Levels of Serum C-Reactive Protein during Oral and Transdermal Estradiol in Postmenopausal Women with and without a History of Intrahepatic
http://www.100md.com
《临床内分泌与代谢杂志》
Abstract
Liver dysfunction may affect the production and release of C-reactive protein (CRP). We designed a double-blind prospective crossover study involving 40 postmenopausal women with or without a history of intrahepatic cholestasis of pregnancy (ICP), where we compared the basal levels of CRP and their responses to increasing doses of oral and transdermal estradiol (E2), followed by addition of oral medroxyprogesterone acetate (MPA). Serum samples collected at baseline, on the last day of each E2 period, and on the last day of the E2 + MPA combination were assayed for CRP, estrogens, and liver enzymes. There was no difference in basal CRP between the study groups. Both regimens (oral and transdermal E2) were accompanied by significant rises in estrone and E2 concentrations; the former were 16 times higher during the oral than during the transdermal regimen. Oral E2 elevated CRP dose dependently, and this response was unaffected by a history of ICP or the use of MPA. The activities of liver transaminases varied but were in normal ranges during E2 use, in women with and without a history of ICP. In conclusion, the synthesis of CRP is not affected by a history of ICP. It is readily and dose dependently stimulated by oral but not by transdermal E2 in as soon as 2 wk.
Introduction
C-REACTIVE PROTEIN (CRP) is primarily derived from the liver under the control of IL-6 stimulation (1). This protein has been widely studied because small elevations in concentration within the range not indicative of infection predict cardiovascular risk (2, 3, 4, 5). CRP, especially its activated form in the blood vessel wall (3, 6), induces the expression of various adhesion molecules in the endothelium, which in turn accelerates vascular inflammatory reactions and may accelerate the development of atherosclerosis (7, 8). Long-term oral postmenopausal hormone therapy (HT) has been shown to be associated with elevated levels of CRP (5, 9, 10, 11, 12, 13, 14, 15, 16), and some authors feel that this rise in CRP may account for the lack of effect of oral HT in the primary [Women’s Health Initiative (WHI)] (17) or secondary [Heart and Estrogen-Progestin Replacement Study (HERS)] (18) prevention of ischemic cardiac events. Transdermal HT may be safer in this regard because it has been shown to reduce (19, 20) or have no effect on circulating CRP levels (12, 15). The difference in CRP response to oral and transdermal HT is possibly a result of higher estrogenicity in the liver during oral than during transdermal regimens (21). It is also known that a phytoestrogen regimen used as an alternative to HT does not affect the circulating levels of CRP in postmenopausal women (23). Data on the impact of progestins on CRP are controversial because reductions (19), elevations, or no changes (10, 11, 13, 16, 24) in circulating CRP concentrations have been reported after the addition of progestin to an estrogen regimen.
Intrahepatic cholestasis of pregnancy (ICP), which complicates approximately one in 100 pregnancies, manifests itself as itching skin and elevated activities of liver transaminases, usually in the third trimester of pregnancy (25, 26). The cause of ICP is largely unknown, although some women with mutations in the ABCB4 gene have been described (27, 28, 29, 30). ICP is most likely an etiologically heterogeneous condition, which has been suggested to be a consequence of insufficient hepatic capacity to metabolize the high amounts of placenta-derived sex steroids in genetically predisposed individuals (31). The condition disappears soon after delivery but the increased risk of gallstones (26, 32) suggests permanent hepatobiliary dysfunction in these women. This is also supported by data indicating that the use of high-dose oral contraceptives often triggers elevations in liver enzyme activities in these women (33, 34, 35). Whether postmenopausal HT affects liver function in women with a previous history of ICP has not to our knowledge been studied before.
No data exist on CRP in women with a history of ICP. Moreover, in previous studies on HT and CRP (5, 9, 10, 11, 12, 13, 14, 15, 16), CRP has been assayed at 1–6 months after the start of HT. Thus, we do not yet know how soon CRP may be stimulated after the start of HT. Therefore, we designed a trial to assess the effects of relatively short-term increasing doses of oral and transdermal estradiol (E2) treatment on CRP, with and without progestin, in women with and without a history of ICP.
Subjects and Methods
Subjects
With the permission of the local Ethics Committee, we studied 40 postmenopausal women who had entered the menopause 6–7 yr before the study (Table 1). The volunteers received thorough written and verbal information on the conduct of the study, and an informed consent was obtained from all of them. They all complained of hot flashes and other climacteric symptoms, and 28 women had previously used HT; this had been stopped at least 4 wk before recruitment. The women were otherwise healthy and reported no allergic skin reaction to patches, but two of them used calcium channel blockers, and two other women used angiotensin-converting enzyme blockers to control hypertension. Three other women used thyroxine. None of these drugs should have any effect on liver function. The baseline characteristics of the patients and controls are shown in Table 1. Twelve women had undergone hysterectomy, and both ovaries had been removed in nine of them. All women had a history of at least one full-term pregnancy, and this was complicated by ICP in 20 women (in 13 women repeatedly). On average, ICP had occurred 30 yr (26–33 yr) before the study. The diagnosis of ICP had been based on typical skin itching and on elevations in liver enzyme activities in the third trimester of pregnancy; all symptoms and signs had vanished 6 wk after delivery.
Study design
The women were randomized to receive increasing doses (2 mg for 14 d; 4 mg for 14 d) of E2 either orally (Progynova, estradiol valerate, Schering AG, Berlin, Germany) or transdermally from a patch (50 μg/24 h for 14 d; 100 μg/24 h for 14 d) (FemSeven, Merck KgaA, Darmstadt, Germany) (Fig. 1). The study was blind in that during both treatments, each woman took tablets (active or placebo) and used patches (placebo or active). After using E2 only for 4 wk, each subject took 10 mg medroxyprogesterone acetate (MPA) (Lutopolar, Orion Pharma, Espoo, Finland) for 14 d concomitantly with the highest E2 dose (Fig. 1). After a 4-wk washout period, the subjects crossed over to the other treatment (Fig. 1). Thus, each woman used E2 both orally and transdermally. The women were seen at baseline (also after the 4-wk washout period) and at the end of each 2-wk treatment period, and blood samples were collected on these occasions (a total of eight times during the trial). Blood samples were drawn after an overnight fast, and centrifugally separated serum was stored at –20 C until analyzed. General and pelvic examinations were carried out before and after both 6-wk treatment periods.
