| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Thomas Jefferson University (J.L.R.) and Hahnemann University (H.K.), Philadelphia, Pennsylvania 19107; Pennsylvania State College of Medicine (D.R.), Hershey, Pennsylvania 17033; and Susquehanna Health Systems (D.R.), Williamsport, Pennsylvania 17701; and the Developmental Endocrinology Branch, National Institute of Child Health and Human Development, National Institutes of Health (P.F., G.B.C.), Bethesda, Maryland 20892
Address all correspondence and requests for reprints to: Judith L. Ross, M.D., Department of Pediatrics, Thomas Jefferson University, 1025 Walnut Street, Philadelphia, Pennsylvania 19107.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The major result of this study was the positive estrogen treatment effect on nonverbal processing speed and speeded motor performance in 12-yr-old TS girls. That motor performance would be slower in estrogen-deficient TS females is consistent with previous studies of the influence of estrogen on motor function. Estrogen replacement is thus the most likely explanation for the improved motor speed and nonverbal processing time in the estrogen-treated TS girls compared to that in the placebo-treated TS girls. Whether these findings will influence the psychoeducational outcome or quality of life of females with TS is not yet known.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This motor deficit in TS could be due at least in part to the absent ovarian estrogen production. The ovaries in normal girls produce very small amounts of estrogen before puberty at levels that are difficult to measure in the usual assays. Postmortem studies of normal ovarian development in the first 2 yr of life (11) indicate that females have increased ovarian estrogen production in the first year of life, analogous to the testosterone surge in the first 6 months of life in males (11). Girls with TS would therefore lack this surge, secondary to their dysgenetic ovaries. This very early estrogen deficiency may impact on later neurocognitive development. Additionally, production of estrogen by the ovaries increases significantly throughout puberty until adult levels are reached. In general, TS girls also lack these pubertal levels of estrogen and require estrogen replacement therapy in adolescence.
Estrogen alterations during the perinatal period and puberty have been shown to influence cognitive function and behavior in animal models. Animal studies have shown both structural (organizational) and activational effects of estrogen on subcortical nuclear regions such as the hypothalamic/preoptic area as well as forebrain regions that are related to behavior (12, 13). McEwen and colleagues have demonstrated specific morphological and chemical changes in neurons of the ventromedial nucleus of the hypothalamus, hippocampus, and basal forebrain in rats treated with estrogen that result in altered memory and motor speed (13). Oophorectomized rats treated with estrogen had higher activity levels on an exercise wheel compared to untreated control rats. Increased exploratory and wheel-running activities are observed in the higher estrogen phase of the estrous cycle in rats (14).
Girls with TS represent an intriguing model for the study of hormonal contributions to cognitive development. Previously, we demonstrated potential positive effects on mood in estrogen-treated TS girls, aged 12–16 yr (15). The goal of this study was to determine whether estrogen replacement therapy would reverse selected deficits in motor function and nonverbal processing speed, a measure of the time required to perform certain disparate nonverbal tasks, in adolescent girls with TS. The study used a randomized, placebo- controlled design. Additional goals were to contrast motor performance and nonverbal processing speed in estrogen-treated TS girls and age-matched control girls and to examine estrogen treatment effects on the lateralized asymmetry in right-handed subjects for various motor tasks.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The TS girls who participated in the study were drawn from an on-going, placebo-controlled, double-blind study of the effects of estrogen and/or GH on final adult height. The investigators remain blinded to most of the major end points of the study and received specific permission to analyze the results presented below from the Data and Safety Board that monitors this ongoing study. Therefore, only selected data can be shown. Girls were eligible to start the study between the ages of 5–11 yr and came from all parts of the United States. Children were randomized into 1 of 4 treatment groups: estrogen (ethinyl estradiol, 12.5–50 ng/kg·day), GH (Humatrope, 0.1 mg/kg, three times weekly, sc), the combination of estrogen and GH, or placebo. Girls aged 5–8 yr received 25 ng/kg·day, and girls aged 8–12 yr received 50 ng/kg·day. The girls were evaluated at 6-month intervals. If they experienced excessive bone age advancement or premature breast development, the dosage of estrogen was reduced by 50%. Approximately 20% of children in the 5–8 yr age range and 70% of children in the 8–12 yr age range had a 50% dose reduction during their participation in the study. The evaluation described below occurred when the girls reached the age of 10–12 yr. All girls in the larger growth study participated in this cognitive study when they reached the appropriate age. A total of 47 girls participated in this cognitive study. The number of girls is somewhat smaller for a subset of cognitive tests that was added later, in the second and third years of this study.
