- Email: [email protected]
Insulin sensitivity is unaltered by the use of the Norplant® subdermal implant contraceptive
Insulin S,ensitivity is Unaltered by the Use of the Norplant@ Subdermal Implant Contraceptive Tina B. Koopersmith
and Rogerio A. Lobo
Prior evaluations of carbohydrate metabolism in Norplant implant users, using the oral glucose tolerance test, have shown mild but clinically insignificant deterioration of insulin sensitivity. Using the more sensitive insulin tolerance test, the effects of the Norplant implant system on insulin sensitivity was studied in normal women, Insulin tolerance tests were performed before Norplant implant insertion in ten ovulatory female volunteers and repeated after 12 weeks of use. Both fasting glucose and fasting insulin values were similar before and after the use of Norplant implants. There was no significant difference in either the Kitt glucose value or the Kitt insulin value at baseline and at 3 months of use. Furthermore, the insulinto-glucose ratio did not differ before and after the use of Norplant implants. There was a significant correlation with BMl and Kitt glucose value (r = -0.45, p < .05), as well as between fasting insulin and BMI (r = 0.6, p = 0.006). In the first three months of use, the levonorgestrelcontaining implant system, Norplant, does not affect sensitivity to insulin or glucose in normal, nondiabetic women. CONTRACEPTION 1995; 5 1: 197-200 KEY WORDS: carbohydrate, insulin resistance, levonorgestrel, Norplant, insulin tolerance test
Introduction It is well known that steroid hormones can influence carbohydrate metabolism. Abnormalities in carbohydrate metabolism are important clinically and may lead to vascular disease. Progestins have been shown to exhibit a variety of effects on insulin actions, facilitating insulin action in some tissues, while at other organ sites, promoting insulin resistance.’ The balance of effects determines if a state of insulin senDivision of Reproductive Endocrinology and lnfenility, Department of Obstetrics and Gynecology, University of Southern California School of Medicine, Los Angeles, California Address for correspondence and reprint requests: Tina 6 Koopersmith, M.D., Department of Obstetrics and Gynecology, Women’s and Children’s Hospital, Room 1 M2, Los Angeles County-University of Southern California Medical Center, 1240 North Missio Road, Los Angeles, CA 90033 @‘Norplant is a registered tra emark of the Population Council, Inc. Submitted for publication S ! ptember 12, 1994 Revised December 8, 1994 Accepted for publication D&ember 20, 1994
0 1995 Elsevier Science Inc. 655 Avenue of the Amerrcas, New York,
NY 10010
sitivity or resistance develops. In humans, it has been shown that insulin sensitivity decreases in the luteal phase of the normal menstrual cycle, compared with the follicular phase, suggesting that in a state of high progesterone (P), insulin resistance may be more pronounced.2 Ever since the introduction of oral contraceptive pills, there have been numerous studies examining the effects of different estrogens and progestins on glucose metabolism. Most studies have demonstrated that progestins adversely affect carbohydrate metabolism.3 However, the magnitude of changes in insulin and glucose appear to be dose-related.4 On balance, the majority of studies have documented a variable degree of deterioration of glucose metabolism in association with increased plasma insulin concentrations. However, most of these observed changes have not been in the diabetic range, thus questioning the clinical relevance of these findings.5 The active steroid in Norplant, the subdermal implantable contraceptive, is a progestin, levonorgestrel (LNG). Hence, if progestins are detrimental to carbohydrate metabolism, it is likely that Norplant implants may adversely affect carbohydrate metabolism. The few studies performed to date, all utilizing the oral glucose tolerance test, indicate that the Norplant system does not markedly alter carbohydrate metabolism.C1i However, in one study,6 Konje reported an increase in the area under the curve for both glucose and insulin, suggesting some element of insulin resistance. It is known that the oral glucose tolerance test is an imprecise measure of in vivo insulin action. More accurate measurements include the euglycemic clamp studies, the IV glucose tolerance test (IVGTT), and the insulin tolerance test (ITT).12-i4 The ITT has been shown to have an excellent correlation with euglycemic clamp studies, as well as the intravenous (-J7y*W6
Therefore, we chose to use the ITT to examine if a continuous release of LNG by subdermal implants in the Norplant system would affect carbohydrate metabolism. lSSN/ OOlO-7824/95/$9.50 SSDV 0010.7824(95)00016-4
198
Koopersmith and Lobo
