This Article Abstract Full Text (PDF) Submit a related Letter to the Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Finkelstein, J. W. Articles by Kulin, H. E. Search for Related Content PubMed PubMed Citation Articles by Finkelstein, J. W. Articles by Kulin, H. E.

Effects of Estrogen or Testosterone on Self-Reported Sexual Responses and Behaviors in Hypogonadal Adolescents¹

Departments of Pediatrics (J.W.F., M.R.D., H.E.K.), Biobehavioral Health (J.W.F., E.J.S.), Health Evaluation Sciences (V.M.C., S.J.K.), Human Development and Family Studies (E.J.S., J.S.), Pathology (L.M.D.), and Psychology (L.S.L.), Pennsylvania State University College of Medicine, Hershey, Pennsylvania 17033; and Colleges of Health and Human Development, and Science, Pennsylvania State University, University Park, Pennsylvania 16802

Address all correspondence and requests for reprints to: Jordan W. Finkelstein, M.D., Pennsylvania State University, East 315 Henderson Building, University Park, Pennsylvania 16802. E-mail: jwf3{at}psu.edu

	Abstract

Top Abstract Introduction Subjects and Methods Results Discussion References

 
The purpose of this study was to investigate the effects of administration of sex steroids on self-reported sexual responses and behaviors in hypogonadal adolescents. We used a randomized, double blind, placebo-controlled, cross-over, clinical trial as the experimental design. The subjects were 39 boys and 16 girls with delayed puberty. We treated girls with oral conjugated estrogen and boys with testosterone enanthate in 3 dose levels intended to simulate early, middle, and late pubertal levels. We administered a modification of the Udry sexual behavior questionnaire after each 3-month placebo and treatment period to detect the effect of sex steroids on self-reported sexual behaviors and responses. We employed a strict intent to treat statistical analytical model. The data showed significant effects of the administration of testosterone to boys causing increases in nocturnal emission and touching behaviors at the mid- and high doses. No other treatment effects on sexual behaviors or responses were seen in boys. For girls, there was a significant increase in necking caused by the administration of estrogen only at the late pubertal dose. No other treatment effects on sexual behaviors or responses were seen in girls. We noted some gender differences for thinking about sex, sexual "turn-on," and the nature of sexual behavior. The administration of physiological doses of sex steroids to boys or girls with delayed puberty have few effects on sexual behaviors and responses.

	Introduction

Top Abstract Introduction Subjects and Methods Results Discussion References

 
SEXUAL behaviors in children occur throughout childhood and are usually self-exploratory and self-stimulatory. Many prepubertal children also engage in exploratory sexual behaviors with prepubertal children of the opposite sex, but they do not experience these behaviors in the same way as individuals who are physically sexually mature (1). At puberty, there is an increase in the frequency, nature, and type of sexual behaviors and responses. Udry has suggested that sex steroids may play a significant role in the development of sexual behaviors during puberty in boys (2) and girls (3). However, these studies were not longitudinal and cannot be used to imply causality. No data exist evaluating the effects of administration of sex steroids on the development of sexual behaviors and responses in adolescence, including the development of nonintercourse (outercourse) sexual behaviors.

The goal of the present study was to investigate the role of sex steroids on the development of various self-reported sexual behaviors and responses by studying hormone-deficient adolescents to whom we could ethically administer and withdraw sex steroid therapy while they undertook sequential behavioral assessments. We posed two questions in the investigation. 1) Would administered sex steroids affect sexual behaviors and responses? 2) Would the effects differ for boys and girls?

	Subjects and Methods

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Subjects

All patients referred to the Pediatric Endocrine Clinic of the Milton S. Hershey Medical Center with complaints of pubertal delay were potential enrollees. Only 2 candidates refused enrollment. We did not include others if their history involved chronic illness, intellectual deficits, or ingestion of medication that might affect behavior. Of 58 enrollees, 3 subjects withdrew from the study before visit 1 because they found the extra time required for behavioral testing burdensome. Six individuals started the protocol at middose levels based on relatively advanced age and/or prior hormone therapy. We entered 6 individuals into the study protocol in a nonrandomized fashion, and we do not include their data in the primary analysis.

