REPRODUCTIVE ENDOCRINOLOGY

Transcription

1 REPRODUCTIVE ENDOCRINOLOGY FERTILITY AND STERILITY VOL. 80, NO. 5, NOVEMBER 2003 Copyright 2003 American Society for Reproductive Medicine Published by Elsevier Inc. Printed on acid-free paper in U.S.A. Effect of gonadotropin-releasing hormone agonist and medroxyprogesterone acetate on calcium metabolism: a prospective, randomized, double-blind, placebocontrolled, crossover trial Bruce R. Carr, M.D., a Neil A. Breslau, M.D., b,c Noel Peng, M.D., a Beverley Adams-Huet, M.S., b Karen D. Bradshaw, M.D., a and Michael P. Steinkampf, M.D. a Division of Reproductive Endocrinology and Infertility, Department of Obstetrics and Gynecology, and Center for Mineral Metabolism and Clinical Research and Department of Internal Medicine at The University of Texas Southwestern Medical Center, Dallas, Texas Received January 3, 2003; revised and accepted April 3, This Clinical Research Center Study was supported by NIH grants RO1 HD and MO1- RR Dr. Bruce Carr had grant support and was on the speaker s bureau of TAP Pharmaceuticals. Reprints requests: Dr. Bruce R. Carr, Department of Obstetrics and Gynecology, University of Texas Southwestern Medical Center, 5323 Harry Hines Boulevard, Dallas, TX (FAX: ; bruce.carr@utsouthwestern. edu). a Division of Reproductive Endocrinology and Infertility, Department of Obstetrics and Gynecology, The University of Texas Southwestern Medical Center. b Center for Mineral Metabolism and Clinical Research and Department of Internal Medicine, The University of Texas Southwestern Medical Center. c Baylor University Medical Center, Houston, Texas /03/$30.00 doi: /s (03) Objective: The purpose of this study was to prospectively compare the effectiveness of administering medroxyprogesterone acetate (MPA; 20 mg/d) in either the first (protocol A) or last (protocol B) 12-week period as well as a 6-month course of the GnRH agonist (GnRH-a; leuprolide acetate; 1 mg/d, SC) on calcium (Ca) metabolism. Design: Prospective, randomized, double-blind, placebo-controlled, crossover trial. Setting: Clinical research center, university hospital. Patient(s): Twenty women were randomized into protocol A or B, received either MPA or placebo along with GnRH-a, and were then crossed over at 12 weeks to placebo or MPA, for the final 12-week interval of GnRH-a therapy. Intervention(s): Collection of serum and urine samples and measurement of bone density. Main Outcome Measure(s): Sex hormone, calcitropic hormone, and bone density studies were performed at baseline and at 12 and 24 weeks. Result(s): In both protocol A and B, LH and E 2 levels declined by 79% 81% and 83% 90% of the baseline, respectively, at 12 and 24 weeks. Serum Ca, phosphorus, alkaline phosphatase, and osteocalcin; 2-h fasting and 24-h urinary Ca excretion; and urinary hydroxyproline levels all increased significantly during GnRH-a treatment alone. Estimated Ca balance decreased significantly during GnRH-a treatment alone. The addition of MPA attenuated the increases in phosphorus, alkaline phosphatase, osteocalcin, and 2-h fasting and 24-h urinary Ca excretion, and the decrease in estimated Ca balance. Comparison of phase order demonstrated that MPA prevented 24-h urinary Ca excretion and urinary hydroxyproline loss and decline in estimated Ca balance when it was added back during the second 12 weeks (protocol B) but not during the first 12 weeks (protocol A). Conclusion(s): We conclude that sequential MPA appears to reverse in part the negative effects of GnRH-a on calcitropic hormones and estimated Ca balance. (Fertil Steril 2003;80: by American Society for Reproductive Medicine.) Key Words: Gonadotropin-releasing hormone agonist, medroxyprogesterone acetate, calcium A number of studies have shown that GnRH agonists (GnRH-a) reduce the symptoms and size of lesions associated with endometriosis, and reduce the size of uterine leiomyomata and associated bleeding disorders (1, 2). However, long-term treatment with GnRH-a results in symptoms of estrogen (E) deficiency and bone loss. The use of add-back therapy with synthetic progestins has been proposed to prevent some of the E-lowering effects of GnRH-a, particularly on calcium metabolism and bone mineral density (1 8). A few studies have investigated the effect of GnRH-a plus add-back therapy using either medroxyprogesterone acetate (MPA) or nor- 1216

