| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Editorial |
Division of Endocrinology, Children’s Hospital of Pittsburgh, Pittsburgh, Pennsylvania 15213
Address all correspondence and requests for reprints to: Selma F. Witchel, M.D., Division of Endocrinology, Children’s Hospital of Pittsburgh, 3705 Fifth Avenue, Pittsburgh, Pennsylvania 15213. E-mail: selma.witchel{at}chp.edu.
Polycystic ovary syndrome (PCOS) is a common heterogeneous familial disorder characterized by hyperandrogenism, oligo/amenorrhea, infertility, and insulin resistance/hyperinsulinemia. Familial clustering of hyperandrogenism, insulin resistance, and hyperinsulinemia in this syndrome is well described (1, 2, 3, 4, 5). Both genetic and environmental factors contribute to the pathogenesis of PCOS. Affected women have a higher risk to develop impaired glucose tolerance and type 2 diabetes mellitus. Although controversy exists regarding extent of the risk for cardiovascular disease, increased carotid artery intima-media thickness, dyslipidemia, obesity, metabolic syndrome, and the presence of inflammatory markers have been described among women with PCOS (6). The clinical features of PCOS often develop during the peripubertal years; premature pubarche due to premature adrenarche is considered to represent the initial manifestation for some girls (7).
Preventative care is a major aspect of pediatrics as illustrated by the development of vaccines to prevent childhood infectious diseases. Thus, premorbid identification of children at risk for PCOS and its associated disorders could be beneficial and enable therapeutic intervention(s). However, the earliest features of PCOS and its temporal progression are best characterized as murky. Prenatal factors such as maternal androgen excess, maternal hyperglycemia due to impaired glucose tolerance or gestational diabetes mellitus, or placental insufficiency may contribute to postnatal development of the metabolic derangements associated with PCOS. Another important factor involved in the pathogenesis of PCOS appears to be an inherent dysregulation of steroid biosynthesis (8).
In this issue, Legro and colleagues (9) hypothesized that "if familial clustering is characteristic of this disorder, then abnormal phenotypic characteristics of PCOS, including both reproductive and metabolic abnormalities, may be more likely in children of PCOS mothers than in those from mothers without PCOS." They theorized that comparison of these two groups of children might provide a window into the pathophysiology and ontogeny of PCOS. Using a case/control design stratified by gender and pubic hair Tanner stage, these investigators compared auxological parameters, timed urinary hormone excretion, and salivary insulin concentrations obtained from children of PCOS mothers (n = 32) to those from children of non-PCOS mothers (n = 38). The children ranged from 4–14 yr of age. Salivary samples were collected before and every 30 min after an oral glucose load for a total of 2 h. Two blood samples at 0 and 120 min were obtained from children over 8 yr of age who assented to venipuncture.
Why did these investigators choose to use these noninvasive techniques? For the past 10 yr, the National Institutes of Health, the Food and Drug Administration, and the U.S. government have directly encouraged pediatric participation in research studies. Participation in ethically sound human research studies follows the principles established by The Belmont Report, which include respect for persons, beneficence, and justice (10). Most national and international guidelines allow for children to participate in research upon parental consent. Many countries including the United States also require child assent (11). Whereas children constitute a vulnerable population entitled to special protections due to their limited decision-making capability, it has been suggested that they are able to assent when they understand the research and what participation entails (12).
Nevertheless, consensus among individual Institutional Review Boards and researchers regarding the assent process does not exist (13). For children, the level of risk and potential direct benefit to the subject are important considerations because federal regulations distinguish between research that does or does not hold direct benefit for the child participant. In general, most Institutional Review Boards will disallow research studies that involve multiple blood samplings that will not directly benefit the child participant because of the pain and distress associated with venipuncture (14). Thus, clinical research involving children is often caught between the proverbial rock and a hard place. To begin to tease out the ontogeny of PCOS in unaffected (but potentially at-risk) children, Legro and colleagues (9) creatively used noninvasive techniques that posed minimal risk to the participants. Findings of these studies included higher salivary insulin concentrations in Tanner 4–5 PCOS daughters, higher testosterone concentrations in Tanner 2–3 PCOS sons, and lower urinary LH excretion in Tanner 4–5 PCOS daughters.
The higher salivary insulin concentrations are consistent with the hypothesis that hyperinsulinemia and insulin resistance are constitutive early components of PCOS and are congruent with data reported by other investigators (15, 16). Comparisons of insulin concentrations during the peripubertal years can be confounded because puberty is associated with decreased insulin sensitivity and compensatory increased insulin secretion (17). One longitudinal study showed decreased insulin sensitivity, increased insulin concentrations, and preservation of hepatic insulin sensitivity during puberty (18).