The diagnosis of TS was confirmed by karyotyping. No subject received any prior treatment with androgen or estrogen. The evaluation took place approximately 12–24 h after the preceding dose of estrogen or placebo. Treatment (estrogen or placebo) durations ranged from 1–7 yr (mean, 4.0 ± 2.1) depending on the age of the child at entry into the growth study and at the time of the cognitive evaluation. No child with a verbal intelligence quotient (VIQ) less than 70 was included, because these children are likely to have more diffuse developmental delay rather than the typical TS neurocognitive profile.
Serum estradiol was measured in each girl by standard RIA; however, levels were generally under the detection limit for the majority of subjects. As the detection limit varied from 5–35 pg/mL, no meaningful statistical analysis could be performed.
Data are also included for comparison purposes from a group of 10- to 12-yr-old, right-handed, female controls recruited from local Pennsylvania school districts. All were healthy and had heights and weights between the 5th and 95th percentiles. Only girls with a VIQ between 70–119 were included for the purpose of VIQ matching the control group to the TS group.
Statistics
One-way ANOVA was used to compare estrogen- and placebo-treated TS subjects and a group of VIQ- and age-matched female controls, aged 10–12 yr. Scheffe’s test was performed for post-hoc analysis where the one-way ANOVA comparison was statistically significant (P < 0.05). Approximately 50% of the children were receiving GH according to the study design. As GH has not been found to affect cognitive function in this group of TS girls (16), the four treatment groups were merged into two treatment groups: estrogen and placebo treated.
2 analysis was performed to compare the distributions of
race and karyotype between the TS and controls and between the two TS
groups, respectively. We used Pearson’s correlation coefficient (r
values) to examine the correlation between duration of treatment and
the outcome variables. All statistical tests were two-tailed tests, and
the results are presented as the mean ± SD.
P values for all comparisons were presented without
adjustment for multiple comparisons. Corrections for simultaneous
multiple comparisons can be performed using Bonferroni-type
adjustments. On that basis, only P < 0.005 would be
considered statistically significant.
Design and procedures
Subjects were administered a battery of cognitive tests,
including the Wechsler Intelligence Scale for Children–Revised
(WISC-R), and the various tasks described below. Several of the motor
tasks (finger tapping, pegboard, and pursuit rotor) were added 1 and 2
yr after the initial study started; therefore, not all subjects had
these tests. The numbers are indicated when subjects represent a subset
of the group numbers indicated in Table 1
. Socioeconomic status levels
were derived from the Hollingshead Two-Factor Index of Social Status
(17), based on education and occupation of parents. All testing was
conducted by two trained psychometricians at Thomas Jefferson
University Hospital and the NIH. They were both supervised for
standardization of methodology.
|
General cognitive ability. The WISC-R (18) is a widely used measure of intellectual functioning for children between the ages of 6.0–16.9 yr. The WISC-R was used rather than the WISC-III because some of the subjects were evaluated before publication of the WISC-III.
Motor function and handedness. Dominant vs. nondominant handedness was assessed before study participation because motor function has clear hemispheric lateralization. Handedness was determined by a modification of the Crovitz method (19). Children were asked to demonstrate which hand they use for eight activities. The hand chosen for six or more activities was defined as the dominant hand. If one hand was chosen for four or five activities only, the child was considered to be mixed dominant. Only right hand-dominant children were considered for the evaluation described below.
Nonspatial, repetitive motor tasks. Finger tapping (20): This test evaluates simple motor function and consists of a mechanical tapping key attached to a counter. Subjects tap with their index finger and are timed for five 10-s trials.