2 ethindrone on calcium metabolism and bone density in premenopausal women presenting with endometriosis. Some results suggest modest but significant effects of norethindrone and MPA on bone mineral content; however, other studies have failed to demonstrate any benefit ofmpaon bone mineral content (1, 5, 6). Some of these differences in bone mineral content may have been due to study design, duration of therapy, or dose of progestin. In addition, these studies did not investigate measurements of calcium homeostasis in depth. In contrast to the model of GnRH-a plus add-back MPA, the use of depot-mpa as a contraceptive is associated in some studies but not all with a modest reduction of bone mineral density (9 12). The use of MPA alone in the absence of E in postmenopausal women appears to inhibit cortical but not trabecular bone loss (13). In postmenopausal women treated with E, the addition of progestin has minimal additional effect on bone density (14, 15). We investigated the effect of the addition of MPA in GnRH-a treated women on calcium metabolism in a doubleblind, placebo controlled, crossover study. MATERIALS AND METHODS Subjects Twenty menstruating women, years of age, presenting with either uterine leiomyomata (n 13) or endometriosis (n 7), signed an informed consent and were randomized into a double-blind, placebo-controlled, crossover designed study, which was approved by the Institutional Review Board of the University of Texas Southwestern Medical Center at Dallas. The design of the study was previously published (8). In brief, 10 women (protocol A) were randomized to receive an initial 12 weeks of MPA (20 mg/d) followed by an additional 12 weeks of placebo. Ten other women (protocol B) were randomized to receive an initial 12 weeks of placebo followed by 12 weeks of MPA (20 mg/d). All women received leuprolide acetate (1 mg/ day) SC injections for 24 weeks. Exclusions included recent use of oral contraceptive (OC) pills, heparin, E, GnRH-a, and history of oligomenorrhea or amenorrhea. Hormonal Studies Serum samples were obtained at baseline (days 2 7ofthe cycle) and at 12 and 24 weeks, aliquoted, and stored at 20 C until assayed. The FSH and LH levels were measured using LKB fluoroimmunometric assay (Delphia assay kit, Wallac, Columbus, MO). The sensitivity of the FSH assay is 0.05 U/L (intra-assay variation, 4.8% 8.6%; interassay variation, 3.7% 8.7%). The sensitivity of the LH assay is 0.12 U/L (intra-assay variation, 2.5% 5.9%; interassay variation, 6.7% 7.9%). Estradiol was measured using a no-extraction, solid phase 125 I RIA supplied by Diagnostic Products Corporation (Los Angeles, CA). The sensitivity of the assay was 0.05 nmol/l (intra-assay variation, 4.0% 7.0%; interassay variation, 4.2% 8.1%). Calcitropic Hormones At baseline and at 12 and 24 weeks, patients were admitted to the General Clinical Research Center at the University of Texas Southwestern Medical Center at Dallas. During the 4-day hospitalization, patients were placed on a constant metabolic diet containing 10.0 mmol of calcium (Ca), 26 mmol of phosphorus, and 100 mmol of sodium daily. On day 1, fractional intestinal 47 Ca absorption ( ) was determined. Fasting venous blood was obtained daily for 3 days for the measurement of serum Ca, phosphorus, alkaline phosphatase, and creatinine. On day 4, the fast and Ca load test was performed. Serum immunoreactive parathyroid hormone (PTH), vitamin D metabolites, and osteocalcin levels were determined on days 1 and 4. Urine was collected in 24-hour pools during 3 days for volume, creatinine, Ca, phosphorus, and hydroxyproline measurements (16). Analysis Fractional intestinal 47 Ca absorption ( ) was measured from fecal recovery of 47 Ca given orally with 2.5 mmol Ca carrier contained in a synthetic meal (17). In 15 normal subjects the mean was 49% 6%. When the measurement was repeated 2 months later, the average relative variability (calculated as the absolute