Materials and Methods Subjects Ten ovulatory women, 18 to 34 years old, volunteered for this study. All subjects were recruited from the family planning clinic after requesting the use of the Nor-plant implantable contraceptive method. All subjects were healthy, nonsmokers and had no history of diabetes or gestational diabetes. None of the subjects was using medications on a routine basis nor had any subject used hormonal medications within eight weeks of the study. All subjects gave written informed consent for the study, which was approved by the Institutional Review Board of the LAC/USC Medical Center. Protocol The subjects were instructed to maintain a high carbohydrate diet for three days before performance of each insulin tolerance test. After an overnight fast, a 19-gauge butterfly needle was placed in an antecubital vein for injection of insulin and for blood sampling. All tests were carried out between 0800 and 1000 A.M. After 3 basal blood samples, 0.1 U/kg of regular insulin (Novolin R; Novo Nordisk A/S, Bagsvaerd, Denmark) was administered as an intravenous bolus. Blood samples were obtained 3, 6, 9, 12, 15,20, and 30 minutes after insulin injection. Thirty minutes after the insulin bolus, a solution of 50 ml of 50% glucose in water was infused to decrease the fall in glucose levels. Blood samples were placed in Vacutainer blood collection tubes (Becton Dickinson Vacutainer Systems, Rutherford, NJ) containing 10 mg sodium heparin. All samples were kept on ice until centrifuged at 4’C, and the plasma aliquot were kept frozen at - 2O’C until analyzed for glucose and insulin. An additional basal blood sample was placed in an empty Vacutainer blood collection tube for determination of serum steroid concentrations. This sample also was centrifuged and then stored at -2O’C until analysis. Baseline insulin tolerance tests were performed during the luteal phase in 7 subjects (days 14 to 22) and in the follicular phase in 3 subjects (days 4, 7, and 11) of the menstrual cycle. During the first menses after the initial testing, the Norplant implants (Wyeth Ayerst Pharmaceuticals) were inserted in a standard fashion in the left upper arm.” Twelve weeks after insertion, the insulin tolerance tests were repeated. Assays Plasma glucose was measured by the glucose oxidase method.i8 Plasma insulin was measured using an lz51-
Contraception 1995:51:197-200
insulin radioimmunoassay (Pharmacia, Sweden). The assay sensitivity was 3 U/mL. Intra-assay coefficients of variation were 3.5% and 7% for glucose and insulin, respectively, and interassay coefficients of variation were 4.5% and 13.5%, respectively. Serum LNG concentration was determined using an lz51-LNG radioimmunoassay after extraction of serum with hexane. l9 The assay sensitivity was 50 pg/ml. Intra-assay and inter-assay coefficients of variation were 4.4% and 4.9%, respectively. 15,16-3H-LNG was used to monitor procedural losses. All samples from each individual were measured in duplicate in the same assay. The rate constant for plasma glucose disappearance (Kitt glucose) was calculated from the formula 0.693/ tl,Z x 100. The plasma glucose ti,T was calculated from the slope by the least squares analysis of plasma glucose concentrations for 3 to 15 minutes after the intravenous injection as previously described. l5 The fractional disappearance rate for plasma insulin (Kitt insulin) after intravenous injection of insulin was calculated in the same way. Body mass index was calculated as weight per height (kg/m’). Statistical Analysis All data are presented as mean 2 SE. Statistical comparisons were calculated by the paired Student’s t-test. Regression analyses were carried out by the method of least squares.
Results There was no correlation between Kitt glucose and weight; although there was a significant negative correlation between body mass index (BMI) and Kitt glucose (r = -0.447, p = 0.049). There was also a significant correlation between fasting insulin and BMI (r = 0.596, p = 0.006). Both mean fasting glucose and fasting insulin values were similar before and after Norplant implant use (84.6 2 2.3 [S.E.] vs 84.7 ? 1.8 mg/dl and 7.3 2 2.2 vs 9.8 ? 3.3 uU/ml, respectively). Furthermore, no significant difference in baseline fasting glucose or fasting insulin was detected among those subjects in the follicular phase compared with those in the luteal phase. There was also no significant difference in insulin-to-glucose ratios before and after Norplant implants (11.7 * 3.8 [S.E.] vs 8.2 2 3.0 UU x102/mg). Baseline fasting insulin levels did not correlate with Kitt glucose values. There was no significant difference in the Kitt glucose value at baseline and after the use of the Norplant implants (3.59 & 0.37 [S.E.] vs 3.49 ? 0.51). This was true whether the initial test was performed in
Insulin Sensitivity and Norplant 1%)
Contraception 199551: 197-200
either the luteal or the follicular phase of the cycle. Similarly, the Kitt insulin value did not differ significantly before and after Norplant implant use (757 ? 0.30 [S.E.] vs 7.95 ? 0.23, respectively). Among the ten subjects, the mean LNG level was 438.5 pg/ml 2 164.15, with a range of 131 to 1900 pg/ml. There was a significant negative correlation between serum LNG levels and Kitt glucose (Figure 1: r = -0.697, p = 0.025). There was also a significant negative correlation between serum LNG concentrations and Kitt insulin (r = -0.643, p = 0.043).
