| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Department of Obstetrics and Gynecology (M.L.U., B.J.V., H.L.J.) and Internal Medicine (J.W.M.), Cedars-Sinai Medical Center and the University of California School of Medicine, Los Angeles, California 90095
Address all correspondence and requests for reprints to: Howard L. Judd, M.D., Department of Obstetrics and Gynecology, Olive View-University of California-Los Angeles Medical Center, 14445 Olive View Drive, Room 2B163, Sylmar, California 91342-1495.
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Estrogen is also thought to promote the formation of gallstones. Epidemiologically, women have more gallstones than men. Gallstone formation is increased in pregnancy, with oral contraceptive usage (12, 13), in men treated with estrogen (14, 15), and in some studies of postmenopausal women on estrogen therapy (16, 17, 18). Mechanistically, estrogens have been shown to increase cholesterol saturation of bile, alter bile acid composition, and decrease bile flow (19, 20). Each or all of these effects could contribute to enhanced stone formation. The oral administration of estrogen elicits enhanced hepatic actions on lipid and protein synthesis compared to nonoral administration. Thus, it has been hypothesized that the nonoral administration of estrogen may reduce the impact this hormone has on the composition of bile and the formation of cholesterol gallstones.
The objectives of the present investigation were 1) to elucidate further the potential mechanisms through which estrogens could promote cholesterol gallstone formation, and 2) to compare the impact of nonoral (transdermal) and oral estrogen administration on a variety of serum, hepatic, and biliary end points of estrogen action.
| Materials and Methods |
|---|
|
|
|---|
Using a double blind and parallel study design, all subjects were randomized to receive either E2 (0.1 mg administered by a transdermal patch that was applied every 3.5 days; Estraderm, Ciba-Geigy Pharmaceuticals, Summit, NJ) or oral conjugated equine estrogens (1.25 mg daily; Premarin, Wyeth-Ayerst Laboratories, Philadelphia, PA). Each was administered for 8 weeks. These doses were chosen because these have been shown to have comparable effects on nonhepatic markers of estrogen action and to increase the likelihood that any biliary effects elicited would be observed (21). A placebo tablet identical to conjugated equine estrogens was given to those subjects receiving an active patch, and a placebo patch identical to the active one was applied to those subjects taking active conjugated equine estrogen tablets. All study medications and placebo patches and tablets were supplied by Ciba-Geigy Pharmaceuticals. The study was approved by the institutional review board at Cedars-Sinai Medical Center (Los Angeles, CA), and each woman gave written informed consent.
All participants underwent a complete history and physical, pelvic examination, complete blood count, serum chemistry and lipid panels, Papanicolaou smear, vaginal cytology, endometrial biopsy, mammogram, and ultrasound of the gallbladder. For entry of a subject into the study, the results of all of these tests had to be normal.
Blood samples were drawn at baseline and after 8 weeks of estrogen administration. The end of treatment blood samples were obtained at 08000900 h on the second day of the application of the last patch and 24 h after the ingestion of the last tablet. Bile was also collected after venipuncture at baseline and again after 8 weeks of estrogen therapy in women who fasted overnight. A nasoduodenal tube was passed, and its position in the duodenum near the opening of the common bile duct was verified by fluoroscopy. Contraction of the gallbladder was stimulated with an iv infusion of the octapeptide of cholecystokinin at a rate of 0.5 ng/kg/min for 20 min. The darkest bile obtained during continuous aspiration was used for analysis (7). After the study was completed, all subjects with an intact uterus received medroxyprogesterone acetate (10 mg, orally; Provera, Upjohn Co., Kalamazoo, MI) for 30 days to avoid the development of endometrial hyperplasia.
Serum FSH, LH, E2, estrone (E1), estrone sulfate (E1S), and sex hormone-binding globulin (SHBG) were measured in both serum samples by methods that have been described previously (21, 22). Serum total cholesterol, high density lipoprotein cholesterol, low density lipoprotein cholesterol, very low density lipoprotein cholesterol, and triglycerides were assessed in both samples by the Physicians Clinical Laboratories (Burbank, CA). Lipoprotein cholesterol fractions were precipitated by dextran sulfate and magnesium, and measured with an Olympus AU5200 (Lake Success, NY) along with total cholesterol and triglycerides (23).