Data from 55 patients contributed to the current report. The actual numbers and genders of participants who provided data for the sexual behaviors and responses at each treatment period are shown in Table 1. Not every subject completed the survey at all visits, which is why the number of subjects in Table 1 is not the same at all visits. The mean age at first visits of all subjects was 13.6 ± 1.85 (SD) yr (range, 10–19 yr). Sixteen individuals had primary gonadal disease (including 10 girls with gonadal dysgenesis); 11 had documented gonadotropin deficiency, and 28 (26 of whom were boys) were classified as having constitutional delay of puberty.

Experimental design

The experimental design was a randomized, double blind, placebo-controlled, cross-over trial. We randomized subjects to initially receive either hormone exposure (arm Y; Table 2) or placebos (arm X; Table 2). We administered sex steroids in three dose levels approximating early, middle, and late pubertal amounts of hormone exposure. The girls took oral conjugated estrogen (Premarin) daily at 0.15 mg (low dose), 0.3 mg (middose), and 0.6 mg (high dose). The boys took testosterone by im injection every 2 weeks at 25 mg (low dose), 50 mg (middose), and 100 mg (high dose). We accompanied each 3-month hormone exposure period by 3 months of placebo administration. They achieved successful adherence to this schedule because the median treatment/placebo interval ranged from 2.9–3.0 months. The protocol lasted for 21 months for all subjects. The 8 boys and 10 girls with permanent hypogonadism (ovarian or testicular agenesis/dysgenesis or hypogonadotropic hypogonadism) were continued during treatment at the high dose for an additional year to determine whether there were significant changes in sexual responses or behaviors to longer term treatment.

Hormone analyses of blood for androgens and estrogens and of urine for gonadotropins were performed in each patient at each clinic visit. Measurements included blood levels of total testosterone and estrone sulfate, and urine levels of LH and FSH. The Core Endocrine Laboratory of the Milton S. Hershey Medical Center performed all hormone tests using standard RIA (4, 5, 6). Hormone measurements served to monitor exogenous vs. endogenous sex steroid concentrations (i.e. the on and off treatment periods) and to assess compliance with treatments. Pill and prefilled syringe counts were also used to ensure protocol compliance.

Behavioral testing

We administered a modification of the Udry Sexual Behavior Questionnaire (2, 3) to subjects to assess the effects of sex steroids on self-reported sexual behaviors and responses. The instrument permits an adolescent to rate how much a statement describes his or her own sexual thoughts or feelings, fantasies, dreams, self-stimulatory sexual behaviors, and sexual behaviors with members of the opposite sex. One of two psychologists administered the instrument as a structured interview to be sure that subjects understood all of the sexual behavior terms used. However, after one or two interviews, the subjects completed the written questionnaire by themselves with the psychologist available to answer questions.

Statistical analysis

All clinical, laboratory, and behavioral data were entered into an Ingles database and verified. Sample size calculations revealed that for 80% statistical power and a 5% significance level, 16 male and 16 female subjects would be needed to detect a difference of 1 SD in any of the sexual response or behavioral variables. We used the ² test to evaluate differences in the frequency of sexual responses or behaviors between boys and girls.

As the responses to questions asking whether the subject had experienced specific sexual events during a treatment or placebo period was bivariate (yes or no), an ANOVA model could not be used to estimate treatment effects. Instead, we used a generalized estimating equations model (7) that applies a procedure similar to the ANOVA model but can be used with discrete data. This model compared the number of subjects experiencing specific sexual behaviors and responses during placebo (pooled across all placebo periods) to the number of subjects experiencing the same behaviors during the three levels of treatment by calculating odds ratios and confidence intervals. If the confidence intervals did not include the value 1.0, statistically significance differences were present.

Odds ratios were calculated adjusting for the following model effects. Treatment is the main effect in the design, as it tests the results of hormone administration. Treatment contains the four categories of placebo, low, middle, and high doses. The cross-over design allows this parameter to be a within-subject test. Having each subject act as his or her own control yielded a more powerful test than if comparisons were made between subjects in a parallel design. We also included the following effects in the model: sequence (receiving the drugs in the order treatment-placebo or placebo-treatment), sex (effect of gender), carryover (accounts for any residual or carryover effects of having received prior therapy, either on or off hormone), visit (accounts for changes that take place over time that cannot be accounted for by the treatment), and sex by treatment interactions within subject (allows for the effect of treatment to vary between the sexes).