Discussion These data support the conclusion of prior studies , &lo using the oral glucose tolerance test, in which it is shown that the levonorgestrel-containing implant system does not adversely affect insulin sensitivity in normal, nondiabetic women. The results are also in agreement with the one study that utilized the IV GTT.20 One explanation for the lack of a detrimental effect of levonorgestrel on carbohydrate metabolism, which contradicts the findings obtained with some oral contraceptive agents, is that it may not be the progestin in the contraceptive pill that adversely affects insulin sensitivity, but rather the estrogen component. Although an improvement in insulin sensitivity has been observed in post-menopausal women after ingestion of low doses of conjugated estrogens, the ingestion of higher doses of conjugated estrogens appears to impair Kitt values.21 We have also shown that Kitt glucose decreased with the ingestion of low doses of ethinyl estradiol (20, 35, and 50 mcg), but not with 1 Kitt Insulin ia
r= - 0.643 9
.
.
a
6
5 ! 0
400
SD0
1,200
1,800
2,000
LNG (pg/mL) Figure 1. Correlation Glucose values.
‘between
serum LNG
levels
and Kitt
mg of norethindrone alone.22 Similar results by Godsland et al.23 showed that the effects of the oral contraceptive pills on carbohydrate metabolism appeared to be due to a combination of estrogen-induced insulin resistance and progestin-associated changes in insulin half-life. Alternatively, the lack of a detrimental response on carbohydrate metabolism may be a dose-response phenomenon. Since the daily release rate of levonorgestrel in the implants is only 30-50 pg/day, about l/2 to l/3 of the daily dose of levonorgestrel in the combination contraceptive pills, the serum levels of levonorgestrel in implant users are lower than those of women ingesting levonorgestrel-containing pills.” This supposition is supported by Godsland’s study23 in which oral contraceptive formulations with ethinyl estradiol and progestins increased insulin resistance whereas the formulations with progestin alone, with a lower progestin dose, equivalent to the level from the implants, did not affect insulin resistance. A dose-response phenomenon may also explain why the Kitt insulin levels were not altered with the use of Norplant implants, despite prior studies suggesting that progestins alter metabolic clearance of insulin.z3 No change in Kitt insulin levels was observed with a 1 mg norethindrone formulation containing no estrofound between segen. ” Moreover, the correlation rum LNG levels and Kitt glucose and Kitt insulin levels is supportive of a dose-response relation between progestin levels and glucose impairment. It appears that in users of Norplant implants who have serum LNG concentrations in the 200-400 pg/ml range, which is found in the majority of women, the Kitt glucose and Kitt insulin levels remain normal. However, those women who have an abnormally high serum level of LNG, the Kitt glucose and Kitt insulin levels may be adversely affected (Figure 1). It is important to recognize that this study utilized the ITT, which has been shown to be a more accurate measure of in vivo carbohydrate metabolism than the oral glucose test. lzPi5 Data using the euglycemic clamp have suggested that factors other than insulin resistance influence plasma insulin responses after oral glucose administration. No consistent relationship has been observed between results using the OGTT and those studies using the euglycemic clamp method of assessing glucose metabolism.z4 The minimal model IVGTT has been shown to correlate well with both the euglycemic clamp studies12 and the 17y-~y.lkl6 The results of this study also corroborate prior studies which have shown that as the body mass index increases, there is an increase in insulin resistance as manifested by a decrease in Kitt glucose values as well as by an elevation in fasting insulin.
200
Koopersmith
and Lobo
Conclusions This study, which utilized the insulin tolerance test as a measure of insulin sensitivity, indicates that the use of the contraceptive levonorgestrel implants does not adversely affect carbohydrate metabolism in normal nondiabetic females in the first three months of use. This corroborates prior studies with the Norplant system which utilized the oral glucose tolerance test. Further research is needed to determine if carbohydrate metabolism in women with impaired glucose tolerance or overt diabetes mellitus will be altered by the Norplant system.