Measurements
Serum CRP was assayed immunochemically by use of a monoclonal antibody specific to human CRP (Sensitive CRP IEMA TEST, Oy Medix Biochemica Ab, Kauniainen, Finland). To minimize the impact of interassay variation, the CRP assessments were carried out in one assay batch. The coefficient of intraassay variation was less than 5%. Because serum CRP concentrations rise in acute infection, we excluded those values exceeding 10 mg/liter (nine values altogether). Levels of estrone (E1) were determined by RIA (20 subjects) (Estron, Diagnostic Systems Laboratories, Sinsheim, Germany), and levels of E2 were measured by means of time-resolved fluoroimmunoassay (DELFIA Estradiol, Wallac Oy, Turku, Finland) (40 subjects). The E2 antibody shows a 0.75% crossreactivity with E1. The activities of liver enzymes, such as aspartate transaminase, alanine transaminase, and -glutamyl transferase, were measured by means of routine laboratory methods.
Statistical analyses
Analyses were carried out using SPSS version 11.0 (SPSS Inc., Chicago, IL). Data are expressed as medians (with 95% confidence intervals according to Altman’s Statistics). Because we used the crossover design, the possibility of a period effect was tested by Mann-Whitney test, where we compared the differences between the periods in the two groups of patients (those beginning with the oral and those beginning with the transdermal E2). No period effect was detected. The possibility of a treatment-period interaction was investigated by Mann-Whitney test, where we compared the average responses to the two treatments and found patients’ average responses to the two treatments to be the same regardless of the order of treatments. Because of the skewed and not normal distribution of CRP, E2, and E1 data, a nonparametric test (Wilcoxon’s signed rank test) was used to compare values at baseline and during follow-up. To better illustrate the responses of CRP with HT, we calculated a change in CRP from basal to 2 (to 4 and to 6) wk of treatment. This change is also expressed as percentages. Nonparametric repeated measures ANOVA (Friedman test) was used to test variation among groups during treatment, and Wilcoxon’s signed-rank test was used to test changes between the groups. Correlation analyses were performed using Spearman’s nonparametric correlation coefficient. A two-tailed P < 0.05 was accepted as the level of significance.
Results
The study groups were comparable in age, body mass index, and other clinical variables at baseline (Table 1). Baseline levels of CRP, E1, E2, and liver enzyme activities also did not differ (Table 2). One subject in the control group used nimesulid (COX-2 inhibitor) during the study, and owing to its possible effect on liver function, she was excluded from data analysis. None of the subjects experienced any side effects during treatment. All women reported reductions in hot flashes, and nonhysterectomized women experienced withdrawal bleeding after the MPA course.
The order of the HT regimen (oral or transdermal first) had no influence on the biochemical changes. Biomarkers at baseline and during the trial in four women using antihypertensive drugs did not differ from those not using antihypertensive drugs; therefore, no further subgroup analysis was done.
Oral E2 led to significant dose-dependent rises (51 and 87%) in CRP levels in the ICP group, but these rises did not differ from those in the control group (39 and 95%) (Table 3). Median dose-dependent rises ranged between 49 and 91% when both ICP and control women were analyzed as one group. The rises were not significantly affected by the addition of MPA in either group (Table 3).
Transdermal E2 did not affect the level of CRP in either group, even at the highest dose. A trend toward a fall in CRP concentrations was seen, and this was augmented by MPA (Table 3).
Smoking, body mass index, age, and previous use of HT were not determinants of baseline CRP or CRP responses to oral E2.
The absolute median rises (with 95% confidence intervals) in serum E1 concentrations during oral E2 treatment were 539 ng/liter (360–871) [1994 pmol/liter (1332–3223)] and 963ng/liter (462-2656) [3561 pmol/liter (1709–9824)] in ICP and control women, respectively. Transdermal E2 did not result in significantly increased E1 concentrations. The relative rises of serum E1 levels during use of oral and transdermal E2 were 2030 and 24% in the ICP group and 1470 and 55% in the control group (data not shown).
Oral E2 led to similar rises in serum E2 levels in the ICP and control group (Table 4). These elevations were greater during oral treatment than during transdermal administration, with no difference between the groups. Oral E2 at 2 mg/d and transdermal E2 at 100 μg/d led to similar rises in the circulating levels of E2 (Table 4).
During oral E2, the changes in CRP concentrations correlated positively with those in E2 levels (r = 0.326, P = 0.007; Fig. 2) but not with those in E1 levels (r = 0.08, P = 0.7). During transdermal E2, no correlations were seen between changes in circulating concentrations of CRP, E1, or E2.
Oral and transdermal E2 administration was not accompanied by significant changes in liver enzyme activities. The levels fluctuated but remained within the normal range in each subject.
Discussion
The liver and enterohepatic circulation play a key role in estrogen metabolism, and this ultimately determines the estrogenicity of a given HT regimen (21). Therefore, liver dysfunction, even if subclinical, may become a factor in the metabolism of estrogen in postmenopausal women. The liver also produces CRP, an important marker, and perhaps even a contributing factor as regards cardiovascular disease (1). It is stimulated by oral estrogens, given for 1–6 months (5, 9, 10, 15, 16). The possible role of progestin on CRP has remained uncertain (10, 13, 19, 24). We studied the effects of short-term use of E2 given either orally or transdermally on CRP in women with and without a history of ICP. We first gave increasing doses of E2 alone, both orally and transdermally for 4 wk, followed by the addition of a routine course of oral MPA for the next 2 wk. This study design allowed us to determine the time and dose dependency of the CRP-stimulating effect of E2. Moreover, the effect of MPA on CRP when added to oral or transdermal E2 could be studied.