difference between the two measures/mean 100) was 2.60% 1.95% ( SD). Serum Ca, phosphorus, alkaline phosphatase, and creatinine were measured as part of a multichannel screen. Urinary fasting Ca excretion and response to a 1-g Ca load were determined as described previously (16). Serum intact PTH (intra-assay variation, 1.4% 5.4%; interassay variation, 3.6% 7.6%) was measured by the immunoradiometric assay with antiserum with micromolecular specificity (Endocrine Metabolic Center, Oakland, CA). Serum vitamin D metabolites (25-hydroxyvitamin D [25- (OH)D] and 1,25-dihydroxyvitamin D [1, 25-(OH) 2 D]) were quantitated after resolution and purification on high performance liquid chromatography (16). The intra-assay and interassay variations were 8% and 15%, respectively. Serum osteocalcin was determined by an RIA kit (IncStar, Stillwater, MN) (intra-assay variation, 4.8%; interassay variation, 6.1%). Urinary Ca (U Ca ) was determined by atomic absorption spectrophotometry and phosphorus by the method of Fiske and Subbanow (18). Based on 10 normal subjects who had two sets of three 24-hour U Ca measurements collected 2 weeks apart on the constant metabolic diet and not taking any drugs, the average relative variability of the mean 24- hour U Ca levels was 7% 3% (16). Urinary hydroxyproline (U OHP ) was determined according to the method of Kivirikko et al. (19). These measurements were performed at baseline, 12 weeks, and 24 weeks (mean SD). FERTILITY & STERILITY 1217

3 TABLE 1 Changes in Ca homeostasis in women at baseline and during either GnRH-a alone or GnRH-a plus MPA treatment. Baseline GnRH-a GnRH-a MPA n Mean SD Median Mean SD Median P value vs. baseline Mean SD Median P value vs. baseline P value vs. GnRH-a alone Serum Ca (mg/dl) Phosphorus (mg/dl) Alk phos (IU/L) Osteocalcin (ng/ml) PTH (pg/ml) (OH)D (ng/ml) ,25-(OH) 2 D (pg/ml) Urine 24-h U Ca (mg/d) h U OHP (mg/d) h Fasting urinary Ca (mg/dl GF) Intestinal 47 Ca absorption, fraction Estimated Ca balance (mg/d) Estimated Ca Balance The Ca balance was estimated as the difference between the absorbed and mean U Ca (A Ca U Ca ) (20). Absorbed Ca was determined by multiplying the fractional intestinal 47 Ca absorption ( ) by the 400-mg daily Ca intake. This estimate does not take into account fecal Ca secretion, which is approximately 100 mg/d. The measurement was performed at baseline, 12, and 24 weeks and is expressed in milligrams per day (mean SD). Bone Density Studies Bone density studies were obtained at baseline, 12, and 24 weeks. Bone density of distal third of the radius was measured by 125 I photon absorptiometry (Norland Cameron bone mineral analyzer, Norland Instruments, Fort Atkinson, WI). The same instrument was used to measure bone density in each patient before and after treatment. The results are presented as grams per cm 2 (mean SD). Statistical Analysis Differences between the baseline, GnRH-a, and GnRH-a MPA were assessed using a two-factor repeated-measures analysis of variance for serum, urine, and other variables of calcium metabolism. A between-group factor to indicate treatment sequence (phase order difference), protocol A or B, was included in the model. Because an interaction between protocol and treatment phase was found for a few responses, comparisons of treatment phases within each protocol were made using the least square means contrasts derived from the analysis of variance model. Results are presented as mean and standard deviation. Statistical analysis was performed with SAS v8.2 (SAS Institute, Cary, NC). RESULTS Randomization of subjects provided no significant difference between weight in women in protocol A (67 16 kilograms [mean SD]) vs. protocol B (61 18 kg) or age in protocol A (32 4 years) vs. protocol B (31 9 years). To evaluate the effectiveness and compliance of GnRH-a on hormonal milieu, LH, FSH, and E 2 levels were evaluated. Serum LH levels declined approximately by 79% 81% in both protocols by 12 and 24 weeks (P.01) (data not shown). There were no differences between protocols A and B. Serum FSH levels declined 18% 25% in both protocols by 12 and 24 weeks, but these results and the differences between