References 1. Kalkoff R. Metabolic effects of progesterone. Am J Ob Gyn 1982; 142: 735-8. 2. Valdes C, Elkind-Hirsch K. Intravenous glucose tolerance test-derived insulin sensitivity changes during the menstrual cycle. J Clin Endocrinol Metab 1991; 72: 642-5. 3. Krauss R, Burkman RT. The metabolic impact of oral contraceptives. Am J Ob Gyn 1992j 167: 1177-84. 4. Ramcharan S, ed. The Walnut Creek Contraceptive Drug Study. Bethesda: National Institutes of Health. 1974/1976/1981: DHEW publication no. 5 (NIH), 74522, 76-563, 81-564. 5. Kjos S, Shoupe D, Douyan S et al. Effect of low-dose oral contraceptives on carbohydrate and lipid metabolism in women with recent gestational diabetes: Results of a controlled randomized prospective study. Am J Ob Gyn 199Oj 163: 1822-7. 6. Konje J, Odukuya 0, Otolorin E, Ewings P, Ladipo 0. Carbohydrate metabolism before and after Norplant removal, Contraception 1992j 46: 61-9. 7. Konje JC, Otolorin EO, Ladipo OA. The effect of continuous subdermal levonorgestrel (Norplant) on carbohydrate metabolism . Am J Ob Gyn 1992j 166: 15-9. 8. Konje JC, Otolorin EO, Ladipo OA. Changes in carbohydrate metabolism during 30 months on Norplant. Contraception 199lj 44: 163-72. 9. Singh K, Viegas 0, Ratnam S. Effect of Norplant implants on liver, lipid and carbohydrate metabolism. Contraception 1992j 45: 141-53. 10. Singh K, Viegas 0, Liew D, Singh I’, Ratnam S. Twoyear follow-up of changes in clinical chemistry in Singaporean Norplant acceptors: Metabolic changes. Contraception 1989j 39: 129-36. 11. Singh K, Viegas 0, Loke D, Ratnam R. Effect of Norplant-2 Rods on liver, lipid and carbohydrate metabolism. Contraception, 1992j 45: 463-72.
Contraception 1995:51: 197-200
12. Bergman R, Prager R, Volund A, Olefsky J. Equivalence of the insulin sensitivity index in man derived by the minimal model method and the euglycemic glucose clamp. J Chn Invest 1987j 79: 790-800. 13. Bergman RN, Phillips LS, Cobelli C. Physiologic evaluation of factors controlling glucose tolerance in man. Measurement of insulin sensitivity and B-cell glucose sensitivity from the response to intravenous glucose. J Clin Invest 198lj 68: 1456-67. 14. Beard JC, Bergman RN, Ward K, Porte D, Jr. The insulin sensitivity index in man: Correlation between clampderived and IVGTT-derived values. Diabetes 1986j 35: 362-9. 15. Bonora E, Moghetti P, Zacanaro C et al. Estimates of in vivo insulin action in man: Comparison of insulin tolerance tests with euglycemic and hyperglycemic glucose clamp studies. J Clin Endocrinol Metab 1989j 68: 374-8. 16. Lindheim SR, Buchanan TA, Duffy DM et al. Comparison of estimates of insulin sensitivity in pre- and postmenopausal women using the insulin tolerance test and the frequently sampled intravenous glucose tolerance test. J SOCGynecol Invest. 1994j 1: 15&4. 17. Shoupe D, Mishell DR, Jr. Norplant: Subdermal implant system for long-term contraception. Am J Ob Gyn 1989j 100: 128~91. 18. Trinder G. Determination in blood using glucose oxidase with an alternative oxygen receptor. An Clin Biothem 1969j 6: 24-8. 19. Stanczyk FZ, Hiroi M, Goebelsmann U, Brenner PF, Lumkin M, Mishell DR, Jr. Radioimmunoassay of serum d-norgestrel in women following oral and intravaginal administration. Contraception 1975j 12: 27998. 20. Harper MA, Meis PJ. Continuous subdermal levonorgestrel does not affect insulin sensitivity. Presented at the Society for Gynecologic Investigation: Chicago, Illinois, March, 1994j 240. 21. Lindheim SR, Presser SC, Ditkoff EC, Vijod BS, Stanczyk FZ, Lobo RA. A possible bimodal effect of estrogen on insulin sensitivity in postmenopausal women and the attenuating effect of added progestin. Fertil Steril 1993j 60: 66~7. 22. Kojima T, Lindheim SR, Duffy DM, Vijod MA, Stanczyk FZ, Lobo RA. Insulin sensitivity is decreased in normal women by doses of ethinyl estradiol used in oral contraceptives. Am J Ob Gyn 1993; 169: 1540-4. 23. Godsland IF, Walton C, Felton A, Proudler A, Pate1 A, Wynn V. Insulin resistance, secretion, and metabolism in users of oral contraceptives. J Clin Endocrinol Metab 1992j 74: 6~70. 24. Kasdorf G, Kalkoff R. Prospective studies of insulin sensitivity in normal women receiving oral contraceptive agents. J Clin Endocrinol Metab. 1988; 66: 846-52.
