| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH |
From the Département d’Information Médicale and the Service de Gynécologie et d’Obstétrique, Hôtel-Dieu, Hospices Civils de Lyon, Lyons, France; the Centro Ambrosiano Metodi Naturali, Milan, Italy; and Quidel Corporation, San Diego, California.
Address reprint requests to: René Ecochard, MD, PhD, Hospices Civils de Lyon, Département d’Information Médicale, 162, Avenue Lacassagne, 69424 Lyons Cedex 03, France, E-mail: rene.ecochard{at}chu-lyon.fr
 |
Abstract |
|---|
Top
 Abstract Materials and MethodsResultsDiscussionReferences |
|---|
Methods: We compared baseline characteristics, cycle characteristics, follicle sizes, and daily hormone levels among women with low (n = 22), normal (n = 63), or high (n = 22) BMIs over 326 cycles.
Results: There were no significant differences in age or other lifestyle characteristics between groups. High BMI was significantly associated with younger age at menarche and less sleeping time. No differences were observed between high- and low-BMI groups in cycle length or diameter of the dominant follicle. Luteinizing hormone levels were significantly higher only in the beginning of the cycle in women with low BMIs than in those with high BMIs. Follicle-stimulating hormone levels were also significantly higher but were high during all three phases of the cycle (early follicular, periovulatory, and luteal phases). Mean levels were approximately 1.9, 1.8, and 1.2 times higher, respectively, in the low-BMI group than the high-BMI group.
Conclusion: Luteinizing hormone levels and BMI were inversely associated in normally cycling women during the early follicular phase. Follicule-stimulating hormone levels and BMI were inversely associated during the whole cycle, independent of age.
The relationship between gonadotropin levels and body mass index (BMI) in polycystic ovary syndrome (PCOS) have been discussed thoroughly.1,2 Women with PCOS can be divided into two subgroups: those with high BMIs, insulin resistance, and normal to moderately increased LH levels; and those with normal BMIs and increased LH levels.3,4 Polycystic ovary syndrome is a complex clinical entity that probably includes several other syndromes. In PCOS, there is a constant inverse relationship between BMI and LH levels but no clear modification of FSH levels.
Results of studies of the relationship between gonadotropin levels and BMI in normally cycling women have conflicted. In some studies, LH levels did not differ between lean and obese women,5,6 whereas in other studies, they differed significantly.7 Only recently were relationships between BMI and LH levels and BMI and LH-FSH ratios specifically assessed in a large sample of normally menstruating women.8 Body mass index was inversely associated with LH levels over most of the BMI range and was inversely associated with LH-FSH ratios, but the latter association was of borderline significance. The investigators used blood specimens taken during the first 3 days of menstrual cycles, and results were confirmed after considering age.
The aims of the present study were to investigate whether the relationship between LH levels and BMI in normally cycling women is limited to the early preovulatory phase or extends throughout the cycle and to assess the relationship between FSH levels and BMI.
|
Materials and Methods |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Participants were ostensibly healthy menstruating women aged 18–45 years with previous menstrual cycle lengths of 24–34 days. Subjects were excluded if they were breast-feeding, postpartum (first 3 months), or perimenopausal; had frequent anovulatory cycles; were participating in programs to stimulate hormonal responses for fertility reasons; or had abnormal cycles (polycystic ovary disease or luteal defect), pelvic inflammatory disease, or histories of fertility problems or special habits such as running. One hundred seven subjects meeting those criteria were selected between June and November 1996. Mean age was 32.4 years (range 19–44 years) and mean BMI was 21.2 (range 17.1–28.3). Each subject participated for at least one cycle. On average, each subject participated for three menstrual cycles. Precisely, six, five, 74, and 22 women participated for one, two, three, and four cycles, respectively. The total number of cycles was 326; mean cycle length was 28.46 days.
On study entry, each subject was administered a standard questionnaire and underwent a physical examination. Data sought included age, age at menarche, parity, past oral contraceptive (OC) use, lifestyle habits such as smoking, special diets, sports activity (number of hours per week), and usual stress (general subjective feeling). Height and weight were measured and BMI was calculated.
Early-morning urine samples were collected daily in tubes containing gentamicin sulfate. These tubes were frozen at -20C the day of collection. Urine samples were assayed for quantitative detection of estrone (E1)-3-glucuronide, pregnanediol-3
-glucuronide, FSH, and LH using time-resolved fluorometric immunosorbent assays.
Ovulation was monitored using ultrasound. Investigations were performed after the onset of fertile-type cervical mucus9 or detection of LH surge by rapid assay (Bluetest Ovulation Test; Quidel Corp., San Diego, CA). Once 16-mm follicles were observed, scanning was done daily until the presumed day of ovulation. The dominant follicle was defined as the largest-diameter cystic structure. The ultrasound-estimated day of ovulation was defined as the day of maximum follicular enlargement followed the next day by ultrasound evidence of follicle rupture. The follicular phase was the delay from the onset of menses to the estimated day of ovulation, inclusive. The luteal phase was the delay from the 1st day with evidence of follicle rupture to the day preceding the onset of the following menses, inclusive.