protocols A and B were not statistically significant. Serum E 2 decreased by 83% 90% by 12 and 24 weeks (P.01 compared to baseline). The differences between protocols were not significant. Parameters of Ca Homeostasis Changes in Ca homeostasis were evaluated by assessing the data during GnRH-a alone treatment or during GnRH-a plus MPA treatment. In addition, we compared the phase order in this crossover study, that is, whether MPA was given during first 12 weeks (protocol A) or during second 12 weeks (protocol B). In general there were few phase order differences and where significant, these were noted in the text. As seen in Table 1, serum Ca increased from Carr et al. GnRH-a and MPA on Ca metabolism Vol. 80, No. 5, November 2003

4 FIGURE 1 The effect of GnRH agonist in the presence or absence of medroxyprogesterone acetate (MPA) on serum phosphorus levels at 12 and 24 weeks. Protocol A: an initial 12 weeks of MPA (20 mg/d) followed by an additional 12 weeks of placebo; protocol B: an initial 12 weeks of placebo followed by 12 weeks of MPA (20 mg/d). All women received leuprolide acetate (1 mg/d) SC injections for 24 weeks. *P.05 compared to baseline; P.001 compared to baseline; ( )P.001 compared to 12 weeks. Points and error bars represent mean and standard deviation. mg/dl (mean SD) to mg/dl during GnRH-a alone treatment and mg/dl during GnRH-a plus 12 weeks of MPA. The observed increase in Ca in both treatment groups was statistically significant compared to baseline. Serum phosphorus levels parallel those of Ca, increasing from mg/dl at baseline to mg/dl in GnRH-a alone treated subjects (P.0001 vs. baseline) and mg/dl during GnRH-a plus MPA treatment (P.0001 from baseline) (Table 1). However, the addition of MPA attenuated the increase in phosphorus levels compared to GnRH-a alone (P.005). When evaluating the effect of phase order of MPA on phosphorus levels (i.e., protocol A vs. protocol B), the use of MPA plus GnRH-a for the first 12 weeks (protocol A) suppressed phosphorus levels compared to protocol B (the absence of the MPA from baseline to 12 weeks). When phosphorus was measured at 24 weeks, the addition of MPA significantly lowered levels (P.001) from that observed at 12 weeks (Fig. 1). Alkaline phosphatase levels, a marker of bone turnover, increased during GnRH-a alone treatment compared to baseline (80 16 IU/L; IU/L, respectively) and these differences were statistically significant, P.001 (Table 1). In contrast, during treatment with GnRH-a plus MPA, alkaline phosphatase levels did not increase (74 18 IU/L), and this was significantly less compared to GnRH-a treatment alone (P.02). Serum osteocalcin, also representing an estimate of bone turnover, increased significantly in the GnRH-a alone treatment period compared to baseline ( ng/ml, ng/ml, respectively, P.001). In contrast, during the GnRH-a plus 3 months MPA treatment, serum osteocalcin ( ng/ml) was unchanged from baseline, and the level was less compared to GnRH-a treatment alone (P.0005). Serum PTH did not change significantly between groups. The vitamin D status was also evaluated and there were no significant differences in 25-(OH)D levels, but 1,25- (OH) 2 D levels declined significantly from baseline in both groups (29 9 pg/ml in GnRH-a alone and 29 8 pg/ml in GnRH-a plus MPA treatment group) (P.0001). There was no difference between treatment groups (Table 1). Next, we evaluated 2-hour fasting U Ca, 24-hour U Ca, and urinary OHP levels (Table 1). The 2-hour fasting U Ca increased significantly during GnRH-a alone ( mg/dl GF) and GnRH-a plus MPA ( mg/dl) treatment compared to baseline ( mg/dl) (P.0001). However, during GnRH-a plus MPA treatment, this increase was significantly less compared to GnRH-a alone (P.0006). With respect to the 24-hour excretion of U Ca, there was significant urinary loss of Ca during the GnRH-a alone treatment group compared to baseline ( mg/d vs mg/d; respectively, P.001). During GnRH-a plus MPA treatment the levels of 24-hour U Ca also increased significantly from baseline ( mg/d, FERTILITY & STERILITY 1219