All study subjects tolerated both oral and transdermal HT well, and none discontinued the regimen. Moreover, liver transaminase activities remained within normal ranges in both groups. As expected, both oral and transdermal E2 led to rises in the serum levels of E1 and E2. The response of E1 was much greater during oral than transdermal treatment, but E1 and E2 levels showed no differences between women with and without a history of ICP. This novel finding can be seen as evidence that in women with a history of ICP the liver can metabolize postmenopausal oral and transdermal E2 normally. Possible liver dysfunction in women with a history of ICP may be so minimal that it could not be triggered by the doses of E2 we used in our study or previously when 12 of 20 such women had used HT with no known untoward side effects. However, our data cannot exclude the possibility that more prolonged use and even higher doses of HT would have had a different effect.
These are the first data showing that oral E2 at 2 mg/d is capable of inducing rises in serum CRP concentrations as soon as within 2 wk. The response of CRP in our study was of the same order as in some previous ones where HT has been administered for periods of 1–6 months (10, 11, 12, 14, 16, 20). We also demonstrated that by increasing the dose of E2 from 2–4 mg/d we could further increase the level of CRP, although it is possible that some of this could be due to a prolonged effect of the preceding treatment with 2 mg of E2 as previously suggested by van Baal (16). The increase in CRP concentration was related to the rises in E2 only during oral use, lending support to the role of the liver in E2-induced elevation of CRP concentrations. Moreover, our data demonstrate that the effect of oral E2 on CRP vanishes totally in 4 wk and that a history of ICP has no effect on CRP levels. In contrast, transdermal E2 failed to affect CRP levels, although the dose was raised to 100 μg/d; if anything, a decreasing trend in serum CRP concentrations was seen. This was unexpected because the circulating levels of E2 were similar during oral administration of E2 at 2 mg/d and during use of transdermal E2 at 100 μg/d. Perhaps CRP stimulation in the liver also requires the presence of E1 (36), resulting in no rise in serum CRP levels during transdermal use of E2.
Previous data are not uniform as regards the effects of various progestins on CRP levels. The protocols include sequential use of micronized progesterone (10), sequential use of dydrogesterone (11), continuous use of trimegestrone (11), norethisterone acetate (13, 15, 22), gestodene (16) and MPA (2.5 mg) (10, 14, 24), and intrauterine administration of levonorgestrel (13). In general, they have had no effect on estrogen-induced rises in CRP levels, but continuous use of MPA (5 mg) (24), gestodene (25 μg) (16), or sequential use of MPA (20 mg) (13) has been reported to abolish oral E2-induced rises in CRP concentrations. We demonstrate here that MPA at 10 mg/d for 2 wk did not affect CRP levels, which were already stimulated by preceding and concomitant E2 use. Our data may be of significance because MPA has been suggested to be one factor in the lack of protective effect of conjugated equine estrogen + MPA in primary (WHI) (17) and secondary (HERS) (18) prevention of cardiac events.
Taken together, a history of ICP is not accompanied by any differences in basal CRP or in its responses to HT. Oral E2 stimulates CRP synthesis dose dependently within 2 wk, both in women with and without a history of ICP, and the addition of short-term progestin (MPA) had no effect.
Footnotes
This study was supported by grants from the research funds of the Helsinki University Central Hospital.
First Published Online November 2, 2004
Abbreviations: CRP, C-reactive protein; E1, estrone; E2, estradiol; HT, hormone therapy; ICP, intrahepatic cholestasis of pregnancy; MPA, medroxyprogesterone acetate.
Received April 26, 2004.
Accepted October 25, 2004.
References
Straub RH, Hense HW, Andus T, Sch?lmerich J, Riegger GAJ, Schunkert H 2000 Hormone replacement therapy and interrelation between serum interleukin-6 and body mass index in postmenopausal women: a population-based study. J Clin Endocrinol Metab 85:1340–1344
Koenig W, Sund M, Fr?hlich M, Fischer HG, L?wel H, D?ring A, Hutchinson WL, Pepys MB 1999 C-reactive protein, a sensitive marker of inflammation, predicts future risk of coronary heart disease in initially healthy middle-aged men. Circulation 99:237–242
Lagrand WK, Wisser CA, Hermens WT, Niessen HWM, Verheugt FWA, Wolbink GJ, Hack CE 1999 C-reactive protein as a cardiovascular risk factor. Circulation 100:96–102
Ridker PM, Hennekens CH, Buring JE, Rifai N 2000 C-reactive protein and other markers of inflammation in the prediction of cardiovascular disease in women. N Engl J Med 342:836–843
Pradhan AD, Manson JE, Rossouw JE, Siscovick DS, Mouton CP, Rifai N, Wallace RB, Jackson RD, Pettinger M, Ridker PM 2002 Inflammatory biomarkers, hormone replacement therapy, and incident coronary heart disease. JAMA 288:980–987
Bhakdi S, Torzewski M, Klouche M, Hemmes M 1999 Complement and atherogenesis: binding of CRP to degraded, nonoxidized LDL enhances complement activation. Arterioscler Thromb Vasc 19:2348–2354
Pasceri V, Willerson JT, Yeh ETH 2000 Direct proinflammatory effect of C-reactive protein on human endothelial cells. Circulation 102:2165–2168
Mikkola TS, Clarkson TB 2002 Review. Estrogen replacement therapy, atherosclerosis, and vascular function. Cardiovasc Res 53:605–619
Ridker PM, Hennekens CH, Rifai N, Buring JE, Manson JAE 1999 Hormone replacement therapy and increased plasma concentration of C-reactive protein. Circulation 100:713–716
Cushman M, Legault C, Barrett-Connor E, Stefanick ML, Kessler, Judd HL, Sakkinen PA, Tracy RP 1999 Effect of postmenopausal hormones on inflammation-sensitive proteins. Circulation 100:717–722
van Baal WM, Kenemans P, Mooren MJ, Kessel H, Emeis JJ, Stehouwer CDA 1999 Increased C-reactive protein levels during short-term hormone replacement therapy in healthy postmenopausal women. Thromb Haemost 81:925–928
