| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Section of Pediatric Endocrinology, The University of Chicago, Chicago, Illinois 60637
Address all correspondence and requests for reprints to: Elizabeth Littlejohn, University of Chicago Comer Children’s Hospital, Section of Pediatric Endocrinology, 5841 South Maryland Avenue (M/C 5053), Chicago, Illinois 60637. E-mail: ebaumann{at}peds.bsd.uchicago.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Objective: The objective was to determine the functional significance of polycystic-size ovaries (PSO) in healthy adolescents.
Design/Setting/Participants/Interventions: Healthy 11- to 18-yr-old postmenarcheal volunteers (n = 22) were recruited and divided into groups with normal size ovaries (VNSO; n = 10) or a polycystic-size ovary (VPSO; n = 12). They were secondarily compared with adolescents with polycystic ovary syndrome (PCOS; n = 8) matched for gynecological age and a PSO. All underwent GnRH agonist (GnRHag), oral glucose tolerance, and ACTH1–24 testing in our General Clinical Research Center.
Results: VPSO had a higher peak 17-hydroxyprogesterone (17PROG) response to GnRHag than VNSO (146 ± 14 ng/dl, mean ± SEM, vs. 85 ± 11; P = 0.008), as well as larger ovaries (13.3 ± 0.7 cc vs. 8.5 ± 0.8 cc). VPSO peak 17PROG was elevated (>137 ng/dl) in 42% (5 of 12). However, VPSO and VNSO androgen levels were similar, with the exception of one VPSO subject who had hyperandrogenemia and thus met criteria for PCOS. VPSO were similar to VNSO in LH, FSH, estradiol, and adrenal androgenic function. Although the VPSO group resembled the PCOS group in their 17PROG response to the GnRHag test, they differed in having significantly smaller ovaries and lower body mass index and in lacking evidence of peripheral androgen excess and of insulin resistance.
Conclusion: A PSO in asymptomatic adolescents seems typically to be a normal variant. However, about half have a subclinical PCOS type of ovarian dysfunction; it is unknown whether this indicates a genetic carrier state or a risk for anovulation.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
It is not clear whether the adult criteria for a polycystic ovary pertain to adolescents because the number of large antral follicles reaches its maximum around the time of menarche (9, 10). The criteria for an adult polycystic ovary are a polycystic-size ovary (PSO; volume > 10.5 cc) or a polyfollicular ovary (>12 follicles per ovary) (11). In recruiting healthy adolescent volunteers for our studies of adolescent PCOS, we found a high prevalence of a PSO. We report here the functional significance of the PSO in healthy adolescents.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Healthy 11- to 18-yr-old volunteers (n = 22) were recruited by advertisement: menarche had occurred within 0.6–4 yr, they were in the early follicular phase of cycles that were eumenorrheic by adult standards (those within 1 yr of menarche had monthly cycles since onset), and they did not have hirsutism or clinically significant acne or acanthosis nigricans. An adolescent PSO was defined traditionally as an ovary with a volume greater than 10.8 cc (3, 12). Volunteers fell into groups with normal-sized ovaries (VNSO; n = 10) or with a PSO (VPSO; n = 12); none had polyfollicular morphology. They were secondarily compared with a group of adolescents diagnosed with PCOS (n = 8) on the basis of otherwise unexplained menstrual dysfunction, elevated plasma free testosterone, and a PSO without polyfollicular morphology, thus meeting all Rotterdam diagnostic criteria (11), matching for gynecological age and the presence of a PSO lacking a polyfollicular appearance. These studies were approved by the University of Chicago Institutional Review Board and were performed after obtaining assent of the patients and consent of the parents.
Methods
All subjects were admitted to the University of Chicago General Clinical Research Center. Examination included determination of body mass index (BMI). After obtaining a baseline early morning fasting blood sample for steroids, an oral glucose tolerance test was performed according to American Diabetes Association guidelines (13), during which blood was taken at 30-min intervals for pooling to obtain mean androgen and steroid intermediate levels (14). Dexamethasone (DEX) 0.25 mg/m2 was then given at 1200 h daily. At 1600 h, a basal control sample was drawn; then ACTH1–24 1.0 µg/1.7 m2 was administered iv (15, 16), and blood samples were obtained 15 and 30 min later. At 1700 h, blood sampling began at 15-min intervals for 1 h to pool for gonadotropins; then another basal steroid sample was drawn, and leuprolide acetate (GnRHag) 10 µg/kg was administered sc (17), followed by sampling at 30- to 60-min intervals for 4 h, and at 18, 20, 22, and 24 h after injection. Real-time pelvic ultrasound imaging was performed by the abdominal route using an Acuson Sequoia with a 4 Mhz transducer (Acuson, Mountain View, CA), and a bone age radiograph was obtained.