An aliquot of bile obtained from the duodenum was centrifuged at 100,000 x g for 2 h. The supernatant was used for biochemical analyses. Cholesterol crystals were identified in the sediment by light microscopy at x400. The concentration of cholesterol in bile was measured by gas-liquid chromatography (24), total bile acids by the steroid dehydrogenase enzymatic method of Talalay (25), and total phospholipids according to the method of Fiske and Subbarow (as lipid phosphorus; recovery, 97.6%) (26). The cholesterol saturation index was determined using Careys critical tables (27). Total glycoproteins (which include mucous glycoproteins) were determined by the method of Yamazaki and LaRusso (28) and Perdigoto et al. (29). PGE2 levels in bile were measured by RIA as previously described (30). Nucleation time was determined as previously described (11). The use of duodenal bile for biochemical determinations and nucleation time has been validated previously (7, 10, 11, 31).
Sample size calculations were based on the biliary cholesterol
saturation index data from Van Erpecum et al. (32) where the
SE of the cholesterol saturation index for each of the two
treatment groups was approximately 0.33 (based on an overall
approximate SE of 0.08 and a sample size of 17). In this
parallel group trial, to achieve an
level of 0.05, a ß level of
0.20 (study power of 0.80), and a
of 0.20 would require a minimum
of 43 participants in each treatment group to be studied. All values
are expressed as the mean ± SD unless otherwise
noted. Comparison of baseline values between groups was performed by a
two-sample t test. A paired t test was used to
determine whether the percent change or the change from baseline of
each treatment was significantly different from zero. Between-treatment
comparisons were carried out by examining the differences between
treatments in percent change and change from baseline. Two sample
t tests or McNemar tests were used for these comparisons.
Significance was accepted at P < 0.05.
| Results |
|---|
|
|
|---|
The baseline laboratory values by treatment groups were not significantly different for serum hormones, lipid profiles, and biliary measurements. The levels of both gonadotropins were elevated, whereas the concentrations of the three estrogens were within the ranges observed in hypogonadal women, reflecting the postmenopausal status of the participants. Mean total cholesterol levels (219.3 ± 29.2 mg/dL for transdermal E2 and 222.9 ± 31.1 mg/dL for oral conjugated equine estrogens) were in the borderline elevated range, whereas the cholesterol fractions and triglyceride levels were in the normal ranges for women in this age group. For the biliary end points, the cholesterol saturation indexes for transdermal E2 and oral conjugated equine estrogens were 1.0 ± 0.2 and 0.99 ± 0.2, respectively, and the nucleation times for transdermal E2 and oral conjugated equine estrogens were 17.8 ± 5.8 and 18.1 ± 5.6 days, respectively. These values were similar to those observed in healthy perimenopausal women and obese women without gallstones (7, 33). Eight women in each group had cholesterol crystals at baseline. Levels of archidonate (4.5 ± 2.3% for transdermal E2 and 4.5 ± 2.4% for oral conjugated equine estrogens), PGE2 (262.6 ± 141.4 pg/mL for transdermal E2 and 256.6 ± 179.1 pg/mL for oral conjugated equine estrogens), and total glycoproteins (2.5 ± 3.2 mg/mL for transdermal E2 and 1.8 ± 1.4 mg/mL for oral conjugated equine estrogens) in these postmenopausal women were also comparable to values in obese patients without gallstones (10). The values of all of these end points were not significantly different between groups, again indicating that randomization of the participants was successful in separating two similar groups.
The percent change with each treatment from baseline as well as a
comparison of the percent change for each parameter between groups are
listed in Table 1
. Both treatments
elicited significant and similar decreases from baseline in the values
of serum FSH and LH. Both also showed significant increases from
baseline in the values of serum E2, E1, and
E1S. The mean increase in E2 was significantly
greater with transdermal E2, whereas the increases in
E1 and E1S were significantly greater with oral
conjugated equine estrogens. For SHBG, no significant change from
baseline was observed with transdermal E2, whereas a
significant increase was seen with oral conjugated equine estrogens.
This between-treatment difference was statistically significant
(P < 0.05) for SHBG.
|
|
|
| Discussion |
|---|
|
|
|---|
It was also anticipated that conjugated equine estrogens would have a greater impact than transdermal E2 on hepatic markers of estrogen action because of its oral route of administration. This was confirmed by the significantly greater changes in SHBG and circulating lipids with conjugated equine estrogens than with transdermal E2. These findings also confirmed results reported previously by us and others (21, 34, 35).