Decensi A, Omodei U, Robertson C, Bonanni B, Guerrieri-Gonzaga A, Ramazzotto F, Johansson H, Mora S, Sandri MT, Cazzaniga M, Franchi M, Pecorelli S 2002 Effect of transdermal estradiol and oral conjugated estrogen on C-reactive protein in retinoid-placebo trial in healthy women. Circulation 106:1224–1228
Skouby SO, Gram J, Andersen LF, Sidelmann J, Petersen KR, Jespersen J 2002 Hormone replacement therapy: estrogen and progestin effects on plasma C-reactive protein concentrations. Am J Obstet Gynecol 186:969–977
Silvestri A, Gebara O, Vitale C, Wajngarten M, Leonardo F, Ramires JAF, Fini M, Mercuro G, Rosano GMC 2003 Increased levels of C-reactive protein after oral hormone replacement therapy may not be related to an inflammatory response. Circulation 107:3165–3169
Strandberg TE, Ylikorkala O, Tikkanen MJ 2003 Differing effects of oral and transdermal hormone therapy on cardiovascular risk factors in healthy postmenopausal women. Am J Cardiol 92:212–214
Post MS, van der Mooren MJ, Stehouwer CDA, van Baal WM, Mijatovic V, Schalkwijk CG, Kenemans P 2002 Effects of transdermal and oral oestrogen replacement therapy on C-reactive protein levels in postmenopausal women: a randomized, placebo-controlled trial. Thromb Haemost 88:605–610
Manson JAE, Hsia J, Johnson KC, Rossouw JE, Assaf AR, Lasser NL, Trevisan M, Black HR, Heckbert SR, Detrano R, Strickland OL, Wong ND, Crouse JR, Stein E, Cushman M 2003 Estrogen plus progestin and the risk of coronary heart disease. N Engl J Med 349:523–534
Hulley S, Grady D, Bush T, Furberg C, Herrington D, Riggs B, Vitinghoff E 1998 Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women. JAMA 280:605–613
Sattar N, Perera M, Small M, Lumsden MA 1999 Hormone replacement therapy and sensitive C-reactive protein concentrations in women with type-2 diabetes. Lancet 354:487–488
Modena MG, Bursi F, Fantini G, Cagnacci A, Carbonieri A, Fortuna A, Rossi R 2002 Effects of hormone replacement therapy on C-reactive protein levels in healthy postmenopausal women: comparison between oral and transdermal administration of estrogen. Am J Med 113:31–334
Nachtigall LE 1995 Emerging delivery systems for estrogen replacement, aspects of transdermal and oral delivery. Am J Obstet Gynecol 173:993–997
Ylikorkala O, Evi? S, V?lim?ki M, Tiitinen A 2003 Effects of hormone therapy and alendronate on C-reactive protein, E-selectin, and sex hormone-binding globulin in osteoporotic women. Fertil Steril 80:541–545
Nikander E, Mets?-Heikkil? M, Tiitinen A, Ylikorkala O 2003 Evidence of a lack of effect of a phytoestrogen regimen on the levels of C-reactive protein, E-selectin and nitrate in postmenopausal women. J Clin Endocrinol Metab 88:5180–5185
Wakatsuki A, Okatani Y, Ikenoue N, Fukaya T 2002 Effect of medroxyprogesterone acetate on vascular inflammatory markers in postmenopausal women receiving estrogen. Circulation 105:1436–1439
Davidson KM 1998 Intrahepatic cholestasis of pregnancy. Semin Perinat 22:104–111
Lammert F, Marschall HU, Glantz A, Matern S 2000 Intrahepatic cholestasis of pregnancy: molecular pathogenesis, diagnosis and management. J Hepatol 33:1012–1021
Jacquemin E 2001 Role of multidrug resistance 3 deficiency in pediatric and adult liver disease: one gene for three diseases. Semin Liver Dis 21:551–562
Dixon PH, Weerasekera N, Linton KJ 2000 Heterozygous MDR3 missense mutation associated with intrahepatic cholestasis of pregnancy: evidence for a defect in protein trafficking. Hum Mol Genet 9:1209–1217
Lucena JF, Gallego M, Cardenas S 2003 A multidrug resistance gene mutation causing cholelithiasis, cholestasis of pregnancy and adulthood biliary cirrhosis. Anal Chem 75:1425–1429
Mullenbach R, Linton KJ, Wiltshire S 2003 ABCB4 gene sequence variation in women with intrahepatic cholestasis of pregnancy. J Med Genet 40:e70
Reyes H 1997 Review. Intrahepatic cholestasis. A puzzling disorder of pregnancy. J Gastroenterol Hepatol 12:211–216
Savander M, Ropponen A, Avela K, Weerasekera N, Cormand B, Hirvioja ML, Riikonen S, Lehesjoki AE, Ylikorkala O, Williamson C, Aittom?ki K 2003 Genetic evidence of heterogeneity in intrahepatic cholestasis of pregnancy. Gut 52:1025–1029
Kreek MJ, Weser E, Sleisenger MH, Jeffries GH 1967 Idiopathic cholestasis of pregnancy. N Engl J Med 277:1391–1395
Adlercreutz H, Eisalo A, Heino A, Luukkainen T, Penttil? I, Saukkonen H 1968 Investigations of the effect of an oral contraceptive and its components on liver function, serum proteins, copper, coeruloplasmin and -glutamyl peptidase in postmenopausal women. Scand J Gastroenterol 3:273–284
Drill VA 1974 Review. Benign cholestatic jaundice of pregnancy and benign cholestatic jaundice from oral contraceptives. Am J Obstet Gynecol 119:165–174
Barinas-Mitchell E, Cushman M, Meilahn EN, Tracy RP, Kuller LH 2001 Serum levels of C-reactive protein are associated with obesity, weight gain, and hormone replacement therapy in healthy postmenopausal women. Am J Epidemiol 153:1094–1101(A. Ropponen, K. Aittom?ki)
Liver dysfunction may affect the production and release of C-reactive protein (CRP). We designed a double-blind prospective crossover study involving 40 postmenopausal women with or without a history of intrahepatic cholestasis of pregnancy (ICP), where we compared the basal levels of CRP and their responses to increasing doses of oral and transdermal estradiol (E2), followed by addition of oral medroxyprogesterone acetate (MPA). Serum samples collected at baseline, on the last day of each E2 period, and on the last day of the E2 + MPA combination were assayed for CRP, estrogens, and liver enzymes. There was no difference in basal CRP between the study groups. Both regimens (oral and transdermal E2) were accompanied by significant rises in estrone and E2 concentrations; the former were 16 times higher during the oral than during the transdermal regimen. Oral E2 elevated CRP dose dependently, and this response was unaffected by a history of ICP or the use of MPA. The activities of liver transaminases varied but were in normal ranges during E2 use, in women with and without a history of ICP. In conclusion, the synthesis of CRP is not affected by a history of ICP. It is readily and dose dependently stimulated by oral but not by transdermal E2 in as soon as 2 wk.