Physical and neurological examination of soft signs (Paness) selected tests (21): This test is a standardized, quantitative test of motor function. The dominant and nondominant hands and feet are tested individually. The time measurements are the time (seconds) required to 1) complete 20 taps by 4 sequential fingers to thumb (index finger first), 2) complete 20 taps of index finger to thumb, and 3) complete 20 taps of each foot to the ground.
Spatial motor tasks. Lafayette pegboard (22): This motor test consists of timing subjects as they place pegs in 25 round holes on a board as quickly as possible. Both the dominant and the nondominant hand are evaluated.
The pursuit rotor (23): This task measures hand-eye coordination and requires the subject to use a computer mouse on a pad to control the movement of the graphic mouse on the computer screen. The goal is to keep the graphic mouse within a target circle that moves around a circular track at the rate of 35 mm/s (Macintosh version). The average time spent off-target and an average score derived from the distance traveled and the time off-target are calculated for both dominant and nondominant hands.
Money street map (24): This task requires the child to determine left-right orientation while following a complex "street map." The test assesses visual-spatial and spatial orientation skills and has some motor demands. This test has both accuracy and time components.
Developmental test of visual-motor integration (25): This task evaluates the child’s ability to rapidly and accurately reproduce a series of simple figures after practicing conceptualizing and reproducing each figure.
Nonverbal tasks for measurement of processing speed. Matching familiar figures (26): This task is commonly used for assessing reflective vs. impulsive reactions to nonverbal, spatial-perceptual stimuli.
Test of facial recognition (27): This task requires the subject to match a face with a series of target faces. It assesses visual discrimination of the complex, nonverbal stimuli associated with facial features.
Motor-free visual perception test–revised (28): This task uses a forced choice paradigm that assesses different components of spatial ability, including closure, discrimination, memory, figure-ground discrimination, and mental manipulation.
Nonverbal processing speed. The processing time required for performing nonverbal tasks is important in itself. We would argue that this processing time is different from the accuracy of performing the task. We have previously demonstrated that the correlation between output and latency is poor. The nonverbal tasks timed included Money street map, Motor-free visual perception test, Matching familiar figures test, and Test of facial recognition.
Behavior questionnaires. Child Behavior Checklist (CBCL) (29): The CBCL was developed by Achenbach and colleagues as a standardized measure of academic and social competency as well as of behavior problems in children aged 4–18 yr. The parent version of the CBCL was completed by one or both parents for each subject. We focused on the three social competency subscales (activities, social, and school) that comprise the Social Competency Scale for purposes of this manuscript. The activities scale relates to participation in hobbies, sports, and other outside activities. The school subscale relates to parental assessment of school difficulties, grade retention, etc. Lastly, the social subscale relates to participation in outside organizations and contacts with friends. Higher scores generally reflect more frequent participation or activities.
Children’s Self-Concept Scale (Piers-Harris) (30): The Piers-Harris Self-Concept Scale is a self-report measure of self-concept. It consists of 80 statements that children rate as "yes" or "no" in relationship to how they feel about themselves. The total overall score and the physical subscale score were analyzed for the purposes of this manuscript. The reliability and validity of both the CBCL and Piers-Harris are well established, and both measures are widely used in child development studies. The overall score reflects global self-concept, with higher scores reflecting more positive self-concepts. We focused on the physical subscale, which reflects the child’s perception of her face and body. Higher scores generally reflect more positive self-image.
Self-Perception Profile (SPP) (31): This instrument measures five areas of perceived functioning: 1) social acceptance, 2) scholastic competence, 3) athletic competence, 4) physical appearance, and 5) behavioral conduct. Higher scores generally reflect more positive self-concepts. We focused on the athletic and physical scales for this analysis. Typical questions for the athletic competence subscale relate to the child’s perception of sports competence. The physical subscale related to the child’s perception of how she looks and to her comfort with her face and body. Higher scores generally reflect a more positive self-image.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
). In
addition, the controls were matched to the TS groups for age, SES,
race, and VIQ. Analysis of general cognitive function revealed no
differences in IQ scores for the estrogen- and placebo-treated TS
groups. The performance IQ was significantly higher in the control
group (by post-hoc test, P < 0.05; Table 1).