Subjects were separated into three groups: the low-BMI (lower quintile, less than 19.2), normal-BMI, and high-BMI (upper quintile, greater than 23.4) groups.
Descriptive summaries of baseline characteristics were generated to examine distributions and calculate means with standard errors for continuous variables (such as age, age at menarche, and cycle characteristics) and rates for categoric variables (such as smoking, special diet, sports activity, and stress).
In the tables, we limited our description of average hormone levels to three periods of the menstrual cycle: days 3 ± 1 (early follicular phase), estimated day of ovulation ± 1 (periovulatory phase), and estimated day of ovulation +5, +7, and +9 (luteal phase). However, for graphic representations, daily hormone levels were used. Geometric mean values of hormone concentrations during the menstrual cycle were expressed as percentages of the maximum values, with the estimated day of ovulation as center point.
To compare baseline characteristics between groups, we used one-way analysis of variance for continuous variables and the Pearson 
2 test for categoric variables. To compare characteristics of cycles between groups, we used analysis of variance in which the cycle was treated as a repeated-measures factor. Hormone levels were not normally distributed; a log transformation was done to obtain a normal distribution before statistical analysis.
This univariate analysis was followed by a stepwise multivariable hierarchic regression analysis10 with forward selection. At the top of the hierarchy were the centers, followed by the women within the centers, then the cycles within the women, and finally the days within the cycles. Cycles and days were introduced as dummy variables, and the linearity of the relationship between BMI or age and hormones was tested using a quadratic regression. To test whether the association between differences in FSH or LH levels and BMI were maintained through the menstrual cycle, we introduced a term of interaction between differences in the hormone levels and the rank of the day within the cycle.
Univariate analyses were done using the S-Plus statistical package (MathSoft Inc., Seattle, WA). Multilevel regression analyses were done using the SAS statistical package (SAS Institute, Cary, NC). Significance was defined as P < .05.
|
Results |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
). High BMI was significantly associated with younger age at menarche and less sleeping time. Univariate comparisons of menstrual cycle characteristics and hormone levels between groups are shown in Table 2. There were no differences between the high-BMI and low-BMI groups in cycle length or dominant follicle diameter.
|
|
-glucuronide levels were significantly higher in the low-BMI group than in the high-BMI group, but the differences did not persist during the periovulatory and luteal phases of the cycle. We observed no difference in E1-3-glucuronide levels between low-BMI and high-BMI groups. Day-3 LH levels were significantly higher in the low-BMI group than in the high-BMI group, but the differences did not persist during the periovulatory and luteal phases of the cycle.
Day-3 FSH levels were also significantly higher in the low-BMI group than in the high-BMI group, but in contrast to the differences in LH levels, the large differences in FSH levels persisted during all three phases of the cycle. Thus, during the early follicular, periovulatory, and luteal phases of the cycle, FSH mean levels were approximately 1.9, 1.8, and 1.2 times higher, respectively, in the low-BMI group than in the high-BMI group.
Using daily measurements and multivariable hierarchic regression analysis, we clearly demonstrated that differences in FSH levels between low-BMI and high-BMI groups were maintained through the menstrual cycle (Figure 1). The high LH levels measured in the low-BMI group compared with the levels in the high-BMI group were limited to the early follicular phase. The association of greater age with high LH levels and high FSH levels was clear and significant during the early follicular phase but not during the periovulatory and luteal phases (Figure 2). In the figures, log-transformed hormone data were compared and differences appeared more important after that transformation, which is known to be useful for those comparisons.
|
|
Multivariable analysis also confirmed the trend for LH levels after considering age, but only during the early follicular phase. Levels of LH increased with age (P < .001) and decreased with BMI (P < .001).
|
Discussion |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Recent results obtained by Bohlke et al8 raised the question of the relationship between gonadotropin levels and BMI in normally cycling women. Our results concur with those earlier findings, that LH levels are higher during the early follicular phase in lean women, independent of age. Conflicting results obtained in previous studies might have been due to lack of power, because healthy women were used as controls in these studies of PCOS.3,5–7,12,13 Our results showed that the difference was not maintained during the periovulatory and luteal phases. We observed the same trend of high FSH levels in lean women, and this trend was maintained throughout the cycle in younger and older women.
Regarding the distinction between weight and diet, are the low levels of LH and FSH in women with high BMIs due to current diet disorders? Previous studies did not support that hypothesis. Drezgic et al14 observed that a 21-day fast with significant weight loss did not affect gonadotropin pulse pattern. Taylor et al15 also found no significant differences in follicular or luteal phase lengths after 3 days of normal or binge eating. Thus, the relationship between LH or FSH levels and BMI seems to be stable. A nonnegligible proportion of milder forms of PCOS in the normal population could explain previous results8 and our results concerning the relationship between gonadotropin levels and BMI. Whereas PCOS occurs in at least 5% of the population and was an exclusion criterion in our study, polycystic-appearing ovaries are found in 16–25% of the population.16 A subgroup of women with polycystic-appearing ovaries (up to 30% of all women) could have subtle abnormalities resembling PCOS.17 Nevertheless, normally ovulating women with polycystic-appearing ovaries cannot be considered to have PCOS.18 The absence of such disorders was confirmed in our study by measurement of hormone levels and ultrasound.
