|
|
||||||||
1 Gynecological Endocrinology Unit, Division of Endocrinology, Hospital de
Clínicas de Porto Alegre, Avenida Ramiro Barcelos 2350, 90035-903,
Porto Alegre, Brazil.
2 Radiology Service, Hospital da Criança Santo Antônio, Avenida
Ceará 1549, 90240-512, Porto Alegre, Brazil.
3 Graduate Program in Clinical Medicine, Universidade Federal do Rio Grande do
Sul. Avenida Ramiro Barcelos 2350, 90035-903, Porto Alegre, Brazil.
Received April 23, 2001;
accepted after revision December 27, 2001.
Presented in part at the XIII World Congress of Pediatric and Adolescent
Gynecology, Buenos Aires, Argentina, April 2001.
Abstract
|
|
|---|
SUBJECTS AND METHODS. In this prospective study, 139 consecutive patients between 1 and 13 years old (mean ± SD, 6.0 ± 3.4 years) underwent sonography and hand radiography (for bone-age determination). Pubertal development was classified according to Tanner stages (prepubertal, 5.0 ± 2.7 years [n = 117]; pubertal, 11.2 ± 1.2 years [n = 22]). Uterine and ovarian longitudinal, transverse, and anteroposterior diameters were measured. Uterine fundalcervical ratio was determined. Ovaries were morphologically classified as homogeneous, paucicystic, multicystic, macrocystic, and presenting isolated cysts. A p value less than or equal to 0.05 was considered significant.
RESULTS. The uterus was identified in 96% of the patients (mean, 6.1 ± 3.4 years). One ovary was visible in 93% (mean, 6.2 ± 3.4 years), and both ovaries, in 81% (mean, 6.5 ± 3.3 years). Neither ovary was visualized in 10 girls (mean, 2.5 ± 2.2 years). Uterine parameters and ovarian volume were smaller in patients without thelarche (p < 0.0001). Mean ovarian and uterine size was smaller in girls until 8 years, intermediate between 9 and 11 years, and larger after 12 years (p < 0.0001). Chronologic age, bone age, and Tanner stage were correlated even before 7 years. Patients with and without thelarche presented different ovarian morphology (p = 0.01).
CONCLUSION. Uterine and ovarian growth was associated with age and puberty. Uterine length presented the best correlation with age. Multicystic ovaries seemed to be correlated with normal or premature pubertal stimuli.
|
|
|---|
Although sonography is routinely performed, there is a wide variation in the results of several studies concerning sonographic pelvic findings in children and adolescents. Uterine and ovarian growth patterns during infancy and puberty are not completely understood; also, no consensus exists about the normal measures and morphologic appearance of the ovaries. This may be attributed to the methodologic limitations of most studies, such as inclusion of both pubertal and prepubertal girls in the same analysis [4], small populations [5,6,7], absence of prepubertal girls younger than 8 years from the analyses [4, 8, 9], lack of statistical analyses [4, 5, 7], and use of formulae associated with logarithmic transformation, the application of which is not practical [10]. In addition, although cystic ovarian structures are also commonly observed on sonography, the classification of these structures is confusing and nonuniform. Investigators use either identical terms for different concepts or different terms for identical concepts, making the comparison among studies more difficult. For these reasons, physicians often do not know how to interpret sonographic results.
The aim of our study was to correlate ovarian volume, ovarian morphology, uterine volume, uterine area, and uterine length as observed on sonography with chronologic age, bone age, and breast pubertal status (Tanner scale [11]) in a sample of healthy girls.
|
|
|---|
Pubertal development was classified according to Tanner puberty stages [11]. The presence of thelarche (onset of breast development) was the criterion used to distinguish pubertal girls from prepubertal girls. One hundred seventeen patients were classified as prepubertal (Tanner 1 for breast development; age range, 1-12 years; mean, 5.0 ± 2.7 years) and 22 as pubertal (Tanner score 2-4 for breast development; age range, 9-13 years; mean, 11.2 ± 1.2 years).