	Results

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Results of the administration of sex steroids to boys and girls in this study showed that statistically significant effects of treatment on sexual behaviors and responses occurred for nocturnal emission and touching behaviors in boys and for necking in girls.

Boys

Nocturnal emission occurred at the low testosterone dose in 6 subjects, at the middle dose in 15 subjects, and at the high dose in 23 boys. There was a significant (P < 0.05) increase in the number of boys reporting nocturnal emission for only the middle dose of testosterone (50 mg) compared with that during the placebo periods (Table 3). The effect was not statistically significant at either the high or the low dose vs. the placebo.

As the largest number of boys (n = 26) in this study had constitutional delay in growth and development, we monitored their secretion of endogenous testosterone during placebo periods. A secondary data analysis was performed, eliminating the sexual response and behavior data of those boys after the time when their endogenous testosterone concentrations (assessed at the end of placebo periods) exceeded 50 ng/dL. The results for the boys whose endogenous testosterone remained below 50 ng/mL and including those four boys who were not initially randomized were essentially unchanged at the low and middle doses. At the high dose, there was a significant increase in necking (odds ratio, 2.54; 95% CI, 1.12,5.76) based on the secondary analysis.

Girls

A significant increase in necking (odds ratio, 5.29; 95% CI, 1.52,18.53) was seen only at the high dose (0.6 mg; Table 5). No other significant treatment effects at any dose for any other sexual behavior were seen including the two nonrandomized girls in the secondary analysis.

Boys engaged in more advanced sexual behaviors than girls (Tables 4 and 5). For instance, only 1 girl reported any activity beyond necking at all doses, whereas several boys engaged in all behaviors, including intercourse, by the middle dose. These gender differences were also found when subjects were asked "how often do you think about sex?" The range of responses for boys reporting this activity more than several times a week was 19% (5 boys) before treatment and at the low dose and rose to a high of 31–41% (9 and 13 boys) after the middle and high doses, respectively. No girls reported sexual thoughts before treatment. One girl experienced sexual thoughts at the low dose, and only two did so at middle and high doses.

Median hormone values for subjects in this study are shown in Table 7. The increases in concentrations of testosterone (in the male) and estrone sulfate (in the female) along with decreases in gonadotropins suggested that subjects were receiving the appropriate amounts of hormone or placebo material. The reciprocal relationship between sex steroids and gonadotropins confirmed that exogenous sex steroids suppressed endogenous gonadotropins.

	Discussion

Top Abstract Introduction Subjects and Methods Results Discussion References

The first question formulated in this study is whether administered sex steroids can influence self-reported sexual responses or behaviors. There was a significant increase in the number of male subjects experiencing nocturnal emissions and participating in sexual touching behaviors at the middle dose and in kissing behaviors at the high dose of testosterone administration. There was a significant increase in the number of female subjects experiencing necking behaviors at the high dose. These findings suggest that there may be an effect of testosterone or conjugated estrogen at least for these sexual behaviors and responses. The relation of the reported responses and behaviors we observed to the dose of administered hormones suggests that a threshold may exist below which a response or behavior will not appear. Once that dose has been reached, further increases in the dose will not affect further increases in the response or behavior (9). The possibility that the subjects’ exposure to sex steroids was not long enough prompted us to reevaluate behaviors and responses in those subjects in whom continued treatment was clinically indicated (those with permanent hypogonadism). No significant changes were detected after 1 yr of continuous treatment at the high dose.

Testosterone has been reported to play a role in the frequency of nonerotic or reflex erections, including nocturnal penile erections and for the production of seminal fluid (10). There is no literature on the role of testosterone on nocturnal emission at any age. It seems reasonable to speculate that as testosterone is involved in causing nocturnal erections and production of seminal fluid, testosterone may also cause increases in nocturnal emission. It seems possible that nocturnal emissions may be induced/facilitated by factors other than hormones that prime the adolescent for this physiological response, as not all boys experienced nocturnal emission but all boys did receive testosterone.

Udry et al. (2) reported that nocturnal emissions, masturbation to orgasm, thinking about sex, sexual "turn-on," and coitus were all significantly correlated with a calculated free testosterone index in a nonclinical group of Caucasian boys in the 9th and 10th grades.