In our study, pregnanediol-3-glucuronide levels on day 3 of the cycle were significantly higher in the low-BMI group than in the high-BMI group, but those differences did not persist during the periovulatory and luteal phases of the cycle, and there was no difference in E1-3-glucuronide levels between the low-BMI and high-BMI groups. The sample size was sufficient to show diversity between women and cycles but was not calculated statistically. Thus, statistically significant differences are reliable, but nonsignificant results do not preclude effects.
|
Footnotes |
|---|
Received October 25, 1999. Received in revised form January 25, 2000. Accepted February 10, 2000.
|
References |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
2. Carmina E, Lobo RA. Do hyperandrogenic women with normal menses have polycystic ovary syndrome? Fertil Steril 1999;71:319–22.[Medline]
3. Dale PO, Tanbo T, Vaaler S, Abyholm T. Body weight, hyperinsulinemia, and gonadotropin levels in the polycystic ovarian syndrome: Evidence of two distinct populations. Fertil Steril 1992;58: 487–91.[Medline]
4. Arroyo A, Laughlin GA, Morales AJ, Yen SSC. Inappropriate gonadotropin secretion in polycystic ovary syndrome: Influence of adiposity. J Clin Endocrinol Metab 1997;82:3728–33.
5. Morales AJ, Laughlin GA, Butzow T, Maheshwari H, Baumann G, Yen SSC. Insulin, somatotropic, and luteinizing hormone axes in lean and obese women with polycystic ovary syndrome: Common and distinct features. J Clin Endocrinol Metab 1996;81:2854–64.[Abstract]
6. Acién P, Quereda F, Matallin P, Villarroya E, Lopez-Fernandez JA, Acién M, et al. Insulin, androgens, and obesity in women with and without polycystic ovary syndrome: A heterogeneous group of disorders. Fertil Steril 1999;72:32–40.[Medline]
7. Taylor AE, McCourt B, Martin KA, Anderson EJ, Adams JM, Schoenfeld D, et al. Determinants of abnormal gonadotropin secretion in clinically defined women with polycystic ovary syndrome. J Clin Endocrinol Metab 1997;82:2248–56.
8. Bohlke K, Cramer DW, Barbieri RL. Relation of luteinizing hormone levels to body mass index in premenopausal women. Fertil Steril 1998;69:500–4.[Medline]
9. Hilgers TW, Abraham GE, Cavanagh D. Natural family planning. I. The peak symptom and estimated time of ovulation. Obstet Gynecol 1978;52:575–82.[Abstract]
10. Goldstein H. Multilevel mixed linear model analysis using iterative generalized least squares. Biometrika 1986;73:43–56.
11. Helm P, Munster K, Schmidt L. Recalled menarche in relation to infertility and adult weight and height. Acta Obstet Gynecol Scand 1995;74:718–22.[Medline]
12. Holte J, Bergh T, Gennarelli G, Wide L. The independent effects of polycystic ovary syndrome and obesity on serum concentrations of gonadotrophins and sex steroids in premenopausal women. Clin Endocrinol (Oxf) 1994;41:473–81.[Medline]
13. Tropeano G, Vuolo IP, Lucisano A, Liberale I, Barini A, Carfagna P, et al. Gonadotropin levels in women with polycystic ovary syndrome: Their relationship to body weight and insulin levels. J Endocrinol Invest 1996;19:139–45.[Medline]
14. Drezgic M, Penezic Z, Zarkovic M, Vujovic S, Ciric J, Trbojevic B, et al. Influence of three-week fasting on gonadotropin pulsatility in obese menstruating women. Int J Obes Relat Metab Disord 1996; 20:608–12.[Medline]
15. Taylor AE, Hubbard J, Anderson EJ. Impact of binge eating on metabolic and leptin dynamics in normal young women. J Clin Endocrinol Metab 1999;84:428–34.
16. Polson DW, Adams J, Wadsworth J, Franks S. Polycystic ovaries—A common finding in normal women. Lancet 1988;1:870–2.[Medline]
17. Carmina E, Wong L, Chang L, Paulson RJ, Sauer MV, Stanczyk FZ, et al. Endocrine abnormalities in ovulatory women with polycystic ovaries on ultrasound. Hum Reprod 1997;12:905–9.
18. Carmina E, Lobo RA. Polycystic ovary syndrome (PCOS): Arguably the most common endocrinopathy is associated with significant morbidity in women. J Clin Endocrinol Metab 1999;84:1897–9.
This article has been cited by other articles:
|
|
 |
|
  D. A. Dumesic, M. A. Damario, D. R. Session, A. Famuyide, T. G. Lesnick, A. R. Thornhill, and A. S. McNeilly Ovarian Morphology and Serum Hormone Markers as Predictors of Ovarian Follicle Recruitment by Gonadotropins for in VitroFertilization J. Clin. Endocrinol. Metab., June 1, 2001; 86(6): 2538 - 2543. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