Simple repetitive motor tasks (Table 2)
The Turner females were compared to control subjects for the
following tasks: tapping and three tasks from the Paness. The
performance of the estrogen-treated vs. that of the
placebo-treated TS groups did not significantly differ (Table 2).
However, the sum of times required to perform certain of these tasks
was significantly less in the estrogen-treated TS group (by
post-hoc test, P < 0.05; Table 3).
|
|
).
Spatially mediated motor tasks (Tables 3 and 4)
The Turner females were compared to normal controls in the
following tests: Pursuit rotor, Nongrooved pegboard (Lafayette), Money
street map, and Test of visual-motor integration. The estrogen-treated
TS girls performed more quickly than the placebo-treated girls for the
spatially mediated motor task, the pegboard (by post-hoc
test, P < 0.05; Table 3). The controls differed only
from the placebo-treated TS group for this task (by post-hoctest, P < 0.05). Performance of the spatially
mediated motor task, the Test of Visual-motor integration, was
significantly better in the estrogen vs. the placebo TS
group (by post-hoc test, P < 0.05; Table 3). The control group differed significantly only from the placebo
group for this task. The sum of times required for performance of
certain motor tasks (Paness, pegboard, and pursuit rotor) was
significantly less in the estrogen-treated group (by
post-hoc test, P < 0.05; Table 3). In
addition, the sum of the times required for performance of four
nonverbal tasks (Nonverbal processing speed, Money street map,
Motor-free visual perception test, Matching familiar figures test, and
Test of facial recognition) was also less in the estrogen-treated group
(by post-hoc test, P < 0.05; Table 4). The
accuracy of performance for these tasks, however, was similar in the
two groups (Money, Table 2; other results not shown; see
Materials and Methods). The times required for performance
of nonverbal tasks in the estrogen-treated TS group approached the
results seen in the controls.
|
and 3) was also similar. No selective effect of
estrogen on the dominant vs. the nondominant hand was
observed.
Self-image questionnaires and physical activities (Table 5)
Both the Turner and control subjects answered three
questionnaires, including the SPP, the CBCL, and the Piers-Harris
Self-Concept Scale, to examine their sense of physical well-being and
competence. We focused on scales reflecting athletic competence,
physical self-image, and activities for purposes of this manuscript. A
significant finding was found only for the SPP athletic scale and the
CBCL Social Competency subscale, where only the placebo-treated TS
group differed from the controls (by post-hoc test,
P < 0.05; Table 5). There was a trend for improved
self-image with regard to SPP physical and Piers physical scales in the
estrogen-treated vs. the placebo-treated TS groups that did
not reach statistical significance.
|
We examined the correlation between treatment duration and motor/nonverbal task performance time in the estrogen-treated and placebo-treated groups separately. For many of the timed variables, the time required for performance diminished with increasing duration of estrogen but not with placebo therapy. For example, the r and P values for the Paness four-finger test (nondominant hand) were r = -0.49, P = 0.10 and r = 0.30, P = 0.32, respectively, for the two groups (estrogen and placebo). The trend was similar for motor sum (nondominant hand): r = -0.53, P = 0.08 and r = 0.44, P = 0.15, respectively.
In contrast, the values for the dominant hand Paness four-finger were r = -0.33, P = 0.30 and r = 0.11, P = 0.72; those for motor sum (dominant hand) were r = -0.13, P = 0.69 and r = -0.07, P = 0.83 for the two groups (estrogen and placebo), respectively. These results suggest that the correlation of estrogen treatment duration and nonverbal task time requirement was stronger for the nondominant hand. However, after the Bonferroni correction was applied, few of the values achieved statistical significance (data not shown).