and Rogerio A. Lobo
Prior evaluations of carbohydrate metabolism in Norplant implant users, using the oral glucose tolerance test, have shown mild but clinically insignificant deterioration of insulin sensitivity. Using the more sensitive insulin tolerance test, the effects of the Norplant implant system on insulin sensitivity was studied in normal women, Insulin tolerance tests were performed before Norplant implant insertion in ten ovulatory female volunteers and repeated after 12 weeks of use. Both fasting glucose and fasting insulin values were similar before and after the use of Norplant implants. There was no significant difference in either the Kitt glucose value or the Kitt insulin value at baseline and at 3 months of use. Furthermore, the insulinto-glucose ratio did not differ before and after the use of Norplant implants. There was a significant correlation with BMl and Kitt glucose value (r = -0.45, p < .05), as well as between fasting insulin and BMI (r = 0.6, p = 0.006). In the first three months of use, the levonorgestrelcontaining implant system, Norplant, does not affect sensitivity to insulin or glucose in normal, nondiabetic women. CONTRACEPTION 1995; 5 1: 197-200 KEY WORDS: carbohydrate, insulin resistance, levonorgestrel, Norplant, insulin tolerance test
Introduction It is well known that steroid hormones can influence carbohydrate metabolism. Abnormalities in carbohydrate metabolism are important clinically and may lead to vascular disease. Progestins have been shown to exhibit a variety of effects on insulin actions, facilitating insulin action in some tissues, while at other organ sites, promoting insulin resistance.’ The balance of effects determines if a state of insulin senDivision of Reproductive Endocrinology and lnfenility, Department of Obstetrics and Gynecology, University of Southern California School of Medicine, Los Angeles, California Address for correspondence and reprint requests: Tina 6 Koopersmith, M.D., Department of Obstetrics and Gynecology, Women’s and Children’s Hospital, Room 1 M2, Los Angeles County-University of Southern California Medical Center, 1240 North Missio Road, Los Angeles, CA 90033 @‘Norplant is a registered tra emark of the Population Council, Inc. Submitted for publication S ! ptember 12, 1994 Revised December 8, 1994 Accepted for publication D&ember 20, 1994
0 1995 Elsevier Science Inc. 655 Avenue of the Amerrcas, New York,
NY 10010
sitivity or resistance develops. In humans, it has been shown that insulin sensitivity decreases in the luteal phase of the normal menstrual cycle, compared with the follicular phase, suggesting that in a state of high progesterone (P), insulin resistance may be more pronounced.2 Ever since the introduction of oral contraceptive pills, there have been numerous studies examining the effects of different estrogens and progestins on glucose metabolism. Most studies have demonstrated that progestins adversely affect carbohydrate metabolism.3 However, the magnitude of changes in insulin and glucose appear to be dose-related.4 On balance, the majority of studies have documented a variable degree of deterioration of glucose metabolism in association with increased plasma insulin concentrations. However, most of these observed changes have not been in the diabetic range, thus questioning the clinical relevance of these findings.5 The active steroid in Norplant, the subdermal implantable contraceptive, is a progestin, levonorgestrel (LNG). Hence, if progestins are detrimental to carbohydrate metabolism, it is likely that Norplant implants may adversely affect carbohydrate metabolism. The few studies performed to date, all utilizing the oral glucose tolerance test, indicate that the Norplant system does not markedly alter carbohydrate metabolism.C1i However, in one study,6 Konje reported an increase in the area under the curve for both glucose and insulin, suggesting some element of insulin resistance. It is known that the oral glucose tolerance test is an imprecise measure of in vivo insulin action. More accurate measurements include the euglycemic clamp studies, the IV glucose tolerance test (IVGTT), and the insulin tolerance test (ITT).12-i4 The ITT has been shown to have an excellent correlation with euglycemic clamp studies, as well as the intravenous (-J7y*W6
Therefore, we chose to use the ITT to examine if a continuous release of LNG by subdermal implants in the Norplant system would affect carbohydrate metabolism. lSSN/ OOlO-7824/95/$9.50 SSDV 0010.7824(95)00016-4
198
Koopersmith and Lobo