5 FIGURE 2 The effect of GnRH agonist in the presence or absence of medroxyprogesterone acetate (MPA) on 24-hour urinary calcium excretion levels at 12 and 24 weeks. Protocol A: an initial 12 weeks of MPA (20 mg/d) followed by an additional 12 weeks of placebo; protocol B: an initial 12 weeks of placebo followed by 12 weeks of MPA (20 mg/d). All women received leuprolide acetate (1 mg/d) SC injections for 24 weeks. *P.05 compared to baseline; P.001 compared to baseline; ( )P.001 compared to 12 weeks. Points and error bars represent mean and standard deviation. P.001), but the effect of add-back MPA reduced the loss of U Ca compared to GnRH-a alone treatment (P.008). When these data were evaluated by phase order (i.e., protocol A vs. B over time), MPA did not prevent U Ca loss when used initially with GnRH-a (protocol A) vs. protocol B (GnRH-a alone) at 12 weeks. When MPA was added back during the second 12 weeks (protocol B), Ca loss decreased significantly by 24 weeks (P.001) (Fig. 2). Urinary OHP, an estimate of bone turnover, increased with GnRH-a alone treatment compared to baseline (25 7 mg/d vs mg/d; respectively, P.0009), whereas U OHP levels associated GnRH-a plus MPA treatment did not differ from baseline (23 8 mg/d) (Table 1). When evaluated by phase order (i.e., comparing protocol A vs. B over time), MPA did not prevent the increased urinary OHP excretion seen with GnRH-a alone by week 12 (protocol A) (Fig. 3), but by week 24 urinary OHP decreased to baseline levels when MPA was added back (protocol B). Intestinal 47 Ca absorption fraction measurement was not affected by treatment. Estimated Ca balance decreased significantly during GnRH-a alone treatment compared to baseline ( mg/d vs mg/d, respectively; P.0001). During GnRH-a plus MPA, estimated Ca balance ( mg/d) declined significantly from baseline, but was attenuated compared to GnRH-a alone treatment (P.002) (Table 1, Fig. 4). When evaluated by phase order, estimated Ca balance declined similarly, by 12 weeks in both protocol A and B (Fig. 5). By week 24, when MPA was added to GnRH-a (protocol B), a significant increase in estimated Ca balance was observed compared to week 12 (P.001) and further decline was prevented. Measurements of Bone Density Changes in cortical bone as measured by single photon absorption did not change significantly in either treatment group or by protocol A or B over time (data not shown). DISCUSSION A number of studies have reported the potential usefulness of the hypoestrogenic state induced by GnRH-a for treatment of uterine leiomyomata and endometriosis (1 8). Although the pathogeneses of uterine leiomyomata and endometriosis have not been clearly established, it is generally agreed that uterine leiomyomata and endometriosis are hormonally dependent, primarily through the trophic effects of E. Progestin treatment alone provides relief of endometriosis-associated pelvic pain (1, 4); moreover, the addition of high-dose MPA (100 mg/d) supplementation to GnRH-a diminishes the hypoestrogenic side effects of GnRH-a with Carr et al. GnRH-a and MPA on Ca metabolism Vol. 80, No. 5, November 2003

6 FIGURE 3 The effect of GnRH agonist in the presence or absence of medroxyprogesterone acetate (MPA) on 24-hour urinary hydroxyproline excretion levels at 12 and 24 weeks. Protocol A: an initial 12 weeks of MPA (20 mg/d) followed by an additional 12 weeks of placebo; protocol B: an initial 12 weeks of placebo followed by 12 weeks of MPA (20 mg/d). All women received leuprolide acetate (1 mg/d) SC injections for 24 weeks. *P.05 compared to baseline; (*)P.05 compared to 12 weeks. Points and error bars represent mean and standard deviation. out changing its efficacy on extent of endometriosis and pain (21). The MPA appears to be a suboptimal progestin for add-back regimens for GnRH-a therapy directed toward reduction in size of uterine leiomyomata (8). However, a multicenter case-control study demonstrated a protective effect [reduced risk of developing uterine leiomyoma (odds ratio 0.44; 95% confidence interval )] of depot- MPA on surgically treated uterine leiomyomata (22). A multicenter prospective, randomized study has demonstrated reduced incidence of myoma and myoma-related surgery associated with the