Sonography, in all patients, was performed and interpreted by the same physician. All patients underwent examinations with a full bladder, but a Foley catheter was not used. Although transvaginal sonography offers better image quality, it is not used in children because of the dimensions of the equipment. Sonography was thus performed using an SL 1 Ultrasonographer (Sono Line; Siemens, Erlangen, Germany) with 3.5- and 5-MHz transducers. Uterine measurements included the following: longitudinal diameter (A), transverse diameter (B), anteroposterior diameter (C), and diameter at the fundus divided by the diameter at the cervix (fundalcervical ratio). Uterine volume was calculated in cubic centimeters using the ellipse formula A x B x C x 0.5233. Uterine area was obtained in square centimeters by multiplying longitudinal diameter by transverse diameter (A x B). Fundalcervical ratio was measured to investigate the existence of a correlation between uterine morphology and thelarche (fundalcervical ratio = 1 indicates prepubertal status; fundalcervical ratio >1 indicates pubertal status).
The volume of each ovary was calculated using the same formula as that for calculation of uterine volume. Because the volumes of the right ovary (0.7 ± 0.8 cm3) and left ovary (0.7 ± 0.7 cm3) were not significantly different (Wilcoxon's test, p = 0.87), the final volume considered was the average volume for both ovaries [(right ovary volume + left ovary volume) / 2].
The morphologic appearance of the ovaries was also assessed. Because in most adults the average dimensions of a mature follicle range from 15 to 20 mm [12], completely cystic structures measuring up to 20 mm were classified as follicles, and larger structures were classified as cysts. To simplify and unify the varied nomenclature used to classify ovaries in terms of morphology, we used the following definitions: homogeneous ovaries, absence of cysts or visible follicles [13]; paucicystic ovaries, up to five follicles measuring less than 10 mm in diameter [8]; multicystic ovaries, more than six follicles measuring less than 10 mm in diameter [8, 14]; macrocystic ovaries, one or more follicles measuring between 10 and 19 mm in diameter [4, 8, 15]; and ovaries presenting isolated cysts, one or more cysts measuring 20 mm or more in diameter [8].
After sonography, hand and wrist radiography was performed for the assessment of bone age. The radiologic tests of 73 girls were performed and interpreted by the same radiologist using the Greulich and Pyle scoring system [16]. Thirty-one radiographs (42%) were later reviewed by the same physician, with a good correlation (r = 0.911, p < 0.0001).
A significance level of p value less than or equal to 0.05 was used for the statistical analyses. Results were expressed as mean plus or minus standard deviation. For variables without a normal distribution, we used the following nonparametric tests: Wilcoxon's test for related samples, Mann-Whitney test for comparison between continuous variables in two groups, Kruskal-Wallis test for comparison among continuous variables in more than two groups, Fisher's exact test for comparison between categorical variables in two groups, and Spearman's correlation for testing ordinal variables. For variables with a normal distribution (bone age), the Pearson's correlation coefficient for testing continuous variables was used.
The curve estimation for uterine length, uterine area, and uterine and ovarian volume according to chronologic age was calculated using the linear, logarithmic, quadratic, cubic, and exponential models (8.0 for Windows; SPSS, Chicago, IL).
|
|
|---|
The best curve estimation for uterine length, uterine area, and uterine and
ovarian volume according to age was achieved with the cubic model. In the
cubic equation y = a + bx + cx2
+ dx3, y stands for uterine length, uterine area,
uterine volume, or ovarian volume; x stands for age in years; and
a, b, c, and d are constants calculated using the SPSS
software for each of the variables. Therefore, we used the following
equations:
![]() |
Uterine volume, area, and length and ovarian volume were smaller in patients without thelarche than in patients with thelarche (Mann-Whitney test, p < 0.0001): uterine volume was 1.8 ± 1.2 cm3 versus 8.1 ± 6.6 cm3 in patients without and with thelarche, respectively; uterine area was 2.7 ± 1.2 cm2 versus 7.1 ± 3.3 cm2; uterine length was 3.2 ± 0.7 cm versus 4.9 ± 1.1 cm; and ovarian volume was 0.5 ± 0.3 cm3 versus 1.7 ± 1.2 cm3. The group between 9 and 12 years old included pre- and postpubertal patients. In this age group, mean uterine length and ovarian volume were higher in patients with thelarche (Mann-Whitney test, p < 0.03) when compared with age-matched girls without thelarche.