Materials and Methods Subjects Ten ovulatory women, 18 to 34 years old, volunteered for this study. All subjects were recruited from the family planning clinic after requesting the use of the Nor-plant implantable contraceptive method. All subjects were healthy, nonsmokers and had no history of diabetes or gestational diabetes. None of the subjects was using medications on a routine basis nor had any subject used hormonal medications within eight weeks of the study. All subjects gave written informed consent for the study, which was approved by the Institutional Review Board of the LAC/USC Medical Center. Protocol The subjects were instructed to maintain a high carbohydrate diet for three days before performance of each insulin tolerance test. After an overnight fast, a 19-gauge butterfly needle was placed in an antecubital vein for injection of insulin and for blood sampling. All tests were carried out between 0800 and 1000 A.M. After 3 basal blood samples, 0.1 U/kg of regular insulin (Novolin R; Novo Nordisk A/S, Bagsvaerd, Denmark) was administered as an intravenous bolus. Blood samples were obtained 3, 6, 9, 12, 15,20, and 30 minutes after insulin injection. Thirty minutes after the insulin bolus, a solution of 50 ml of 50% glucose in water was infused to decrease the fall in glucose levels. Blood samples were placed in Vacutainer blood collection tubes (Becton Dickinson Vacutainer Systems, Rutherford, NJ) containing 10 mg sodium heparin. All samples were kept on ice until centrifuged at 4’C, and the plasma aliquot were kept frozen at - 2O’C until analyzed for glucose and insulin. An additional basal blood sample was placed in an empty Vacutainer blood collection tube for determination of serum steroid concentrations. This sample also was centrifuged and then stored at -2O’C until analysis. Baseline insulin tolerance tests were performed during the luteal phase in 7 subjects (days 14 to 22) and in the follicular phase in 3 subjects (days 4, 7, and 11) of the menstrual cycle. During the first menses after the initial testing, the Norplant implants (Wyeth Ayerst Pharmaceuticals) were inserted in a standard fashion in the left upper arm.” Twelve weeks after insertion, the insulin tolerance tests were repeated. Assays Plasma glucose was measured by the glucose oxidase method.i8 Plasma insulin was measured using an lz51-
Contraception 1995:51:197-200
insulin radioimmunoassay (Pharmacia, Sweden). The assay sensitivity was 3 U/mL. Intra-assay coefficients of variation were 3.5% and 7% for glucose and insulin, respectively, and interassay coefficients of variation were 4.5% and 13.5%, respectively. Serum LNG concentration was determined using an lz51-LNG radioimmunoassay after extraction of serum with hexane. l9 The assay sensitivity was 50 pg/ml. Intra-assay and inter-assay coefficients of variation were 4.4% and 4.9%, respectively. 15,16-3H-LNG was used to monitor procedural losses. All samples from each individual were measured in duplicate in the same assay. The rate constant for plasma glucose disappearance (Kitt glucose) was calculated from the formula 0.693/ tl,Z x 100. The plasma glucose ti,T was calculated from the slope by the least squares analysis of plasma glucose concentrations for 3 to 15 minutes after the intravenous injection as previously described. l5 The fractional disappearance rate for plasma insulin (Kitt insulin) after intravenous injection of insulin was calculated in the same way. Body mass index was calculated as weight per height (kg/m’). Statistical Analysis All data are presented as mean 2 SE. Statistical comparisons were calculated by the paired Student’s t-test. Regression analyses were carried out by the method of least squares.
Results There was no correlation between Kitt glucose and weight; although there was a significant negative correlation between body mass index (BMI) and Kitt glucose (r = -0.447, p = 0.049). There was also a significant correlation between fasting insulin and BMI (r = 0.596, p = 0.006). Both mean fasting glucose and fasting insulin values were similar before and after Norplant implant use (84.6 2 2.3 [S.E.] vs 84.7 ? 1.8 mg/dl and 7.3 2 2.2 vs 9.8 ? 3.3 uU/ml, respectively). Furthermore, no significant difference in baseline fasting glucose or fasting insulin was detected among those subjects in the follicular phase compared with those in the luteal phase. There was also no significant difference in insulin-to-glucose ratios before and after Norplant implants (11.7 * 3.8 [S.E.] vs 8.2 2 3.0 UU x102/mg). Baseline fasting insulin levels did not correlate with Kitt glucose values. There was no significant difference in the Kitt glucose value at baseline and after the use of the Norplant implants (3.59 & 0.37 [S.E.] vs 3.49 ? 0.51). This was true whether the initial test was performed in
Insulin Sensitivity and Norplant 1%)
Contraception 199551: 197-200
either the luteal or the follicular phase of the cycle. Similarly, the Kitt insulin value did not differ significantly before and after Norplant implant use (757 ? 0.30 [S.E.] vs 7.95 ? 0.23, respectively). Among the ten subjects, the mean LNG level was 438.5 pg/ml 2 164.15, with a range of 131 to 1900 pg/ml. There was a significant negative correlation between serum LNG levels and Kitt glucose (Figure 1: r = -0.697, p = 0.025). There was also a significant negative correlation between serum LNG concentrations and Kitt insulin (r = -0.643, p = 0.043).