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The estrogen-treated TS group had an interesting pattern of results for performance of timed spatial tasks. Although accuracy was not consistently changed, the speed of performance of these tasks consistently was affected by estrogen treatment. The sum of times required for processing the nonverbal tests, including Motor-free visual perception test, Test of facial recognition, Money street map, and Matching familiar figures, was significantly decreased in the estrogen-treated group. The estrogen-treated Turner girls’ speed on the spatial-dependent motor task, the pegboard, was significantly faster than that of the placebo-treated girls for the dominant (right) hand. Additionally, performance of the estrogen-treated group was better than that of the placebo-treated TS group on the Test of visual motor integration. In contrast, the two groups were similar for both time and performance for the Money street map. The sum of time required for performance of the motor tasks, including pegboard, Paness, and pursuit rotor, was also significantly less in the estrogen-treated TS group.
The relatively superior performance of the dominant (right) vs. the nondominant (left) hand was similar for the TS and control groups. This particular result is of interest, as it pertains to proposed neuroanatomical models of TS. Some investigators (32) have proposed that cognitive dysfunction in TS is related to right hemisphere abnormalities, whereas others (33) suggest that it reflects diffuse or generalized brain dysfunction. In studies of TS subjects, motor dysfunction for both hands was observed in tasks with greater spatial demands. For both speeded motor tasks and spatially loaded motor tasks, performance in the dominant relative to the nondominant hand was similar for the TS and control girls. However, there was some subtle suggestion of relatively greater estrogen duration effects on the nondominant vs. the dominant hand, perhaps implying relatively greater cumulative right hemisphere estrogen effects. This pattern of performance by the TS girls of motor tasks suggests that TS is neither a generalized brain disorder nor a generalized abnormality of the right hemisphere. The results appear to be consistent with dysfunction of the right posterior systems.
Slower motor performance in estrogen-deficient Turner females is consistent with the known influence of estrogen on motor function. To our knowledge, this study represents the first study to demonstrate that estrogen affects motor performance in this population. Intact ovaries in normal girls in this age range already produce low levels of estrogen (34). Estrogen appears to positively influence performance on many motor tasks studied in several models, including the menstrual cycle and postmenopausal women. Studies of the menstrual cycle have shown relatively improved performance on the finger-tapping and pegboard tests as well as rapid alternating movements, in association with the higher estrogen phase of the menstrual cycle (35, 36, 37). Hampson reported relatively facilitated verbal memory, speed and accuracy of articulatory performance, fine motor speed, and manual dexterity in the higher estrogen phase of the menstrual cycle (37). Surgically menopausal women treated with estrogen had improved memory, mood, digit span performance, abstract reasoning, and processing speed compared to women treated with placebo (38, 39). In addition, Denckla found that girls performed successive motor movements faster than boys at ages 5–10 yr (40, 41).
In our study, relatively few significant differences were present between the TS and control girls in tests evaluating simple motor function. Given the impact of estrogen on motor function in other studies, the failure to find a greater effect in our study was unexpected. We would have expected to observe more generalized, across the board changes in association with estrogen treatment. The placebo-treated group generally had slower performance times for simple motor tasks; however, none was statistically significant. These results, in the context of previous studies, are consistent with an influence of estrogen on motor speed that is sufficiently subtle so that it may be overridden by other aspects of the task, such as variance among subjects (fairly large) or sample size. In addition, the dose of estrogen used in this study (25–50 ng/kg·day) is approximately 1/10th or less of that usually used for estrogen replacement therapy in adult postmenopausal women.
In contrast to the absence of marked estrogen effects on simple motor task performance, the estrogen-treated TS girls were significantly faster in performance of a spatially mediated motor task, the pegboard. This was true for the dominant (right) hand. Clinical studies of adults with right vs. left hemisphere focal lesions showed that right hemisphere damage affected performance on the pegboard for both hands, whereas left hemisphere damage affected performance for only the right hand (22). Their results indicate that performance by either hand of motor tasks with greater visual-spatial requirements is determined by the integrity of the right hemisphere. We postulate that this is the explanation for the results of our study. Most explanations for the characteristic cognitive profile in TS argue that the spatial deficits are best explained by dysfunction of the right posterior hemisphere. Therefore, this right posterior dysfunction produces slower, spatially mediated motor function for both hands. This performance outcome appears to be improved by estrogen treatment. Such results are consistent with the premise that the estrogen effects in girls with TS are mediated via right hemispheric spatial-motor function.