In another report by Udry et al. (3), a nonclinical sample of Caucasian girls in the 9th and 10th grades showed significant correlations between androgens (free testosterone index, androstenedione, and dehydroepiandrosterone sulfate) and thinking about sex, sexual turn-on, and masturbation, but not intercourse. They reported a significant effect of physical pubertal maturation only for masturbation. The only effect of estrogen was a negative correlation with sexual turn-on. These two cross-sectional, observational studies suggest a significant correlation between the concentration of endogenous sex hormones and sexual behaviors and responses, but causation cannot be inferred.

Boys with true precocious puberty are often reported by parents to engage in masturbatory behaviors that decrease if testosterone-lowering treatments are given. This suggests that masturbation may be testosterone dependent, and that other factors may play a limited role in the inhibition of sexual behaviors and responses, at least for precocious puberty in boys.

Girls with precocious puberty are not thought to exhibit increases in sexual behaviors during the ages (<8 yr) when their sexual development is considered precocious (11). However, girls with precocious puberty may exhibit statistically significantly earlier engagement in sexual behavior (2 yr) at the usual time of adolescence for kissing, necking, breast petting, masturbation, and intercourse than controls (12).

The literature describing the relationship between hormones and sexual behavior or response in adults is controversial, but a recent review (11) suggests that in males, androgens are necessary, but not sufficient, for sexual desire. Supraphysiological doses of testosterone given to healthy adult men may increase sexual interest or arousability, but do not increase the frequency of overt sexual behaviors, i.e. masturbation or intercourse (13). In adult males with experimentally induced hypogonadism, there was a significant decrease in the frequency of sexual desire, sexual fantasies, and intercourse after 4–6 weeks in the hypogonadal state. Sexual behaviors and responses returned to normal during the administration of testosterone at a dose of 50 mg/week (14). All of these studies support the concept that testosterone is necessary, but not sufficient, for sexual behaviors and responses to occur, and that other factors play important roles in sexual behaviors and responses in adult males.

The differences between boys’ and girls’ sexual turn-on reports in the current study are similar to what has been reported for adults. Boys and men are aroused by visual erotic stimuli, whereas girls and women are aroused more by mental processes (14). Boys engaged in more advanced behaviors than girls (15). These gender differences between boys and girls suggest that androgen administration to males produces different effects on sexual behavior than estrogen administration to girls. However, as only touching and kissing behaviors were affected by hormone administration, these conclusions should be interpreted cautiously.

Four methodological issues should be noted. First, there is the possibility that a type 2 error may exist to explain the paucity of significant hormone effects on sexual behaviors and responses, especially among the girls. For our sample size, the probability of a type 2 error was 20% for an effect size of 1 SD.

Second, the duration of treatment was 3 months at each dose level alternating with placebo. This may not have been adequate exposure for some of the behaviors and responses. Therefore, we treated those subjects affected with permanent hypogonadism continuously for 1 yr with the high dose and found no additional significant effects. These results could also reflect a type 2 error due to the sample size, which was now smaller than that in the main study or to a very small effect of treatment. However, large changes in behaviors or responses might be detected even with a small sample size, but they were not.

Third, the information provided by our subjects was all obtained by self-report. It is possible that the findings of few significant changes in sexual behaviors and responses are related to our subjects’ inability to report their own sexual thoughts or activities comfortably. This might be particularly true for the boys, who were at least initially interviewed by a woman. No girls were interviewed by a man.

Fourth, the gender differences noted may be related to gender differences in reporting rather than to true gender differences in behaviors.

The presence of only a few significant hormone effects in this study strongly suggests that factors other than hormones play an important role in the development of sexual behaviors and responses in subjects with delayed puberty. Concentrations of sex steroids in the adolescent range may be necessary for engaging in sexual responses, thinking, and behaviors, but they are not sufficient. Although ruling out a contribution of sex steroids in the production of small changes in or a larger variety of sexual responses and behaviors is not possible, it is highly likely that the development of those responses and behaviors in boys and girls progressing through induced puberty is controlled to a great extent by nonbiological factors (16) that may play a permissive role for the effects of sex steroids.

	Acknowledgments

 
We thank Ayerst-Wyeth for supplying Premarin and placebo.