Introduction
C-REACTIVE PROTEIN (CRP) is primarily derived from the liver under the control of IL-6 stimulation (1). This protein has been widely studied because small elevations in concentration within the range not indicative of infection predict cardiovascular risk (2, 3, 4, 5). CRP, especially its activated form in the blood vessel wall (3, 6), induces the expression of various adhesion molecules in the endothelium, which in turn accelerates vascular inflammatory reactions and may accelerate the development of atherosclerosis (7, 8). Long-term oral postmenopausal hormone therapy (HT) has been shown to be associated with elevated levels of CRP (5, 9, 10, 11, 12, 13, 14, 15, 16), and some authors feel that this rise in CRP may account for the lack of effect of oral HT in the primary [Women’s Health Initiative (WHI)] (17) or secondary [Heart and Estrogen-Progestin Replacement Study (HERS)] (18) prevention of ischemic cardiac events. Transdermal HT may be safer in this regard because it has been shown to reduce (19, 20) or have no effect on circulating CRP levels (12, 15). The difference in CRP response to oral and transdermal HT is possibly a result of higher estrogenicity in the liver during oral than during transdermal regimens (21). It is also known that a phytoestrogen regimen used as an alternative to HT does not affect the circulating levels of CRP in postmenopausal women (23). Data on the impact of progestins on CRP are controversial because reductions (19), elevations, or no changes (10, 11, 13, 16, 24) in circulating CRP concentrations have been reported after the addition of progestin to an estrogen regimen.
Intrahepatic cholestasis of pregnancy (ICP), which complicates approximately one in 100 pregnancies, manifests itself as itching skin and elevated activities of liver transaminases, usually in the third trimester of pregnancy (25, 26). The cause of ICP is largely unknown, although some women with mutations in the ABCB4 gene have been described (27, 28, 29, 30). ICP is most likely an etiologically heterogeneous condition, which has been suggested to be a consequence of insufficient hepatic capacity to metabolize the high amounts of placenta-derived sex steroids in genetically predisposed individuals (31). The condition disappears soon after delivery but the increased risk of gallstones (26, 32) suggests permanent hepatobiliary dysfunction in these women. This is also supported by data indicating that the use of high-dose oral contraceptives often triggers elevations in liver enzyme activities in these women (33, 34, 35). Whether postmenopausal HT affects liver function in women with a previous history of ICP has not to our knowledge been studied before.
No data exist on CRP in women with a history of ICP. Moreover, in previous studies on HT and CRP (5, 9, 10, 11, 12, 13, 14, 15, 16), CRP has been assayed at 1–6 months after the start of HT. Thus, we do not yet know how soon CRP may be stimulated after the start of HT. Therefore, we designed a trial to assess the effects of relatively short-term increasing doses of oral and transdermal estradiol (E2) treatment on CRP, with and without progestin, in women with and without a history of ICP.
Subjects and Methods
Subjects
With the permission of the local Ethics Committee, we studied 40 postmenopausal women who had entered the menopause 6–7 yr before the study (Table 1). The volunteers received thorough written and verbal information on the conduct of the study, and an informed consent was obtained from all of them. They all complained of hot flashes and other climacteric symptoms, and 28 women had previously used HT; this had been stopped at least 4 wk before recruitment. The women were otherwise healthy and reported no allergic skin reaction to patches, but two of them used calcium channel blockers, and two other women used angiotensin-converting enzyme blockers to control hypertension. Three other women used thyroxine. None of these drugs should have any effect on liver function. The baseline characteristics of the patients and controls are shown in Table 1. Twelve women had undergone hysterectomy, and both ovaries had been removed in nine of them. All women had a history of at least one full-term pregnancy, and this was complicated by ICP in 20 women (in 13 women repeatedly). On average, ICP had occurred 30 yr (26–33 yr) before the study. The diagnosis of ICP had been based on typical skin itching and on elevations in liver enzyme activities in the third trimester of pregnancy; all symptoms and signs had vanished 6 wk after delivery.
Study design
The women were randomized to receive increasing doses (2 mg for 14 d; 4 mg for 14 d) of E2 either orally (Progynova, estradiol valerate, Schering AG, Berlin, Germany) or transdermally from a patch (50 μg/24 h for 14 d; 100 μg/24 h for 14 d) (FemSeven, Merck KgaA, Darmstadt, Germany) (Fig. 1). The study was blind in that during both treatments, each woman took tablets (active or placebo) and used patches (placebo or active). After using E2 only for 4 wk, each subject took 10 mg medroxyprogesterone acetate (MPA) (Lutopolar, Orion Pharma, Espoo, Finland) for 14 d concomitantly with the highest E2 dose (Fig. 1). After a 4-wk washout period, the subjects crossed over to the other treatment (Fig. 1). Thus, each woman used E2 both orally and transdermally. The women were seen at baseline (also after the 4-wk washout period) and at the end of each 2-wk treatment period, and blood samples were collected on these occasions (a total of eight times during the trial). Blood samples were drawn after an overnight fast, and centrifugally separated serum was stored at –20 C until analyzed. General and pelvic examinations were carried out before and after both 6-wk treatment periods.