Both estrogen treatments significantly increased the biliary saturation of cholesterol and decreased the time required to form cholesterol crystals in bile in vitro (nucleation time). Taken together, these findings support the concept that transcutaneous and oral estrogen administration increases the chances of women developing gallstones by increasing the cholesterol saturation index and promoting nucleation of cholesterol into crystals. An increase in the cholesterol saturation index and a decrease in the nucleation time represent two of the three mechanisms theorized to be responsible for stone formation (1, 2). In addition, significant increases in arachidonate and PGE2 were seen (36). These factors may be responsible for mucus glycoprotein production in bile (37). This last factor decreases nucleation time and enhances probable stone formation. These findings continue to support the concept that estrogens secreted in normal or pregnant women, administered in oral contraceptives, or given as hormone replacement increase the chances of gallstone formation. The only clinical trial of estrogens that reported gallbladder disease found two women each with stones in the placebo and estrogen only group. However, the trial was too small and brief to determine differences in gallstone formation between treatment groups (38).
It must be remembered that the doses of estrogen used in this study (0.1 mg transdermal E2 and 1.25 mg conjugated equine estrogens) are prescribed for estrogen replacement, but are twice as much as the lowest doses required to prevent osteoporosis (39). These doses were chosen to increase the chance of seeing changes in biliary function, which was accomplished. It is likely that lower doses of transdermal E2 and oral conjugated equine estrogens would have less effect on biliary end points.
It is of great interest that the enhanced hepatic actions elicited by oral conjugated equine estrogens in comparison to transdermal E2 were not reflected in the biliary lipid alterations. It is unlikely that these findings were the result of an artifact of the study design, because significant and similar changes in biliary end points occurred with both the transdermal and oral estrogen groups. These findings suggest that the nonoral administration of estrogen does not avoid the biliary changes elicited with oral therapy. There are few reports available in the literature examining gallbladder bile after the administration of transdermal or oral estrogens. Heuman and co-workers (19) studied three postmenopausal women who were given 50 µg ethinyl estradiol daily for 3 weeks followed by intervals of 1 week without therapy for three or four treatment cycles. They found that the treatment resulted in an increased fraction of cholesterol in gallbladder bile and a shift in the bile acid composition favoring cholesterol gallstone formation. A single postmenopausal woman was given percutaneous estradiol valerate (5 mg daily) for 6 weeks and after a 6-week washout period, oral estradiol valerate (2 mg daily) for 6 weeks (40). After both methods of administration, there was an increase in the biliary cholesterol concentration but after oral administration, bile flow also increased, and cholesterol crystals appeared in the bile. Everson and colleagues (41) studied 29 anovulatory women, aged 2862 yr, given oral conjugated equine estrogens in doses ranging from 0.6251.25 mg daily for at least 4 weeks. They found that oral estrogens increased the lithogenic index of bile, increased biliary cholesterol secretion, and altered biliary acid composition. The report by Van Erpecum et al. (32) was the first to directly compare transdermal to oral E2 in 17 postmenopausal women. Unfortunately, the small number of patients investigated limited the power of their study.
In summary, transdermal and oral estrogen preparations were administered to postmenopausal women for 8 weeks to determine their impact on a variety of markers of estrogen action. Although the administration of these two medications resulted in very different circulating levels of estrogens, they elicited comparable effects on nonhepatic markers of estrogen action (gonadotropin levels). The oral administration of conjugated equine estrogens induced greater changes in hepatic markers (levels of SHBG and lipids) than transdermal E2. Significant increases in biliary cholesterol saturation index, arachidonate, and PGE2 and a significant decrease in the nucleation time were observed after both the nonoral and oral administration of estrogens. These findings continue to support the hypothesis that estrogens, whether administered transcutaneously or orally, increase the chances of gallstone formation by alteration of biliary lipids and nucleation of cholesterol that have been incriminated in this process.
| Footnotes |
|---|
2 Present address: Department of Obstetrics and Gynecology, Loyola
University Medical Center, 2160 South First Avenue, Maywood, Illinois
60153. ![]()
Received June 24, 1997.
Revised October 8, 1997.
Accepted October 20, 1997.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
D. J. Cirillo, R. B. Wallace, R. J. Rodabough, P. Greenland, A. Z. LaCroix, M. C. Limacher, and J. C. Larson Effect of Estrogen Therapy on Gallbladder Disease JAMA, January 19, 2005; 293(3): 330 - 339. [Abstract] [Full Text] [PDF] |
||||
![]() |
Cholecystitis, Biliary Tract Surgery, and Pancreatitis Obstet. Gynecol., October 1, 2004; 104(4_suppl): 17S - 24S. [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 |