Discussion These data support the conclusion of prior studies , &lo using the oral glucose tolerance test, in which it is shown that the levonorgestrel-containing implant system does not adversely affect insulin sensitivity in normal, nondiabetic women. The results are also in agreement with the one study that utilized the IV GTT.20 One explanation for the lack of a detrimental effect of levonorgestrel on carbohydrate metabolism, which contradicts the findings obtained with some oral contraceptive agents, is that it may not be the progestin in the contraceptive pill that adversely affects insulin sensitivity, but rather the estrogen component. Although an improvement in insulin sensitivity has been observed in post-menopausal women after ingestion of low doses of conjugated estrogens, the ingestion of higher doses of conjugated estrogens appears to impair Kitt values.21 We have also shown that Kitt glucose decreased with the ingestion of low doses of ethinyl estradiol (20, 35, and 50 mcg), but not with 1 Kitt Insulin ia
r= - 0.643 9
.
.
a
6
5 ! 0
400
SD0
1,200
1,800
2,000
LNG (pg/mL) Figure 1. Correlation Glucose values.
‘between
serum LNG
levels
and Kitt
mg of norethindrone alone.22 Similar results by Godsland et al.23 showed that the effects of the oral contraceptive pills on carbohydrate metabolism appeared to be due to a combination of estrogen-induced insulin resistance and progestin-associated changes in insulin half-life. Alternatively, the lack of a detrimental response on carbohydrate metabolism may be a dose-response phenomenon. Since the daily release rate of levonorgestrel in the implants is only 30-50 pg/day, about l/2 to l/3 of the daily dose of levonorgestrel in the combination contraceptive pills, the serum levels of levonorgestrel in implant users are lower than those of women ingesting levonorgestrel-containing pills.” This supposition is supported by Godsland’s study23 in which oral contraceptive formulations with ethinyl estradiol and progestins increased insulin resistance whereas the formulations with progestin alone, with a lower progestin dose, equivalent to the level from the implants, did not affect insulin resistance. A dose-response phenomenon may also explain why the Kitt insulin levels were not altered with the use of Norplant implants, despite prior studies suggesting that progestins alter metabolic clearance of insulin.z3 No change in Kitt insulin levels was observed with a 1 mg norethindrone formulation containing no estrofound between segen. ” Moreover, the correlation rum LNG levels and Kitt glucose and Kitt insulin levels is supportive of a dose-response relation between progestin levels and glucose impairment. It appears that in users of Norplant implants who have serum LNG concentrations in the 200-400 pg/ml range, which is found in the majority of women, the Kitt glucose and Kitt insulin levels remain normal. However, those women who have an abnormally high serum level of LNG, the Kitt glucose and Kitt insulin levels may be adversely affected (Figure 1). It is important to recognize that this study utilized the ITT, which has been shown to be a more accurate measure of in vivo carbohydrate metabolism than the oral glucose test. lzPi5 Data using the euglycemic clamp have suggested that factors other than insulin resistance influence plasma insulin responses after oral glucose administration. No consistent relationship has been observed between results using the OGTT and those studies using the euglycemic clamp method of assessing glucose metabolism.z4 The minimal model IVGTT has been shown to correlate well with both the euglycemic clamp studies12 and the 17y-~y.lkl6 The results of this study also corroborate prior studies which have shown that as the body mass index increases, there is an increase in insulin resistance as manifested by a decrease in Kitt glucose values as well as by an elevation in fasting insulin.
200
Koopersmith
and Lobo
Conclusions This study, which utilized the insulin tolerance test as a measure of insulin sensitivity, indicates that the use of the contraceptive levonorgestrel implants does not adversely affect carbohydrate metabolism in normal nondiabetic females in the first three months of use. This corroborates prior studies with the Norplant system which utilized the oral glucose tolerance test. Further research is needed to determine if carbohydrate metabolism in women with impaired glucose tolerance or overt diabetes mellitus will be altered by the Norplant system.