Measurements
Serum CRP was assayed immunochemically by use of a monoclonal antibody specific to human CRP (Sensitive CRP IEMA TEST, Oy Medix Biochemica Ab, Kauniainen, Finland). To minimize the impact of interassay variation, the CRP assessments were carried out in one assay batch. The coefficient of intraassay variation was less than 5%. Because serum CRP concentrations rise in acute infection, we excluded those values exceeding 10 mg/liter (nine values altogether). Levels of estrone (E1) were determined by RIA (20 subjects) (Estron, Diagnostic Systems Laboratories, Sinsheim, Germany), and levels of E2 were measured by means of time-resolved fluoroimmunoassay (DELFIA Estradiol, Wallac Oy, Turku, Finland) (40 subjects). The E2 antibody shows a 0.75% crossreactivity with E1. The activities of liver enzymes, such as aspartate transaminase, alanine transaminase, and -glutamyl transferase, were measured by means of routine laboratory methods.
Statistical analyses
Analyses were carried out using SPSS version 11.0 (SPSS Inc., Chicago, IL). Data are expressed as medians (with 95% confidence intervals according to Altman’s Statistics). Because we used the crossover design, the possibility of a period effect was tested by Mann-Whitney test, where we compared the differences between the periods in the two groups of patients (those beginning with the oral and those beginning with the transdermal E2). No period effect was detected. The possibility of a treatment-period interaction was investigated by Mann-Whitney test, where we compared the average responses to the two treatments and found patients’ average responses to the two treatments to be the same regardless of the order of treatments. Because of the skewed and not normal distribution of CRP, E2, and E1 data, a nonparametric test (Wilcoxon’s signed rank test) was used to compare values at baseline and during follow-up. To better illustrate the responses of CRP with HT, we calculated a change in CRP from basal to 2 (to 4 and to 6) wk of treatment. This change is also expressed as percentages. Nonparametric repeated measures ANOVA (Friedman test) was used to test variation among groups during treatment, and Wilcoxon’s signed-rank test was used to test changes between the groups. Correlation analyses were performed using Spearman’s nonparametric correlation coefficient. A two-tailed P < 0.05 was accepted as the level of significance.
Results
The study groups were comparable in age, body mass index, and other clinical variables at baseline (Table 1). Baseline levels of CRP, E1, E2, and liver enzyme activities also did not differ (Table 2). One subject in the control group used nimesulid (COX-2 inhibitor) during the study, and owing to its possible effect on liver function, she was excluded from data analysis. None of the subjects experienced any side effects during treatment. All women reported reductions in hot flashes, and nonhysterectomized women experienced withdrawal bleeding after the MPA course.
The order of the HT regimen (oral or transdermal first) had no influence on the biochemical changes. Biomarkers at baseline and during the trial in four women using antihypertensive drugs did not differ from those not using antihypertensive drugs; therefore, no further subgroup analysis was done.
Oral E2 led to significant dose-dependent rises (51 and 87%) in CRP levels in the ICP group, but these rises did not differ from those in the control group (39 and 95%) (Table 3). Median dose-dependent rises ranged between 49 and 91% when both ICP and control women were analyzed as one group. The rises were not significantly affected by the addition of MPA in either group (Table 3).
Transdermal E2 did not affect the level of CRP in either group, even at the highest dose. A trend toward a fall in CRP concentrations was seen, and this was augmented by MPA (Table 3).
Smoking, body mass index, age, and previous use of HT were not determinants of baseline CRP or CRP responses to oral E2.
The absolute median rises (with 95% confidence intervals) in serum E1 concentrations during oral E2 treatment were 539 ng/liter (360–871) [1994 pmol/liter (1332–3223)] and 963ng/liter (462-2656) [3561 pmol/liter (1709–9824)] in ICP and control women, respectively. Transdermal E2 did not result in significantly increased E1 concentrations. The relative rises of serum E1 levels during use of oral and transdermal E2 were 2030 and 24% in the ICP group and 1470 and 55% in the control group (data not shown).
Oral E2 led to similar rises in serum E2 levels in the ICP and control group (Table 4). These elevations were greater during oral treatment than during transdermal administration, with no difference between the groups. Oral E2 at 2 mg/d and transdermal E2 at 100 μg/d led to similar rises in the circulating levels of E2 (Table 4).
During oral E2, the changes in CRP concentrations correlated positively with those in E2 levels (r = 0.326, P = 0.007; Fig. 2) but not with those in E1 levels (r = 0.08, P = 0.7). During transdermal E2, no correlations were seen between changes in circulating concentrations of CRP, E1, or E2.
Oral and transdermal E2 administration was not accompanied by significant changes in liver enzyme activities. The levels fluctuated but remained within the normal range in each subject.
Discussion
The liver and enterohepatic circulation play a key role in estrogen metabolism, and this ultimately determines the estrogenicity of a given HT regimen (21). Therefore, liver dysfunction, even if subclinical, may become a factor in the metabolism of estrogen in postmenopausal women. The liver also produces CRP, an important marker, and perhaps even a contributing factor as regards cardiovascular disease (1). It is stimulated by oral estrogens, given for 1–6 months (5, 9, 10, 15, 16). The possible role of progestin on CRP has remained uncertain (10, 13, 19, 24). We studied the effects of short-term use of E2 given either orally or transdermally on CRP in women with and without a history of ICP. We first gave increasing doses of E2 alone, both orally and transdermally for 4 wk, followed by the addition of a routine course of oral MPA for the next 2 wk. This study design allowed us to determine the time and dose dependency of the CRP-stimulating effect of E2. Moreover, the effect of MPA on CRP when added to oral or transdermal E2 could be studied.