	Footnotes

 
¹ This work was supported by NIH Grants R01-HD-26636 and MO1-RR-10732 and a grant from the Genentech Foundation for Growth and Development.

Received January 27, 1998.

Revised April 1, 1998.

Accepted April 8, 1998.

	References

Top Abstract Introduction Subjects and Methods Results Discussion References

 

1. Friedrich WN, Grambsch P, Broughton D, Kuiper J, Beilke RL. 1991 Normative sexual behavior in children. Pediatrics. 88:456–464.[Abstract/Free Full Text]
2. Udry JR, Billy JOG, Morris NM, Groff TR, Raj MH. 1985 Serum androgenic hormones motivate sexual behavior in adolescent boys. Fertil Steril. 47:90–94.
3. Udry JR, Talbert LM, Morris NM. 1986 Biosocial foundations for adolescent female sexuality. Demography. 23:217–230.[Medline]
4. Kulin HE, Bell PM, Santen RJ, Ferber AJ. 1975 Integration of pulsatile gonadotropin secretion by timed urinary measurements: an accurate and sensitive 3-hour test. J Clin Endocrinol Metab. 40:783–792.[Abstract/Free Full Text]
5. Lookingbill DP, Horton R, Demers LM, Egan N, Marks JG, Santen RJ. 1985 Tissue production of androgens in women with acne. J Am Acad Dermatol. 12:481–87.[Medline]
6. Demers LM, Lipton A, Harvey HA, et al. 1993 The efficacy of CGS 20267 in suppressing estrogen biosynthesis in patients with advanced stage breast cancer. J Steroid Biochem Mol Biol. 44:687–691.[CrossRef][Medline]
7. Liang K-Y, Zegler SL. 1986 Longitudinal data analysis using generalized linear models. Biometrika. 73:13–22.[Abstract/Free Full Text]
8. Davidson JM, Myers LS. 1988 Endocrine factors in sexual psychophysiology. In: Rosen RC, Beck JG, eds. Patterns of sexual arousal. New York: Guilford Press; 1588.
9. Tricker R, Casaburi R, Storer TW, et al. 1996 The effects of supraphysiological doses of testosterone on angry behavior in healthy eugonadal men: a clinical research center study. J Clin Endocrinol Metab. 81:3754–3758.[Abstract]
10. Wu FCW. 1996 Androgen and male sexual function. In: Bhasin S, ed. Pharmacology, biology and clinical application of androgens. New York: Wiley-Liss; 191–196.
11. Money J, Clopper RR. 1974 Psychosocial and psychosexual aspects of errors of pubertal onset and development. Hum Biol. 46:173–181.[Medline]
12. Erhardt AA, Meyer-Bahlberg HFL. 1986 Idiopathic precocious puberty in girls:Long-term effects on adolescent behavior. Acta Endocrinol (Copenh). 279:247–253.
13. Anderson RA, Bancroft J, Wu FCW. 1992 The effects of exogenous testosterone on sexuality and mood of normal men, J Clin Endocrinol Metab. 75:1503.[Abstract]
14. Bagatell CJ, Heiman JR, Rivier JE, Bremner WJ. 1994 Effects of endogenous testosterone and estradiol on sexual behavior in normal young men. J Clin Endocrinol Metab. 78:711–716.[Abstract]
15. Oliver MB, Hyde JS. 1993 Gender differences in sexuality: a meta-analysis. Psychol Bull. 114:29–51.[CrossRef][Medline]
16. Rogers JL, Rowe DC. 1993 Social contagion in adolescent sexual behavior: a developmental EMOSA model. Psychol Rev. 100:479–510.[CrossRef][Medline]

This article has been cited by other articles:

				J.-C. Carel, C. Elie, E. Ecosse, M. Tauber, J. Leger, S. Cabrol, M. Nicolino, R. Brauner, J.-L. Chaussain, and J. Coste Self-Esteem and Social Adjustment in Young Women with Turner Syndrome--Influence of Pubertal Management and Sexuality: Population-Based Cohort Study J. Clin. Endocrinol. Metab., August 1, 2006; 91(8): 2972 - 2979. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Submit a related Letter to the Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Finkelstein, J. W. Articles by Kulin, H. E. Search for Related Content PubMed PubMed Citation Articles by Finkelstein, J. W. Articles by Kulin, H. E.