use of levonorgestrel intrauterine devices (IUD) (23). Furthermore, transdermal hormone replacement therapy (HT [transdermal 17 -E 2 ] patches plus oral MPA) did not increase the size of uterine leiomyomata or affect uterine bleeding patterns in postmenopausal women (24). The purpose of the present investigation was to evaluate the effect of addition of MPA to GnRH-a treatment in women with uterine leiomyomata or endometriosis by conducting a prospective, randomized, double-blind, placebocontrolled, crossover trial to investigate the effects on gonadotropins, Ca homeostasis, and bone density during a 6-month treatment period. We have shown that LH and E 2 levels were suppressed with GnRH-a or GnRH-a plus MPA to demonstrate the effectiveness and compliance of GnRH-a on hormonal milieu. We observed only a minor effect of the two treatments on FSH levels. Although most investigators have demonstrated a reduction in FSH levels on GnRH-a, we and other researchers have not (8, 25). With regard to Ca homeostasis and bone density measurements, a number of observations were made. When patients received GnRH-a alone, the diminution in serum E levels resulted in greater bone turnover. This was reflected in FIGURE 4 The effect of GnRH agonist (GnRH-a) plus or minus medroxyprogesterone acetate (MPA) on estimated calcium balance compared to baseline. P.0001 compared to baseline; ( )P.002 compared to GnRH-a alone. Error bars represent standard deviation. FERTILITY & STERILITY 1221

7 FIGURE 5 The effect of GnRH agonist in the presence or absence of medroxyprogesterone acetate (MPA) on estimated calcium balance at 12 and 24 weeks. Protocol A: an initial 12 weeks of MPA (20 mg/d) followed by an additional 12 weeks of placebo; protocol B: an initial 12 weeks of placebo followed by 12 weeks of MPA (20 mg/d). All women received leuprolide acetate (1 mg/d) SC injections for 24 weeks. **P.01 compared to baseline; P.001 compared to baseline; ( )P.001 compared to 12 weeks. Points and error bars represent mean and standard deviation. the increase in serum Ca, phosphorus, alkaline phosphatase, and osteocalcin. As expected, fasting and 24-hour urine Ca excretion also increased significantly. With increased Ca exiting the bone, serum PTH detected by immunoradiometric assay was surprisingly not suppressed. There was a tendency for a reduction in serum calcitriol and intestinal Ca absorption. Similar biochemical changes suggesting increased bone turnover, namely, increases in serum alkaline phosphatase and urinary hydroxyproline, have been reported by other investigators in women receiving GnRH-a (26 28). As a result of the increased urinary Ca losses, in association with a tendency for reduced intestinal Ca absorption, estimated Ca balance (A Ca U Ca ) declined by 98 mg/d (P.0001). During treatment with GnRH-a plus MPA, there was no increase in markers of bone turnover, specifically alkaline phosphatase and osteocalcin. There was an increase in U Ca excretion and a decrease in estimated Ca balance with GnRH-a plus MPA, but these changes were significantly attenuated compared to GnRH-a alone. Thus, based on metabolic parameters, it seemed as if the addition back of MPA partially prevented the negative Ca balance that otherwise would have ensued from E deficiency as observed by other researchers (21, 29). Similar protective effects during GnRH-a administration have been reported for other progestins, namely norethindrone and norethindrone acetate (28, 30). The use of add-back therapy with medrogestone has an advantageous effect in reducing, although not abolishing, bone mineral loss associated with goserelin monotherapy at lumbar spine, femoral neck, and Ward s triangle (6). The addition of norethindrone acetate, 5 mg or 10 mg daily, to the use of leuprolide acetate up to 1 year for patients with endometriosis or to the use of leuprolide acetate up to 2 years for women with large, symptomatic leiomyomata, respectively, protects against bone loss (4). In contrast with a negative Ca balance, women treated with GnRH-a alone or GnRH-a plus MPA did not have a change in bone density. Although, there might be a trend of increased bone density in the combination treatment group, this was not statistically significant. The study design with a 3-month treatment arm and crossover appears to have been insufficient length of time to observe differences. The lack of decline in bone density at the distal third of the radius with GnRH-a may be secondary to less metabolically active cortical bone (31). A significant change in bone markers can be 1222 Carr et al. GnRH-a and MPA on Ca metabolism Vol. 80, No. 5, November 2003