In our study, there was some evidence of a decreased sense of athletic ability and physical self-image in the Turner girls compared to the controls. It has been suggested that psychological self-percepts may reflect cognitive function. Our results suggest that there may be some support for this perceived association. The Turner girls had some relative impairment of motor function and evidence of a slightly more negative athletic self-concept. Both improved in association with estrogen treatment; however, the results did not achieve statistical significance.
Animal and human studies indicate that estrogen may influence brain development in fetal life and through puberty. However, the mechanism of these effects is not known. Estrogen may function 1) transiently as a neuromodulator by potential mechanisms such as occupying receptors and initiating an enzyme cascade, modifying uptake of neurotransmitters, or altering neuronal electrical activity; 2) permanently by altering synapse formation and remodeling; or 3) by both mechanisms (42). There are some data to suggest how estrogen may work at least in part as a neuromodulator. Processing speed and motor speed appear to be dependent on cerebral dopaminergic systems, in particular the nigrostriatal system (43). Neurophysiological studies have shown that estrogen stimulates dopamine release in the nigrostriatal pathway (43). This provides a possible physiological explanation for estrogen acting as an activator of the motor system and thus for the faster nonverbal processing speed observed in estrogen-treated TS subjects.
We conclude that estrogen replacement therapy in adolescent girls with TS has a positive impact on nonverbal processing speed and certain aspects of the speed of motor performance. Whether these findings will influence the psychoeducational outcome or quality of life of females with TS is not yet known. The influence of estrogen on other aspects of the TS neurocognitive profile is also under investigation by our group.
|
Footnotes |
|---|
2 Current address: Eli Lilly Co., Lilly Research Laboratories,
Indianapolis, Indiana 46285-2055.
Received January 27, 1998.
Revised April 29, 1998.
Accepted May 19, 1998.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  L. Soriano-Guillen, J. Coste, E. Ecosse, J. Leger, M. Tauber, S. Cabrol, M. Nicolino, R. Brauner, the StaTur Study Group, J.-L. Chaussain, et al. Adult Height and Pubertal Growth in Turner Syndrome after Treatment with Recombinant Growth Hormone J. Clin. Endocrinol. Metab., September 1, 2005; 90(9): 5197 - 5204. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. L. Ross, G. A. Stefanatos, H. Kushner, C. Bondy, L. Nelson, A. Zinn, and D. Roeltgen The Effect of Genetic Differences and Ovarian Failure: Intact Cognitive Function in Adult Women with Premature Ovarian Failure Versus Turner Syndrome J. Clin. Endocrinol. Metab., April 1, 2004; 89(4): 1817 - 1822. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. L. Ross, D. Roeltgen, G. A. Stefanatos, P. Feuillan, H. Kushner, C. Bondy, and G. B. Cutler Jr. Androgen-Responsive Aspects of Cognition in Girls with Turner Syndrome J. Clin. Endocrinol. Metab., January 1, 2003; 88(1): 292 - 296. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. A. Quigley, B. J. Crowe, D. G. Anglin, and J. J. Chipman Growth Hormone and Low Dose Estrogen in Turner Syndrome: Results of a United States Multi-Center Trial to Near-Final Height J. Clin. Endocrinol. Metab., May 1, 2002; 87(5): 2033 - 2041. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Elsheikh, D. B. Dunger, G. S. Conway, and J. A. H. Wass Turner's Syndrome in Adulthood Endocr. Rev., February 1, 2002; 23(1): 120 - 140. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Saenger, K. A. Wikland, G. S. Conway, M. Davenport, C. H. Gravholt, R. Hintz, O. Hovatta, M. Hultcrantz, K. Landin-Wilhelmsen, A. Lin, et al. Recommendations for the Diagnosis and Management of Turner Syndrome J. Clin. Endocrinol. Metab., July 1, 2001; 86(7): 3061 - 3069. [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 |