Assays
Plasma total testosterone, estradiol, dehydroepiandrosterone sulfate (DHEAS), and cortisol (Diagnostic Products Corporation, Los Angeles, CA) were measured by commercially available kits and free testosterone and SHBG by binding assay as previously reported (17, 18). Steroid intermediates—17PROG, androstenedione, 11-deoxycortisol, 17-hydroxypregnenolone, and dehydroepiandrosterone (DHEA)–were measured by previously reported RIAs with sensitivities averaging 25 ng/dl (19). Values below this level were interpolated for purposes of statistical analyses. LH and FSH were measured by immunochemiluminometric assay kits (Delphia, Wallach, Finland). Plasma glucose was measured using a glucose analyzer (YSI Model 2300 STAT; Yellow Springs Instruments, Yellow Springs, OH). Serum insulin was assayed by a double-antibody technique in which the cross-reactivity of proinsulin in the RIA for insulin was approximately 40% (20). Insulin resistance index was determined by homeostatic model assessment (HOMA-IR) (21). Bone age was determined by a modification of the Gruelich-Pyle method.
Data analyses
Comparisons between groups were made using Student t tests. The prevalence of glucose abnormalities was compared between groups by Fisher’s exact test (22). Correlations were determined from linear regression analysis. Results are reported as mean ± SEM unless otherwise noted. P values < 0.05 were considered statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
). In two thirds of VPSO subjects, only one ovary met PSO criteria, although the other ovary was also significantly larger (VPSO 9.8 ± 0.9 cc vs. VNSO 6.8 ± 0.7 cc; P < 0.01).
|
). The VPSO had a slightly but significantly lower HOMA-IR than VNSO, although one had impaired glucose tolerance (Table 1
).
|
and 2). No other differences were apparent between the two volunteer groups during the GnRHag test. ACTH testing showed a marginally higher post-DEX basal 17PROG in VPSOs, but no significant differences from VNSOs in the responses of cortisol or steroid intermediates.
PCOS patients had elevated baseline free testosterone levels by definition; baseline androstenedione and DHEAS also were higher, and SHBG was lower than in volunteers (P < 0.05) (Table 1). This PCOS group was matched to the VPSO group for the presence of a PSO; their maximum ovarian volume ranged from 10.9–36.3 cc, which overlapped that of the VPSO group although exceeding theirs in three eighths of cases. Post-DEX testosterone, 17PROG, and androstenedione were also higher (Table 2 basal values); specifically, post-DEX free testosterone was elevated in seven of eight (
9 pg/ml), total testosterone in six of eight (30 ng/dl), 17PROG in six of eight (47 ng/dl), and androstenedione in one of eight (158 ng/dl). In response to GnRHag, PCOS patients had a higher early LH response (30–60 min mean = 58 ± 12 U/liter) than either volunteer group (VNSO 21 ± 4; VPSO 22 ± 6). However, their 17PROG peak responses to GnRHag were similar to those of the VPSO group, whereas higher than those of the VNSO group; the stimulated 17PROG levels were elevated (>137 ng/dl) in half of the PCOS group and half of the VPSO group. Other than testosterone being higher in PCOS after GnRHag, the responses in PCOS were not significantly different than in volunteers. The PCOS group had higher post-ACTH peaks of androstenedione than volunteers, and two had higher DHEA peaks than any volunteer. The PCOS group had significantly higher BMI and a higher prevalence of glucose abnormalities than the volunteers as a whole (P = 0.01); two had impaired glucose tolerance, and two had diabetes mellitus (one type 1).
Linear regression analysis revealed no overall correlation between gynecological age and ovarian size. However, there was an indication that ovarian size may normally maximize about 1 yr after menarche because there was a significant correlation between ovarian size and gynecological age among the healthy volunteers (r = 0.60; P < 0.01), that became insignificant upon removal from the analysis of the two least gynecologically mature girls. Gynecological age did not correlate with free testosterone or 17PROG peak response to GnRHag in either volunteers or PCOS patients.
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Our group of asymptomatic adolescent volunteers with a PSO had 17PROG responses to GnRHag similar to the group of adolescents with PCOS and a PSO. However, they had no other features of PCOS. Most notably, they had absolutely no evidence of the excessive androgen secretion that characterizes PCOS, such as elevated baseline or post-DEX androgen levels (8).