8 observed within months of antiresorptive therapy, before there are changes in bone mineral density (31). In evaluating the phase order of MPA treatment, initial concomitant treatment with MPA did not prevent the increased urinary excretion of calcium and hydroxyproline seen with GnRH-a alone and the decreased estimated calcium balance. When MPA was initiated during weeks 12 24, urinary excretion of calcium and hydroxyproline decreased, and these were significant compared to week 12. By week 24, when MPA was added to GnRH-a (protocol B), a significant increase in estimated calcium balance was observed compared to week 12 (protocol A) and decline seen with GnRH-a alone was prevented. We are unable to explain the phase order difference in this study. Progestin appears to modulate bone remodeling, resulting in partial protection against bone loss caused by E deficiency (32). This effect may be mediated by P receptor expression in human osteoblast-like cells (33), presence of P response element in the 5 -flanking sequence of the mouse osteonectin (bone matrix protein) gene (34), and ability of P to antagonize glucocorticoid-induced bone loss (28). References 1. Surrey ES. Add-back therapy and gonadotropin-releasing hormone agonists in the treatment of patients with endometriosis: can a consensus be reached? Add-Back Consensus Working Group. Fertil Steril 1999;71: Schlaff WD, Zerhouni EA, Huth JA, Chen J, Damewood MD, Rock JA. A placebo-controlled trial of a depot gonadotropin-releasing hormone analogue (leuprolide) in the treatment of uterine leiomyomata. Obstet Gynecol 1989;74: Friedman AJ, Daly M, Juneau-Norcross M, Rein MS, Fine C, Gleason R, et al. A prospective, randomized trial of gonadotropin-releasing hormone agonist plus estrogen-progestin or progestin add-back regimens for women with leiomyomata uteri. J Clin Endocrinol Metab 1993;76: Hornstein MD, Surrey ES, Weisberg GW, Casino LA. Leuprolide acetate depot and hormonal add-back in endometriosis: a 12-month study. Lupron Add-Back Study Group. Obstet Gynecol 1998;91: Sillem M, Parviz M, Woitge HW, Kiesel L, Ulrich U, von Holst T, et al. Add-back medrogestone does not prevent bone loss in premenopausal women treated with goserelin. Exp Clin Endocrinol Diabetes 1999;107: Kiesel L, Schweppe KW, Sillem M, Siebzehnrubl E. Should add-back therapy for endometriosis be deferred for optimal results? Br J Obstet Gynaecol 1996;103(Suppl 14): Caird LE, West CP, Lumsden MA, Hannan WJ, Gow SM. Medroxyprogesterone acetate with Zoladex for long-term treatment of fibroids: effects on bone density and patient acceptability. Hum Reprod 1997; 12: Carr BR, Marshburn PB, Weatherall PT, Bradshaw KD, Breslau NA, Byrd W, et al. An evaluation of the effect of gonadotropin-releasing hormone analogs and medroxyprogesterone acetate on uterine leiomyomata volume by magnetic resonance imaging: a prospective, randomized, double blind, placebo-controlled, crossover trial. J Clin Endocrinol Metab 1993;76: Scholes D, Lacroix AZ, Ott SM, Ichikawa LE, Barlow WE. Bone mineral density in women using depot medroxyprogesterone acetate for contraception. Obstet Gynecol 1999;93: Berenson AB, Radecki CM, Grady JJ, Rickert VI, Thomas A. A prospective, controlled study of the effects of hormonal contraception on bone mineral density. Obstet Gynecol 2001;98: Prior JC, Vigna YM, Barr SI, Rexworthy C, Lentle BC. Cyclic medroxyprogesterone treatment increases bone density: a controlled trial in active women with menstrual cycle disturbances. Am J Med 1994; 96: Perrotti M, Bahamondes L, Petta C, Castro S. Forearm bone density in long-term users of oral combined contraceptives and depot medroxyprogesterone