Hyperglycemic and hyperinsulinemic-euglycemic clamp protocols provide more detailed and thorough assessment regarding insulin sensitivity, insulin secretion, glucose disposition, and insulin-stimulated glucose metabolism. The frequently sampled iv glucose tolerance test (IVGTT) has been successfully used in children; this test involves measurement and calculation of individual parameters of glucose disposal during an IVGTT, considers the hyperbolic relationship between glucose and insulin, and correlates reasonably well with euglycemic clamp determinations (19, 20, 21). Due to the invasiveness, need for multiple blood samplings, and financial burden of clamp studies and the frequently sampled IVGTT, a number of surrogate measures for insulin sensitivity and insulin resistance have been devised. The most widely used surrogate is the homeostasis model assessment (HOMA) index of insulin resistance (HOMA-IR) (22). However, most surrogate measures are calculated using fasting glucose and insulin concentrations, making them valid only in the presence of normal glucose tolerance and β-cell function (23). Surrogate indices derived from oral glucose tolerance test results such as the whole-body insulin sensitivity index and the insulin sensitivity index were reported to be useful in obese children (24).
One contribution of this manuscript is the use of salivary insulin concentrations. Previous studies have demonstrated that salivary insulin concentrations show a positive linear relationship with plasma insulin concentrations (25, 26). However, both the timing and the magnitude of the correlation between serum free insulin and salivary free insulin in these two distinct compartments showed individual variation in nondiabetic subjects (27). Whereas the use of salivary insulin concentrations in this study provides valuable information, it is not ideal. For example, the r value decreased from 0.72 at the beginning to 0.57 at 2 h, consistent with the previously described time lag (25, 27). Unfortunately, this methodology also limits their conclusions regarding insulin sensitivity in the younger children because more rigorous methods to measure insulin sensitivity might have detected hyperinsulinemia or insulin resistance in the younger children. Questions such as what are the molecular mechanisms responsible for insulin resistance in PCOS, how can insulin sensitivity be reliably measured in children, and does insulin resistance precede hyperinsulinemia or vice versa cannot be answered on the basis of this present study. Another unanswered question regarding the ontogeny of PCOS is whether hyperandrogenism precedes or follows the development of insulin resistance.
The finding of lower urinary LH excretion is counterintuitive because increased LH/FSH ratios are characteristic of hyperandrogenic states such as PCOS and nonclassical congenital adrenal hyperplasia. Timed urine samples enable noninvasive determination of integrated hormone excretion. However, timed overnight first morning void urine collections may have provided more information regarding the onset of puberty and LH secretory status because gonadotropin secretion is characteristically elevated at night early in the course of GnRH-dependent puberty. The lack of synchronization in the collection of urine samples from postmenarcheal girls may also have contributed to the unexpected finding of lower LH urinary excretion in PCOS daughters. Poor visualization of ovaries in the premenarcheal girls is not surprising in the absence of gonadotropin stimulation.
Notwithstanding previous reports of elevated dehydroepiandrosterone sulfate concentrations in brothers of women with PCOS (3), the significance of elevated testosterone concentrations in the Tanner 2–3 boys is unclear. The male phenotype still remains poorly characterized. Determination of testosterone concentration warrants discussion. For this study, serum total testosterone concentrations were determined using a commercial RIA (9). Serum free and weakly bound testosterone concentrations were measured after differential precipitation of plasma proteins after equilibrium with tritiated testosterone (9, 28). The urinary testosterone assay involved ether extraction and celite chromatography followed by RIA (9). However, direct competitive RIAs or chemiluminescence immunoassays often lack the sensitivity, specificity, and precision necessary to accurately measure the low testosterone concentrations found among children (29). Use of additional steps such as extraction and chromatography followed by immunoassays increases the accuracy but may still lack the sensitivity requisite to measure the low testosterone concentrations found in children. Yet, all is not lost. The use of liquid chromatography-tandem mass spectrometry in future studies may offer the sensitivity and specificity needed to measure testosterone in children and adolescent girls (29, 30).
This study highlights factors that confound investigations into the pathogenesis and ontogeny of PCOS such as phenotypic heterogeneity, changes in insulin sensitivity during puberty, ethical concerns regarding invasive studies in children, and small numbers of subjects. Using an imaginative noninvasive minimal-risk approach, these investigators demonstrated that salivary insulin concentrations correlated with serum insulin concentrations, salivary insulin concentrations can be used as a surrogate for serum insulin concentrations, and salivary insulin concentrations were higher in the older pubertal daughters of women with PCOS. Based on these data, Legro and colleagues (9) concluded that hyperinsulinism does not appear until the later stages of puberty. However, using more rigorous methods to measure insulin sensitivity and a larger number of subjects, they may have found evidence of hyperinsulinemia or insulin resistance in younger children. Overall, this study provides further support toward the inherited nature of PCOS and suggests early development of the associated metabolic and reproductive phenotypes. Longitudinal studies are needed to test their hypotheses that "hyperinsulinism precedes hyperandrogenism in children" of PCOS mothers and that "these children.... are likely at increased risk for the sequelae of PCOS compared with the normal population" (9).
Footnotes
Disclosure Statement: No relationships, financial or otherwise, exist that could pose a conflict of interest.
Abbreviations: IVGTT, Intravenous glucose tolerance test; PCOS, polycystic ovary syndrome.
Received March 13, 2008.
Accepted March 24, 2008.
References
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