Polycystic ovaries seem to be heritable as an autosomal dominant trait that is not necessarily associated with evidence of PCOS (1, 2). Therefore, a PSO, which is a type of polycystic ovary, appears to represent a carrier state and/or a risk factor for PCOS. The limited available longitudinal data in adolescents suggest that once a PSO develops, it typically persists (12). Polycystic ovaries seem to be a marker for excessive thecal androgen production: isolated theca cells from both symptomatic and asymptomatic subjects with polycystic ovaries secrete similar excessive amounts of 17PROG and androgens, compatible with a primary thecal defect in androgen regulation (5, 23, 24). However, asymptomatic adults have little, if any, elevation of plasma androgens (5, 6, 7). Why this thecal androgen excess is not manifest as hyperandrogenemia is unclear. One possibility is that normal granulosa cells produce factor(s) that restrain thecal androgen production in the intact follicles of such women. One such factor may be activin; within the ovary, activin inhibits ovarian androgen production (25).
Some studies have found that insulin resistance is a concomitant finding with the asymptomatic polycystic ovary in adults (6, 7). However, in agreement with another adult study (26), we did not find evidence for insulin resistance among our regularly menstruating adolescent volunteers with a PSO. Indeed, the VPSO group as a whole had a slightly, but significantly, lower HOMA-IR than the VNSO group; thus, they seemed to be more insulin sensitive. Only one subject (8%) in the VPSO group had impaired glucose tolerance, and her only other feature of PCOS was 17PROG hyperresponsiveness to GnRHag testing.
The limitations of this study should be considered. Our volunteers may not be representative of the general population. Their prevalence of a PSO was greater than reported in the general population (3), although in line with that reported in some other studies of adult volunteer populations (5, 6, 7). Thus, it is possible that a subtle, unreported concern may have led them to volunteer, yet was not clinically evident. However, other than in the ovarian findings, the VPSO group differed from the VNSO group only in having a slightly lower HOMA-IR. It is possible that the greater insulin sensitivity of the VPSO group was a consequence of their tendency to have a lower BMI and to be slightly older; neither of these was statistically significant, however. This lower HOMA-IR is contrary to the expectation if their ovarian findings were related to the insulin resistance of PCOS. Our VPSO group differed from our adolescent PCOS group, which was matched for gynecological age and the presence of a PSO, in the expected ways: PCOS patients had higher BMI and androgen levels, and lower SHBG levels. They also had a higher prevalence of hyperglycemia than the volunteers as a whole. The matching is imperfect, however, because the PCOS group was significantly older in chronological and bone age. This finding may indicate a previously unrecognized delay in menarche in adolescent PCOS.
In summary, the ultrasonic characteristics of the ovary must be taken into consideration when establishing norms for ovarian function testing. Most asymptomatic adolescents in whom a PSO is detected incidentally are normal. However, about half carry a subclinical PCOS type of ovarian dysfunction. It is unclear whether this represents a subgroup that is heterozygous for a polycystic ovary gene or is at risk for developing anovulatory cycles or PCOS later.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
M.M., R.R., and E.L. have nothing to declare.
First Published Online August 8, 2006
Abbreviations: BMI, Body mass index; DEX, dexamethasone; DHEA, dehydroepiandrosterone; DHEAS, DHEA sulfate; GnRHag, GnRH agonist; HOMA-IR, homeostatic model assessment of insulin resistance; PSO, polycystic-size ovary; PCOS, polycystic ovary syndrome; 17PROG, 17-hydroxyprogesterone; VNSO, volunteer with normal-size ovaries; VPSO, volunteer with a PSO.
Received April 19, 2006.
Accepted July 28, 2006.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  S. K. BLANK, K. D. HELM, C. R. MCCARTNEY, and J. C. MARSHALL Polycystic Ovary Syndrome in Adolescence Ann. N.Y. Acad. Sci., June 1, 2008; 1135(1): 76 - 84. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. L. ROSENFIELD Improving Balance in Regulatory Oversight of Research in Children and Adolescents: A Clinical Investigator's Perspective Ann. N.Y. Acad. Sci., June 1, 2008; 1135(1): 287 - 295. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. L. Rosenfield Identifying Children at Risk for Polycystic Ovary Syndrome J. Clin. Endocrinol. Metab., March 1, 2007; 92(3): 787 - 796. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