All study subjects tolerated both oral and transdermal HT well, and none discontinued the regimen. Moreover, liver transaminase activities remained within normal ranges in both groups. As expected, both oral and transdermal E2 led to rises in the serum levels of E1 and E2. The response of E1 was much greater during oral than transdermal treatment, but E1 and E2 levels showed no differences between women with and without a history of ICP. This novel finding can be seen as evidence that in women with a history of ICP the liver can metabolize postmenopausal oral and transdermal E2 normally. Possible liver dysfunction in women with a history of ICP may be so minimal that it could not be triggered by the doses of E2 we used in our study or previously when 12 of 20 such women had used HT with no known untoward side effects. However, our data cannot exclude the possibility that more prolonged use and even higher doses of HT would have had a different effect.
These are the first data showing that oral E2 at 2 mg/d is capable of inducing rises in serum CRP concentrations as soon as within 2 wk. The response of CRP in our study was of the same order as in some previous ones where HT has been administered for periods of 1–6 months (10, 11, 12, 14, 16, 20). We also demonstrated that by increasing the dose of E2 from 2–4 mg/d we could further increase the level of CRP, although it is possible that some of this could be due to a prolonged effect of the preceding treatment with 2 mg of E2 as previously suggested by van Baal (16). The increase in CRP concentration was related to the rises in E2 only during oral use, lending support to the role of the liver in E2-induced elevation of CRP concentrations. Moreover, our data demonstrate that the effect of oral E2 on CRP vanishes totally in 4 wk and that a history of ICP has no effect on CRP levels. In contrast, transdermal E2 failed to affect CRP levels, although the dose was raised to 100 μg/d; if anything, a decreasing trend in serum CRP concentrations was seen. This was unexpected because the circulating levels of E2 were similar during oral administration of E2 at 2 mg/d and during use of transdermal E2 at 100 μg/d. Perhaps CRP stimulation in the liver also requires the presence of E1 (36), resulting in no rise in serum CRP levels during transdermal use of E2.
Previous data are not uniform as regards the effects of various progestins on CRP levels. The protocols include sequential use of micronized progesterone (10), sequential use of dydrogesterone (11), continuous use of trimegestrone (11), norethisterone acetate (13, 15, 22), gestodene (16) and MPA (2.5 mg) (10, 14, 24), and intrauterine administration of levonorgestrel (13). In general, they have had no effect on estrogen-induced rises in CRP levels, but continuous use of MPA (5 mg) (24), gestodene (25 μg) (16), or sequential use of MPA (20 mg) (13) has been reported to abolish oral E2-induced rises in CRP concentrations. We demonstrate here that MPA at 10 mg/d for 2 wk did not affect CRP levels, which were already stimulated by preceding and concomitant E2 use. Our data may be of significance because MPA has been suggested to be one factor in the lack of protective effect of conjugated equine estrogen + MPA in primary (WHI) (17) and secondary (HERS) (18) prevention of cardiac events.
Taken together, a history of ICP is not accompanied by any differences in basal CRP or in its responses to HT. Oral E2 stimulates CRP synthesis dose dependently within 2 wk, both in women with and without a history of ICP, and the addition of short-term progestin (MPA) had no effect.
Footnotes
This study was supported by grants from the research funds of the Helsinki University Central Hospital.
First Published Online November 2, 2004
Abbreviations: CRP, C-reactive protein; E1, estrone; E2, estradiol; HT, hormone therapy; ICP, intrahepatic cholestasis of pregnancy; MPA, medroxyprogesterone acetate.
Received April 26, 2004.
Accepted October 25, 2004.
References
Straub RH, Hense HW, Andus T, Sch?lmerich J, Riegger GAJ, Schunkert H 2000 Hormone replacement therapy and interrelation between serum interleukin-6 and body mass index in postmenopausal women: a population-based study. J Clin Endocrinol Metab 85:1340–1344
Koenig W, Sund M, Fr?hlich M, Fischer HG, L?wel H, D?ring A, Hutchinson WL, Pepys MB 1999 C-reactive protein, a sensitive marker of inflammation, predicts future risk of coronary heart disease in initially healthy middle-aged men. Circulation 99:237–242
Lagrand WK, Wisser CA, Hermens WT, Niessen HWM, Verheugt FWA, Wolbink GJ, Hack CE 1999 C-reactive protein as a cardiovascular risk factor. Circulation 100:96–102
Ridker PM, Hennekens CH, Buring JE, Rifai N 2000 C-reactive protein and other markers of inflammation in the prediction of cardiovascular disease in women. N Engl J Med 342:836–843
Pradhan AD, Manson JE, Rossouw JE, Siscovick DS, Mouton CP, Rifai N, Wallace RB, Jackson RD, Pettinger M, Ridker PM 2002 Inflammatory biomarkers, hormone replacement therapy, and incident coronary heart disease. JAMA 288:980–987
Bhakdi S, Torzewski M, Klouche M, Hemmes M 1999 Complement and atherogenesis: binding of CRP to degraded, nonoxidized LDL enhances complement activation. Arterioscler Thromb Vasc 19:2348–2354
Pasceri V, Willerson JT, Yeh ETH 2000 Direct proinflammatory effect of C-reactive protein on human endothelial cells. Circulation 102:2165–2168
Mikkola TS, Clarkson TB 2002 Review. Estrogen replacement therapy, atherosclerosis, and vascular function. Cardiovasc Res 53:605–619