acetate. Fertil Steril 2001;76: Gallagher JC, Kable WT, Goldgar D. Effect of progestin therapy on cortical and trabecular bone: comparison with estrogen. Am J Med 1991;90: Adachi JD, Sargeant EJ, Sagle MA, Lamont D, Fawcett PD, Bensen WG, et al. A double-blind randomised controlled trial of the effects of medroxyprogesterone acetate on bone density of women taking oestrogen replacement therapy. Br J Obstet Gynaecol 1997;104: Anonymous. Effects of estrogen or estrogen/progestin regimens on heart disease risk factors in postmenopausal women. The Postmenopausal Estrogen/Progestin Interventions (PEPI) Trial. The Writing Group for the PEPI Trial. JAMA 1995;273: Breslau NA, Preminger GM, Adams BV, Otey J, Pak CY. Use of ketoconazole to probe the pathogenetic importance of 1,25-dihydroxyvitamin D in absorptive hypercalciuria. J Clin Endocrinol Metab 1992;75: Breslau NA, McGuire JL, Zerwekh JE, Frenkel EP, Pak CY. Hypercalcemia associated with increased serum calcitriol levels in three patients with lymphoma. Ann Intern Med 1984;100: Fiske CH, Subbanow Y. The colorimetric determination of phosphorus. J Biol Chem 1925;66: Kivirikko KI, Laitinen O, Prockop DJ. Modifications of a specific assay for hydroxyproline in urine. Anal Biochem 1967;19: Preminger GM, Sakhaee K, Pak CY. Hypercalciuria and altered intestinal calcium absorption occurring independently of vitamin D in incomplete distal renal tubular acidosis. Metabolism 1987;36: Makarainen L, Ronnberg L, Kauppila A. Medroxyprogesterone acetate supplementation diminishes the hypoestrogenic side effects of gonadotropin-releasing hormone agonist without changing its efficacy in endometriosis. Fertil Steril 1996;65: Lumbiganon P, Rugpao S, Phandhu-fung S, Laopaiboon M, Vudhikamraksa N, Werawatakul Y. Protective effect of depot-medroxyprogesterone acetate on surgically treated uterine leiomyomas: a multicentre case-control study. Br J Obstet Gynaecol 1996;103: Sivin I, Stern J. Health during prolonged use of levonorgestrel 20 micrograms/d and the copper TCu 380Ag intrauterine contraceptive devices: a multicenter study. International Committee for Contraception Research (ICCR). Fertil Steril 1994;61: Palomba S, Sena T, Noia R, Di Carlo C, Zullo F, Mastrantonio P. Transdermal hormone replacement therapy in postmenopausal women with uterine leiomyomas. Obstet Gynecol 2001;98: Rittmaster RS. Differential suppression of testosterone and estradiol in hirsute women with the superactive gonadotropin-releasing hormone agonist leuprolide. J Clin Endocrinol Metab 1988;67: Johansen JS, Riis BJ, Hassager C, Moen M, Jacobson J, Christiansen C. The effect of a gonadotropin-releasing hormone agonist analog (nafarelin) on bone metabolism. J Clin Endocrinol Metab 1988;67: Riis BJ, Christiansen C, Johansen JS, Jacobson J. Is it possible to prevent bone loss in young women treated with luteinizing hormonereleasing hormone agonists? J Clin Endocrinol Metab 1990;70: Prior JC. Progesterone as a bone-trophic hormone. Endocr Rev 1990; 11: Surrey ES, Fournet N, Voigt B, Judd HL. Effects of sodium etidronate in combination with low-dose norethindrone in patients administered a long-acting GnRH agonist: a preliminary report. Obstet Gynecol 1993; 81: Dambacher MA, Neff M, Kissling R, Qin L. Highly precise peripheral quantitative computed tomography for the evaluation of bone density, loss of bone density and structures. Consequences for prophylaxis and treatment. Drugs Aging 1998;12(Suppl 1): Rosen CJ, Chesnut CH 3rd, Mallinak NJ. The predictive value of biochemical markers of bone turnover for bone mineral density in early postmenopausal women treated with hormone replacement or calcium supplementation. J Clin Endocrinol Metab 1997;82: Graham JD, Clarke CL. Physiological action of progesterone in target tissues. Endocr Rev 1997;18: Wei LL, Leach MW, Miner RS, Demers LM. Evidence for progesterone receptors in human osteoblast-like cells. Biochem Biophys Res Commun 1993;195: Nomura S, Hashmi S, McVey JH, Ham J, Parker M, Hogan BL. Evidence for positive and negative regulatory elements in the 5 flanking sequence of the mouse sparc (osteonectin) gene. J Biol Chem 1989;264: FERTILITY & STERILITY 1223