data analysis also indicated that non-age-matched, hypospadiac boys were lighter than their nonhypospadiac counterparts at time of surgery, even when using age-adjusted percentiles. However, our analysis of boys under the age of 2 years demonstrated no differences in either absolute or relative (weight-for-age percentile) weights.
The low birth weights and normal body weights at surgery for the hypospadiac cohort may denote rebound growth of hypospadiac boys after the neonatal period. In fact, Astbury et al. reported that among a cohort of ex-very-low-birth-weight infants examined at 2 years of age, 29% of patients were at the 10th percentile or lower for weight for age . Somewhat consistent with our data, Yu et al. reported that after 2 years of age some catch-up growth was observed .
In the current study we barred, from our nonhypospadiac cohort, patients with cryptorchidism, ventral skin deficiency of the penis, and/or penile torsion, webbing and curvature. Boys with cryptorchidism were excluded because this disorder has been postulated to have an endocrinologic basis. Penile anomalies, particularly curvature, are suspected by many to be related etiologically to hypospadias. Although we believe our rationale was conservative, definitive proof of these disorders having an endocrinologic cause remains to be established.
Our study featured limitations. This was a small, single-center, retrospective series. Consequently, there were insufficient patient numbers to examine any relationships between the severity of hypospadias and height and weight. The numbers of nonhypospadiac boys were particularly low after restricting the cohort to children under the age of 2 years (Table 3). Likely due in large part to the retrospective nature of the study, there were inadequate data to reliably study head circumference, another morphometric parameter known to be associated with prematurity and possibly hypospadias. Finally, our series may have been subject to selection bias. Specifically, we may have introduced bias by choosing to analyze boys who were not only admitted to our institution's neonatal intensive care unit for prematurity, but also subsequently underwent postneonatal hypospadias repair or circumcision. The patients comprising our cohort may or may not accurately reflect the morphometric characteristics of the general population of ex-premature boys with and without hypospadias.
Initial analysis of our data indicated that hypospadiac boys may be persistently smaller than boys without hypospadias. However, these differences were absent when we subsequently matched boys by age. We conclude that, based on our limited dataset, there is no evidence that hypospadiac boys exhibit persistent growth deficits during infancy and toddlerhood. These findings may be an argument in favor of the placental insufficiency hypothesis of hypospadias.
This work was funded in part by a Research Scholars Award from the American Urological Association Foundation and K08DK087895-01 from the NIDDK.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
This study was approved by our institutional review board.
Conflict of Interest Statement
D.R.R. is a paid instructor for Deflux injection (Oceana Therapeutics). None of the other authors have disclosures.