Figures 1 and 2 show the increase in uterine and ovarian volume according to chronologic age and pubertal stage. Table 1 shows uterine and ovarian volume according to chronologic age. A progressive increase in these parameters can be observed with age. Figure 3 shows the distribution in percentiles 2.5, 50, and 97.5 of uterine length according to chronologic age. The patients with breast development are also identified in this figure. The cubic equation reveals different growth patterns for uterine area and length and uterine and ovarian volume. The mean ovarian and uterine size was smaller in girls up to 8 years, intermediate between 9 and 11 years, and larger after 11 years (Kruskal-Wallis test, p < 0.0001).
|
|
|
|
A statistically significant correlation occurred between chronologic age, bone age, and breast stage when compared with uterine length and area and uterine and ovarian volume, even in girls up to 7 years, as illustrated in Table 2. The association between sonographic variables and chronologic age is present both in the total sample (n = 133), in the prepubertal group (1-12 years old), and in patients up to 7 years old. Table 3 shows sonographic findings according to Tanner stages [11]. All the sonographic variables that were assessed increased with the progression of pubertal development (Spearman's correlation, p < 0.0001).
|
|
In all patients without thelarche, the fundal-cervical ratio was equal to 1 (n = 107). However, in patients with thelarche, 12 (60%) had a fundal-cervical ratio equal to 1, and eight (40%) had a fundal-cervical ratio greater than 1. Thus, fundal-cervical ratio equal to 1 was specific (100%) to identify prepubertal girls but less specific to identify pubertal girls.
Eighteen of 139 ovarian follicles were observed in girls starting at 2 years old, as shown in Figure 4. Ovarian morphology was significantly different in patients with and without thelarche (Fisher's test, p = 0.01). A homogeneous pattern was found in 105 of 117 patients without thelarche and in 16 of 22 patients with thelarche. A paucicystic appearance was observed in 12 of 117 girls without thelarche and in four of 22 with thelarche. Multicystic ovaries were found in only two patients with thelarche (12 and 13 years old). Macrocysts and isolated cysts were not observed among the studied patients.
|
|
|
|---|
In our study, a correlation was found between chronologic age, bone age, Tanner breast stage [11], and ovarian volume and uterine volume, area, and length, as also described in other studies [6, 8, 9, 18]. However, to our knowledge, our data are original in the sense that we show that this increase is also true for the group between 1 and 7 years old, which is the reference population in the study of patients with premature signs of puberty (isolated thelarche or central precocious puberty). In addition, the cubic equations calculated in our study can be used to predict uterine volume, length, and area, and ovarian volume in girls between 1 and 13 years. To our knowledge, there are no previous reports with these formulae.
Although some investigators [9, 10] argue that the ellipse formula is not the best method to calculate uterine volume because the uterus is not shaped like an ellipseand, in fact, its shape changes over timethis is still the most common method used in sonographic studies [2, 4, 6, 17,18,19,20,21]. Therefore, we chose to use the ellipse formula in our study. However, as discussed by others [9, 10], future studies should pay more attention to uterine length, which, in this study, had a better correlation with age than uterine volume.
Bridges et al. [22] suggest that uterine shape, expressed as fundal-cervical ratio, is a better marker of pubertal development than other uterine dimensions. According to our results, the fundal-cervical ratio provided reliable information concerning pubertal status in girls up to 7 years old (fundal-cervical ratio = 1); however, after this age, the fundal-cervical ratio did not differentiate prepubertal and pubertal girls.
Some authors state that ovarian growth is influenced by age and by the onset of puberty [5, 23], whereas others state that it is influenced only by age [4, 17]. In our study, the difference in ovarian volume and uterine length observed in 9- to 12-year-old patients with and without thelarche shows that both age and puberty affect uterine and ovarian size.