Ridker PM, Hennekens CH, Rifai N, Buring JE, Manson JAE 1999 Hormone replacement therapy and increased plasma concentration of C-reactive protein. Circulation 100:713–716
Cushman M, Legault C, Barrett-Connor E, Stefanick ML, Kessler, Judd HL, Sakkinen PA, Tracy RP 1999 Effect of postmenopausal hormones on inflammation-sensitive proteins. Circulation 100:717–722
van Baal WM, Kenemans P, Mooren MJ, Kessel H, Emeis JJ, Stehouwer CDA 1999 Increased C-reactive protein levels during short-term hormone replacement therapy in healthy postmenopausal women. Thromb Haemost 81:925–928
Decensi A, Omodei U, Robertson C, Bonanni B, Guerrieri-Gonzaga A, Ramazzotto F, Johansson H, Mora S, Sandri MT, Cazzaniga M, Franchi M, Pecorelli S 2002 Effect of transdermal estradiol and oral conjugated estrogen on C-reactive protein in retinoid-placebo trial in healthy women. Circulation 106:1224–1228
Skouby SO, Gram J, Andersen LF, Sidelmann J, Petersen KR, Jespersen J 2002 Hormone replacement therapy: estrogen and progestin effects on plasma C-reactive protein concentrations. Am J Obstet Gynecol 186:969–977
Silvestri A, Gebara O, Vitale C, Wajngarten M, Leonardo F, Ramires JAF, Fini M, Mercuro G, Rosano GMC 2003 Increased levels of C-reactive protein after oral hormone replacement therapy may not be related to an inflammatory response. Circulation 107:3165–3169
Strandberg TE, Ylikorkala O, Tikkanen MJ 2003 Differing effects of oral and transdermal hormone therapy on cardiovascular risk factors in healthy postmenopausal women. Am J Cardiol 92:212–214
Post MS, van der Mooren MJ, Stehouwer CDA, van Baal WM, Mijatovic V, Schalkwijk CG, Kenemans P 2002 Effects of transdermal and oral oestrogen replacement therapy on C-reactive protein levels in postmenopausal women: a randomized, placebo-controlled trial. Thromb Haemost 88:605–610
Manson JAE, Hsia J, Johnson KC, Rossouw JE, Assaf AR, Lasser NL, Trevisan M, Black HR, Heckbert SR, Detrano R, Strickland OL, Wong ND, Crouse JR, Stein E, Cushman M 2003 Estrogen plus progestin and the risk of coronary heart disease. N Engl J Med 349:523–534
Hulley S, Grady D, Bush T, Furberg C, Herrington D, Riggs B, Vitinghoff E 1998 Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women. JAMA 280:605–613
Sattar N, Perera M, Small M, Lumsden MA 1999 Hormone replacement therapy and sensitive C-reactive protein concentrations in women with type-2 diabetes. Lancet 354:487–488
Modena MG, Bursi F, Fantini G, Cagnacci A, Carbonieri A, Fortuna A, Rossi R 2002 Effects of hormone replacement therapy on C-reactive protein levels in healthy postmenopausal women: comparison between oral and transdermal administration of estrogen. Am J Med 113:31–334
Nachtigall LE 1995 Emerging delivery systems for estrogen replacement, aspects of transdermal and oral delivery. Am J Obstet Gynecol 173:993–997
Ylikorkala O, Evi? S, V?lim?ki M, Tiitinen A 2003 Effects of hormone therapy and alendronate on C-reactive protein, E-selectin, and sex hormone-binding globulin in osteoporotic women. Fertil Steril 80:541–545
Nikander E, Mets?-Heikkil? M, Tiitinen A, Ylikorkala O 2003 Evidence of a lack of effect of a phytoestrogen regimen on the levels of C-reactive protein, E-selectin and nitrate in postmenopausal women. J Clin Endocrinol Metab 88:5180–5185
Wakatsuki A, Okatani Y, Ikenoue N, Fukaya T 2002 Effect of medroxyprogesterone acetate on vascular inflammatory markers in postmenopausal women receiving estrogen. Circulation 105:1436–1439
Davidson KM 1998 Intrahepatic cholestasis of pregnancy. Semin Perinat 22:104–111
Lammert F, Marschall HU, Glantz A, Matern S 2000 Intrahepatic cholestasis of pregnancy: molecular pathogenesis, diagnosis and management. J Hepatol 33:1012–1021
Jacquemin E 2001 Role of multidrug resistance 3 deficiency in pediatric and adult liver disease: one gene for three diseases. Semin Liver Dis 21:551–562
Dixon PH, Weerasekera N, Linton KJ 2000 Heterozygous MDR3 missense mutation associated with intrahepatic cholestasis of pregnancy: evidence for a defect in protein trafficking. Hum Mol Genet 9:1209–1217
Lucena JF, Gallego M, Cardenas S 2003 A multidrug resistance gene mutation causing cholelithiasis, cholestasis of pregnancy and adulthood biliary cirrhosis. Anal Chem 75:1425–1429
Mullenbach R, Linton KJ, Wiltshire S 2003 ABCB4 gene sequence variation in women with intrahepatic cholestasis of pregnancy. J Med Genet 40:e70
Reyes H 1997 Review. Intrahepatic cholestasis. A puzzling disorder of pregnancy. J Gastroenterol Hepatol 12:211–216
Savander M, Ropponen A, Avela K, Weerasekera N, Cormand B, Hirvioja ML, Riikonen S, Lehesjoki AE, Ylikorkala O, Williamson C, Aittom?ki K 2003 Genetic evidence of heterogeneity in intrahepatic cholestasis of pregnancy. Gut 52:1025–1029
Kreek MJ, Weser E, Sleisenger MH, Jeffries GH 1967 Idiopathic cholestasis of pregnancy. N Engl J Med 277:1391–1395
Adlercreutz H, Eisalo A, Heino A, Luukkainen T, Penttil? I, Saukkonen H 1968 Investigations of the effect of an oral contraceptive and its components on liver function, serum proteins, copper, coeruloplasmin and -glutamyl peptidase in postmenopausal women. Scand J Gastroenterol 3:273–284
Drill VA 1974 Review. Benign cholestatic jaundice of pregnancy and benign cholestatic jaundice from oral contraceptives. Am J Obstet Gynecol 119:165–174
Barinas-Mitchell E, Cushman M, Meilahn EN, Tracy RP, Kuller LH 2001 Serum levels of C-reactive protein are associated with obesity, weight gain, and hormone replacement therapy in healthy postmenopausal women. Am J Epidemiol 153:1094–1101(A. Ropponen, K. Aittom?ki)