References 1. Kalkoff R. Metabolic effects of progesterone. Am J Ob Gyn 1982; 142: 735-8. 2. Valdes C, Elkind-Hirsch K. Intravenous glucose tolerance test-derived insulin sensitivity changes during the menstrual cycle. J Clin Endocrinol Metab 1991; 72: 642-5. 3. Krauss R, Burkman RT. The metabolic impact of oral contraceptives. Am J Ob Gyn 1992j 167: 1177-84. 4. Ramcharan S, ed. The Walnut Creek Contraceptive Drug Study. Bethesda: National Institutes of Health. 1974/1976/1981: DHEW publication no. 5 (NIH), 74522, 76-563, 81-564. 5. Kjos S, Shoupe D, Douyan S et al. Effect of low-dose oral contraceptives on carbohydrate and lipid metabolism in women with recent gestational diabetes: Results of a controlled randomized prospective study. Am J Ob Gyn 199Oj 163: 1822-7. 6. Konje J, Odukuya 0, Otolorin E, Ewings P, Ladipo 0. Carbohydrate metabolism before and after Norplant removal, Contraception 1992j 46: 61-9. 7. Konje JC, Otolorin EO, Ladipo OA. The effect of continuous subdermal levonorgestrel (Norplant) on carbohydrate metabolism . Am J Ob Gyn 1992j 166: 15-9. 8. Konje JC, Otolorin EO, Ladipo OA. Changes in carbohydrate metabolism during 30 months on Norplant. Contraception 199lj 44: 163-72. 9. Singh K, Viegas 0, Ratnam S. Effect of Norplant implants on liver, lipid and carbohydrate metabolism. Contraception 1992j 45: 141-53. 10. Singh K, Viegas 0, Liew D, Singh I’, Ratnam S. Twoyear follow-up of changes in clinical chemistry in Singaporean Norplant acceptors: Metabolic changes. Contraception 1989j 39: 129-36. 11. Singh K, Viegas 0, Loke D, Ratnam R. Effect of Norplant-2 Rods on liver, lipid and carbohydrate metabolism. Contraception, 1992j 45: 463-72.
Contraception 1995:51: 197-200
12. Bergman R, Prager R, Volund A, Olefsky J. Equivalence of the insulin sensitivity index in man derived by the minimal model method and the euglycemic glucose clamp. J Chn Invest 1987j 79: 790-800. 13. Bergman RN, Phillips LS, Cobelli C. Physiologic evaluation of factors controlling glucose tolerance in man. Measurement of insulin sensitivity and B-cell glucose sensitivity from the response to intravenous glucose. J Clin Invest 198lj 68: 1456-67. 14. Beard JC, Bergman RN, Ward K, Porte D, Jr. The insulin sensitivity index in man: Correlation between clampderived and IVGTT-derived values. Diabetes 1986j 35: 362-9. 15. Bonora E, Moghetti P, Zacanaro C et al. Estimates of in vivo insulin action in man: Comparison of insulin tolerance tests with euglycemic and hyperglycemic glucose clamp studies. J Clin Endocrinol Metab 1989j 68: 374-8. 16. Lindheim SR, Buchanan TA, Duffy DM et al. Comparison of estimates of insulin sensitivity in pre- and postmenopausal women using the insulin tolerance test and the frequently sampled intravenous glucose tolerance test. J SOCGynecol Invest. 1994j 1: 15&4. 17. Shoupe D, Mishell DR, Jr. Norplant: Subdermal implant system for long-term contraception. Am J Ob Gyn 1989j 100: 128~91. 18. Trinder G. Determination in blood using glucose oxidase with an alternative oxygen receptor. An Clin Biothem 1969j 6: 24-8. 19. Stanczyk FZ, Hiroi M, Goebelsmann U, Brenner PF, Lumkin M, Mishell DR, Jr. Radioimmunoassay of serum d-norgestrel in women following oral and intravaginal administration. Contraception 1975j 12: 27998. 20. Harper MA, Meis PJ. Continuous subdermal levonorgestrel does not affect insulin sensitivity. Presented at the Society for Gynecologic Investigation: Chicago, Illinois, March, 1994j 240. 21. Lindheim SR, Presser SC, Ditkoff EC, Vijod BS, Stanczyk FZ, Lobo RA. A possible bimodal effect of estrogen on insulin sensitivity in postmenopausal women and the attenuating effect of added progestin. Fertil Steril 1993j 60: 66~7. 22. Kojima T, Lindheim SR, Duffy DM, Vijod MA, Stanczyk FZ, Lobo RA. Insulin sensitivity is decreased in normal women by doses of ethinyl estradiol used in oral contraceptives. Am J Ob Gyn 1993; 169: 1540-4. 23. Godsland IF, Walton C, Felton A, Proudler A, Pate1 A, Wynn V. Insulin resistance, secretion, and metabolism in users of oral contraceptives. J Clin Endocrinol Metab 1992j 74: 6~70. 24. Kasdorf G, Kalkoff R. Prospective studies of insulin sensitivity in normal women receiving oral contraceptive agents. J Clin Endocrinol Metab. 1988; 66: 846-52.