As previously described [24], there are two different stages of ovarian growth: the first one occurs at approximately 8 years old and results from adrenarche. The second stage occurs before the clinical manifestation of puberty and results from gonadotropin pulsatility. However, our results suggest that ovarian growth is influenced by age and by gonadarche.
Ovaries are dynamic organs that present both a stromal component, which increases discreetly from birth to maturity, and a gonadotropin-dependent follicular component [25]. Both the immature and the mature ovary have follicles in several stages of development or in atresia [26,27,28,29,30]. In addition, both types of follicles (atresic or not) may be cystic and can be observed from delivery until puberty on sonography [2, 4, 7, 10, 15, 17, 23, 27] or by histologic analyses [26, 27, 31]. This finding shows that follicular activity starts during intrauterine life [28]. During lactation and during the prepubertal and pubertal periods, higher levels of follicle-stimulating hormone are present [32]. Because microcysts are also frequently observed during these periods, we can postulate that the presence of ovarian follicles may be related to the increase in follicle-stimulating hormone levels.
In addition, we can postulate that the presence of microcysts at any age is physiologic, merely indicating the presence of anovulation and follicle-stimulating hormone stimulus. Thus, in infants or children, the usual ovarian pattern should be homogeneous or paucicystic. On the other hand, multicystic or macrocystic ovaries are usually associated with puberty or premature gonadal activity. Furthermore, in adolescents, ovarian volume and hyperechogenic stroma, rather than the isolated findings of microcysts, seem to be an important variable to differentiate multicystic from polycystic ovaries [33].
In our study, we used a simplified classification of ovarian morphology because the current classifications are not standardized, and their application in clinical practice is compromised by the variations in nomenclature. Therefore, we believe that a simplification in this terminology can be useful for clinicians. According to our classification, homogeneous ovaries were more frequent in patients without thelarche, and multicystic ovaries, in patients with thelarche. No differences were found in the frequency of paucicystic ovaries in patients with or without thelarche. Buzi et al. [8] reported similar findings.
In conclusion, the growth of the uterus and ovaries is proportional to chronologic age, bone age, and pubertal stage. Even in prepubertal girls, age influenced sonographic variables. The mean rate of uterine and ovarian growth was slower until 8 years old, intermediate between 9 and 11 years old, and higher after this age because of the progression of puberty.
The presence of some ovarian follicles is nonspecific and physiologic in infants, children, and adolescents. However, the presence of six or more follicles (multicystic ovaries) seems to be correlated with pubertal stimuli both in normal and precocious puberty.
Acknowledgments
We thank Kristhiane Domenico, Leopoldo Konzen, Leonardo Mylius, Luana
Silveira, and Rossana Marroni for their contributions to data collection and
the Graduate Research Group at Hospital de Clínicas de Porto Alegre for
providing editorial support.
|
|
|---|
This article has been cited by other articles:
![]() |
L. Ibanez, A. Lopez-Bermejo, J. Callejo, A. Torres, S. Cabre, D. Dunger, and F. de Zegher Polycystic Ovaries in Nonobese Adolescents and Young Women with Ovarian Androgen Excess: Relation to Prenatal Growth J. Clin. Endocrinol. Metab., January 1, 2008; 93(1): 196 - 199. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. de Vries, G. Horev, M. Schwartz, and M. Phillip Ultrasonographic and clinical parameters for early differentiation between precocious puberty and premature thelarche. Eur. J. Endocrinol., June 1, 2006; 154(6): 891 - 898. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. D. Herter, E. Golendziner, J. A. M. Flores, M. Moretto, K. Di Domenico, E. Becker Jr, and P. M. Spritzer Ovarian and Uterine Findings in Pelvic Sonography: Comparison Between Prepubertal Girls, Girls With Isolated Thelarche, and Girls With Central Precocious Puberty J. Ultrasound Med., November 1, 2002; 21(11): 1237 - 1246. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |