- Email: [email protected]
Quantitative levonorgestrel plasma level measurements in patients with regular and prolonged use of the levonorgestrel-releasing intrauterine system
Contraception 86 (2012) 345 – 349
Original research article
Quantitative levonorgestrel plasma level measurements in patients with regular and prolonged use of the levonorgestrel-releasing intrauterine system☆ Beata Seeber a , Stephanie C. Ziehr a , Aandrea Gschlieβer a, b, 1 , Christina Moser c , Verena Mattle a , Christoph Seger c , Andrea Griesmacher c , Nicole Concin b , Hans Concin d , Ludwig Wildt a,⁎ a
Department of Gynecological Endocrinology and Reproductive Medicine, Innsbruck Medical University, A-6020 Innsbruck, Austria b Department of Obstetrics and Gynecology Innsbruck Medical University, Innsbruck, Austria c Division of Mass Spectrometry and Chromatography, Institute of Medical and Chemical Laboratory Diagnostics (ZIMCL), University Hospital Innsbruck, Innsbruck, Austria d Department of Obstetrics and Gynecology, Landeskrankenhaus Bregenz, Bregenz, Austria Received 3 August 2011; revised 21 January 2012; accepted 27 January 2012
Abstract Background: The levonorgestrel-releasing intrauterine system (LNG-IUS) is well accepted as an easy-to-use contraceptive with an excellent side-effect profile. It contains a reservoir of 52 mg of levonorgestrel (LNG) with continuous release of the steroid. Its contraceptive use is approved for 5 years. The aim of this study was to determine the plasma concentration of LNG and its variation with time in patients with indwelling LNG-IUS Mirena®. Study Design: In this study, we determined LNG plasma concentrations in 110 women with LNG-IUS at different time points of use. Time from insertion of the system in the study population ranged from 20 days to 11.1 years. Quantitative LNG levels were determined using a validated liquid chromatography–tandem mass spectrometry assay. Results: The mean±SD LNG plasma level in all women was 147±59 pg/mL. A highly significant negative correlation between LNG plasma level and LNG-IUS time of use could be demonstrated. In the first year of use, LNG plasma level was as high as 191±71 pg/mL, decreasing to 157±68 pg/mL in the second year and 134±41 pg/mL in the third year. Even after exceeding the recommended period of LNG-IUS use, systemic LNG concentrations were detectable: 133±38 pg/mL in the sixth year, 133±48 pg/mL in the seventh year and 117±45 pg/mL in the eighth year. Furthermore, a significant negative correlation between LNG plasma level and body mass index could be shown. Conclusion: Systemic LNG concentrations can be found in all patients with LNG-IUS IUS. However, concentrations are much lower than in other forms of LNG application. Moreover, this study demonstrates that a systemic effect of LNG-IUS can also be found after the recommended contraceptive lifespan of 5 years. © 2012 Elsevier Inc. All rights reserved. Keywords: Levonorgestrel; LNG-intrauterine system; Prolonged use; BMI; Plasma level
1. Introduction The levonorgestrel-releasing intrauterine system (LNGIUS) is well known as a highly effective, safe and long☆
Funding: University funding. No external financial support was received for this study. ⁎ Corresponding author. Tel.: +43 512 504 23276; fax: +43 512 504 23277. E-mail address: [email protected] (L. Wildt). 1 Present address: Department of Ophthalmology and Optometry, Vienna Medical University, Vienna, Austria. 0010-7824/$ – see front matter © 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.contraception.2012.01.015
acting contraceptive method with few side effects [1,2]. Besides its contraceptive use, the hormone-releasing IUS has some therapeutical benefits including a reduction in dysmenorrhea and menstrual bleeding and as endometrial protection in women using estrogen therapy [3-5]. The LNG-IUS consists of a T-shaped polyethylene frame body with a cylindrical reservoir containing 52 mg of LNG. Average release rate of LNG is 14 mcg LNG/24 h. Initially, 20 mcg LNG/24 h is released into the uterine cavity, declining progressively to 11 mcg LNG/24 h after 5 years of use [6,7].
346
B. Seeber et al. / Contraception 86 (2012) 345–349
Mirena® is approved for up to 5 years of use. However, there is evidence of effectiveness beyond this period. Even after 7 years of use, the cumulative pregnancy rate is as low as 1.1% [8]. Levonorgestrel is the biologically active form of norgestrel (D-norgestrel) and is a synthetic second-generation gestagen. As part of the LNG-IUS, the hormone exerts its effect primarily locally within the uterine cavity. It exerts an antiproliferative effect on the endometrium leading to endometrial decidualization and atrophy and a thickening of cervical mucus which prevents sperm penetration [9,10]. Side effects of the hormone-releasing IUS are generally rare and, in most cases, caused by the local effect of the IUS. Common causes for premature removal are excessive bleeding and spotting, intrauterine infections and pelvic pain [8,11]. Even though systemic side effects are generally infrequent, hormonal side effects like hirsutism, acne, headache and mood changes are also reasons for elective LNG-IUS removal in the first 36 months of use [12]. LNG-IUS users have more discontinuations because of hair and skin changes and headache than users of the copper intrauterine device (IUD) [13]. In the first 5 years of use, approximately 12% of women prematurely discontinue the LNG-IUS because of hormone-attributable complaints [14]. Hormonal side effects could possibly be explained by systemic LNG distribution. The aim of this study was to determine the extent of and time course of change in LNG plasma distribution in patients with in-dwelling LNG-IUS.
2. Materials and methods This two-center study was performed at the University Hospital in Innsbruck, Austria, and at the General Hospital in Bregenz, Austria. Women were eligible to participate if they currently had a levonorgestrel-containing IUS with known date of insertion. An informed consent form, previously approved by the Innsbruck Medical University Ethical Board (IRB), was obtained to store aliquots of blood samples taken for routine medical reasons and to use them for research purposes. Evaluated were 113 patients with LNG-IUS at different time points of use. Three women were excluded from analyses because of unknown date of insertion (n=1) and unknown date of blood sampling (n=2). In total, 138 LNG plasma level concentrations were measured during the study time period in this retrospective study. If multiple blood samples were obtained from the same patient (n=28), the most recent sample was used for further analysis. Consequently, 110 patients, each with a single LNG plasma concentration, were included in the data analysis. There were 75 women who had used the LNG-IUS within the approved time of use of ≤5 years and 35 women with prolonged use of ≥6 years. In this study population, time of IUS use ranged from 20 days to 11.1 years.
A validated assay was used for LNG plasma concentration measurement. A selective and sensitive liquid chromatography–tandem mass spectrometry method for the determination of LNG plasma levels was developed [15]. An API 4000 tandem mass spectrometer (AB Sciex, Foster City, CA, USA) was used as detector. Lower limit of quantification was shown to be as low as 0.05 ng/mL. This assay was found to be valid in the range from 0.05 to 1.50 ng/mL. At this time, all data resulting from this assay may only be used for research purposes. Statistical analysis of the data was performed using SPSS version 18 (SPSS, Chicago, IL, USA). All concentrations were specified as mean value±SD, including minimum and maximum values. We analyzed the data for all study participants and also performed subgroup analyses according to duration of LNGIUS use and to body mass index (BMI). Since the data were not normally distributed, we used nonparametric tests, namely, Spearman correlation coefficient for testing bivariate correlations, and the Mann– Whitney U test and the Kruskal–Wallis H test, where appropriate. Linear regression was used to test the predictive value of single variables, accounting for possible confounders. A cut-off point of p≤.05 was set as significant and p≤.01 as highly significant.
3. Results 3.1. Demographic data The mean age±SD of the 110 patients was 42.3±8.2 years (range 18.3–55.4 years). Mean duration±SD of LNG-IUS use was 1380±971 days (range 20–4045 days). The mean BMI±SD of the patients was 24.3±4.3 (range 17.8–38.5). The most common bleeding pattern in the population was secondary amenorrhea (42.9%), followed by spotting, defined as very light vaginal bleeding, with 37.7%. Regular menstrual cycles were reported by 6.5% and 11.7% reported irregular menstrual cycles under LNG-IUS. Occasional continuous bleeding lasting N7 days was reported by 1.2% of the patients. 3.2. Relationship between LNG plasma levels and duration of LNG-IUS use The mean LNG plasma level±SD in all patients was 147± 59 pg/mL (range 63–361 pg/mL). There was a highly significant negative correlation between the LNG plasma level and the duration of LNG-IUS use: the longer the period of LNG-IUS use, the lower the LNG plasma level as shown in Fig. 1 (Spearman r=−0.353, pb.001). Fig. 2 shows the average LNG concentrations for subjects grouped according to number of years of use, with the groups differing significantly from each other (p=.029). Pair-wise comparisons showed significant correlations between the first year and third, fifth, seventh and eighth year of LNG-
347
Mean LNG plasma level [pg/mL]
B. Seeber et al. / Contraception 86 (2012) 345–349
Duration of use (days)
Fig. 1. LNG plasma levels according to duration of LNG-IUS use (in days) at the time when blood samples were obtained. There was a slight but significant decline over time as indicated by the regression line of mean values (Spearman rank correlation coefficient −0.353, pb.001).
IUS use and between the second year and third year of use as shown in the figure. The mean LNG level±SD in the first year of LNG-IUS use was 191±71 pg/mL as shown in Fig. 2. In the second year of LNG-IUS use, the mean LNG level±SD was 157± 68 pg/mL and decreased to 134±41 pg/mL in the third year of LNG-IUS use (Fig. 2). In the fourth and fifth year, the LNG level±SD remained approximately stable at 150± 47 and 141±59 pg/mL, respectively. Even after the recommended LNG-IUS use time of 5 years is surpassed, plasma LNG is detectable. LNG plasma levels in the sixth and seventh year do not significantly differ from plasma levels in the second to fifth year. In the group of women who still had their LNG-IUS in the eighth year, the average LNG concentration±SD was measured at its minimum (117±59 pg/mL) as shown in Fig. 2. We further subdivided the study population into those who had used LNG-IUS for fewer than 5 years (n=75) and
those who had the IUS in place for 6 years and longer (n=35) to compare LNG concentrations. Mean LNG level±SD in the group with ≤5 years of application time was 157±62 pg/mL. Mean LNG level±SD in the group with extended application time was significantly lower at 126±44 pg/mL (p=.014). However, no one in either group experienced a pregnancy with a LNG-IUS. 3.3. Relationship between LNG plasma levels and BMI BMI, as determined at the time when blood sample was taken for analysis, showed a significant negative correlation with LNG plasma levels (Spearman rank correlation coefficient −0.239, p=.012). The mean LNG was significantly lower in obese patients with BMI ≥25 compared to patients with BMI b25 (p=.045). The mean LNG level±SD in patients with BMI b20 (n=13) was 165±57 pg/mL and in those with BMI 20–24.9 (n=57), it was 152±59 pg/mL. In
Fig. 2. Time-dependent LNG plasma concentrations. LNG plasma concentrations in all patients grouped according to the year after LNG-IUS insertion. Note the decline from first to third year and the plateau attained thereafter. Data represent mean; the error bars show the 95% confidence interval. Between-group differences are shown within the graph where pb.05 (⁎), pb.01 (⁎⁎) and pb.001 (⁎⁎⁎).
348
B. Seeber et al. / Contraception 86 (2012) 345–349
Fig. 3. LNG plasma levels grouped according to BMI, independent of duration of use. Data represent median and upper and lower quartiles (horizontal lines) with maximum and minimum values shown with whiskers (vertical lines). Outliers are represented by individual open circles. The SPSS output is calculated as follows: the box encompasses the interquartile range; the vertical lines with whiskers represent the values 1.5×the interquartile range; outliers — as defined by SPSS — are outside this range and are indicated as circles. The vertical lines are not of equal length, since they end with the last value within this range.
patients with BMI 25–29.9 (n=28) and BMI ≥30 (n=11), LNG levels were seen to be lower at 141±64 and 119±43 pg/mL, respectively. Data are summarized in Fig. 3. These differences in mean LNG levels between BMI groups were not statistically significant, likely due to the relatively few cases in the extreme weight groups. However both absolute BMI and duration of Mirena® use were independent predictors of LNG plasma concentration. 3.4. Relationship between LNG plasma levels and bleeding patterns No significant difference was found between LNG plasma levels and vaginal bleeding patterns (p=.684). 4. Discussion The present data show that plasma concentrations of LNG are detectable in all patients with LNG-IUS — even after the recommended application period of 5 years. This systemic hormone distribution could be a possible explanation for systemic side effects caused by the LNG IUS like acne, hirsutism, headache and mood changes. However, plasma levels were seen to be much lower than in other application forms of LNG. LNG concentrations are 5 to 10 times higher after oral ingestion (1500 to 2000 pg/mL) and higher with LNG implant/rod (350 pg/mL) for contraceptive use [16,17] and even higher in LNG dosage for emergency contraception (14,600 pg/mL) [18] compared with concentration seen with LNG-IUS (100–200 pg/mL). LNG plasma levels are also influenced by the patients' BMI as determined at the time of blood sampling. This relationship can be explained by the pharmacokinetics of LNG. Like other steroidal hormones, LNG is a lipophilic molecule and partitions into fat tissue [19]. In vitro experi-
ments with LNG concentrations typically seen after LNGIUS insertion showed a rapid uptake of levonorgestrel and an unsaturability of the fat tissue by the steroid [10]. A large volume of adipose tissue increases the volume of distribution for lipophilic substances like LNG and, as a consequence, decreases levels in plasma [20]. Therefore, obese people generally show lower LNG plasma levels. In this study, no difference in LNG plasma levels in patients with and without vaginal bleeding was found. This underscores the fact that changes in bleeding patterns like amenorrhea are caused by local effects of the LNG-IUS and do not depend on plasma levels. Although patients with LNG-IUS use of N6 years continue to have detectable LNG plasma levels, the concentrations are significantly lower than in those who have used the IUS for the recommended time of use of up to 5 years. The contraceptive efficacy of LNG-IUS should not be attributed to the LNG plasma levels because the contraceptive effect of the LNG-IUS is primarily local. Therefore, we still recommend changing the IUS after the recommended 5 years of use. However, we agree with other authors that there is a certain time window for IUS change and, therefore, the 5-year lifespan of the LNG-IUS can be slightly exceeded if necessary for important reasons [21]. Furthermore, in perimenopausal patients who no longer desire to change their IUS and who do not need any further contraception, prolonged use of LNG-IUS could be offered as a convenient and bleeding-free transition into the menopause [22].
References [1] Thonneau PF, Almont T. Contraceptive efficacy of intrauterine devices. Am J Obstet Gynecol 2008;198:248–53.
B. Seeber et al. / Contraception 86 (2012) 345–349 [2] Concin H, Bosch H, Hintermuller P, et al. Use of the levonorgestrelreleasing intrauterine system: an Austrian perspective. Curr Opin Obstet Gynecol 2009;21(Suppl 1):S1–9. [3] ACOG committee opinion. No. 337: Noncontraceptive uses of the levonorgestrel intrauterine system. Obstet Gynecol 2006;107: 1479–82. [4] Varila E, Wahlstrom T, Rauramo I. A 5-year follow-up study on the use of a levonorgestrel intrauterine system in women receiving hormone replacement therapy. Fertil Steril 2001;76:969–73. [5] Suvanto-Luukkonen E, Kauppila A. The levonorgestrel intrauterine system in menopausal hormone replacement therapy: five-year experience. Fertil Steril 1999;72:161–3. [6] Luukkainen T, Lahteenmaki P, Toivonen J. Levonorgestrel-releasing intrauterine device. Ann Med 1990;22:85–90. [7] Bayer Healthcare Pharmaceuticals Inc. Mirena prescribing information. Wayne, NJ, USA; 2009. [8] 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;61:70–7. [9] Xiao BL, Zhou LY, Zhang XL, Jia MC, Luukkainen T, Allonen H. Pharmacokinetic and pharmacodynamic studies of levonorgestrelreleasing intrauterine device. Contraception 1990;41:353–62. [10] Nilsson CG, Haukkamaa M, Vierola H, Luukkainen T. Tissue concentrations of levonorgestrel in women using a levonorgestrelreleasing IUD. Clin Endocrinol (Oxf) 1982;17:529–36. [11] Backman T, Huhtala S, Blom T, Luoto R, Rauramo I, Koskenvuo M. Length of use and symptoms associated with premature removal of the levonorgestrel intrauterine system: a nation-wide study of 17,360 users. BJOG 2000;107:335–9. [12] Toivonen J, Luukkainen T, Allonen H. Protective effect of intrauterine release of levonorgestrel on pelvic infection: three years' comparative
[13]
[14]
[15]
[16] [17]
[18]
[19] [20] [21]
[22]
349
experience of levonorgestrel- and copper-releasing intrauterine devices. Obstet Gynecol 1991;77:261–4. Sivin I, Stern J, Coutinho E, et al. Prolonged intrauterine contraception: a seven-year randomized study of the levonorgestrel 20 mcg/day (LNg 20) and the copper T380 Ag IUDS. Contraception 1991;44:473–80. Andersson K, Odlind V, Rybo G. Levonorgestrel-releasing and copper-releasing (Nova T) IUDs during five years of use: a randomized comparative trial. Contraception 1994;49:56–72. Moser C, Gschliesser A, Mattle V, Wildt L, Griesmacher A, Seger C. An ultra-sensitive online SPE-LC-MS/MS method for the quantification of levonorgestrel released from intrauterine devices. Anal Bioanal Chem 2011;400:2655–62. Orme ML, Back DJ, Breckenridge AM. Clinical pharmacokinetics of oral-contraceptive steroids. Clin Pharmacokinet 1983;8:95–136. Sivin I, Lahteenmaki P, Ranta S, et al. Levonorgestrel concentrations during use of levonorgestrel rod (LNG ROD) implants. Contraception 1997;55:81–5. Kives S, Hahn PM, White E, Stanczyk FZ, Reid RL. Bioavailability of the Yuzpe and levonorgestrel regimens of emergency contraception: vaginal vs. oral administration. Contraception 2005;71:197–201. Göretzlehner G. Praktische Hormontherapie in der Gynäkologie. 4th ed. Berlin: de Gruyter; 2003. Griffin JP. The textbook of pharmaceutical medicine. 6th ed. Oxford: Wiley-Blackwell; 2009. Hidalgo MM, Hidalgo-Regina C, Bahamondes MV, Monteiro I, Petta CA, Bahamondes L. Serum levonorgestrel levels and endometrial thickness during extended use of the levonorgestrel-releasing intrauterine system. Contraception 2009;80:84–9. Ronnerdag M, Odlind V. Health effects of long-term use of the intrauterine levonorgestrel-releasing system. A follow-up study over 12 years of continuous use. Acta Obstet Gynecol Scand 1999;78: 716–21.
Original research article
Quantitative levonorgestrel plasma level measurements in patients with regular and prolonged use of the levonorgestrel-releasing intrauterine system☆ Beata Seeber a , Stephanie C. Ziehr a , Aandrea Gschlieβer a, b, 1 , Christina Moser c , Verena Mattle a , Christoph Seger c , Andrea Griesmacher c , Nicole Concin b , Hans Concin d , Ludwig Wildt a,⁎ a
Department of Gynecological Endocrinology and Reproductive Medicine, Innsbruck Medical University, A-6020 Innsbruck, Austria b Department of Obstetrics and Gynecology Innsbruck Medical University, Innsbruck, Austria c Division of Mass Spectrometry and Chromatography, Institute of Medical and Chemical Laboratory Diagnostics (ZIMCL), University Hospital Innsbruck, Innsbruck, Austria d Department of Obstetrics and Gynecology, Landeskrankenhaus Bregenz, Bregenz, Austria Received 3 August 2011; revised 21 January 2012; accepted 27 January 2012
Abstract Background: The levonorgestrel-releasing intrauterine system (LNG-IUS) is well accepted as an easy-to-use contraceptive with an excellent side-effect profile. It contains a reservoir of 52 mg of levonorgestrel (LNG) with continuous release of the steroid. Its contraceptive use is approved for 5 years. The aim of this study was to determine the plasma concentration of LNG and its variation with time in patients with indwelling LNG-IUS Mirena®. Study Design: In this study, we determined LNG plasma concentrations in 110 women with LNG-IUS at different time points of use. Time from insertion of the system in the study population ranged from 20 days to 11.1 years. Quantitative LNG levels were determined using a validated liquid chromatography–tandem mass spectrometry assay. Results: The mean±SD LNG plasma level in all women was 147±59 pg/mL. A highly significant negative correlation between LNG plasma level and LNG-IUS time of use could be demonstrated. In the first year of use, LNG plasma level was as high as 191±71 pg/mL, decreasing to 157±68 pg/mL in the second year and 134±41 pg/mL in the third year. Even after exceeding the recommended period of LNG-IUS use, systemic LNG concentrations were detectable: 133±38 pg/mL in the sixth year, 133±48 pg/mL in the seventh year and 117±45 pg/mL in the eighth year. Furthermore, a significant negative correlation between LNG plasma level and body mass index could be shown. Conclusion: Systemic LNG concentrations can be found in all patients with LNG-IUS IUS. However, concentrations are much lower than in other forms of LNG application. Moreover, this study demonstrates that a systemic effect of LNG-IUS can also be found after the recommended contraceptive lifespan of 5 years. © 2012 Elsevier Inc. All rights reserved. Keywords: Levonorgestrel; LNG-intrauterine system; Prolonged use; BMI; Plasma level
1. Introduction The levonorgestrel-releasing intrauterine system (LNGIUS) is well known as a highly effective, safe and long☆
Funding: University funding. No external financial support was received for this study. ⁎ Corresponding author. Tel.: +43 512 504 23276; fax: +43 512 504 23277. E-mail address: [email protected] (L. Wildt). 1 Present address: Department of Ophthalmology and Optometry, Vienna Medical University, Vienna, Austria. 0010-7824/$ – see front matter © 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.contraception.2012.01.015
acting contraceptive method with few side effects [1,2]. Besides its contraceptive use, the hormone-releasing IUS has some therapeutical benefits including a reduction in dysmenorrhea and menstrual bleeding and as endometrial protection in women using estrogen therapy [3-5]. The LNG-IUS consists of a T-shaped polyethylene frame body with a cylindrical reservoir containing 52 mg of LNG. Average release rate of LNG is 14 mcg LNG/24 h. Initially, 20 mcg LNG/24 h is released into the uterine cavity, declining progressively to 11 mcg LNG/24 h after 5 years of use [6,7].
346
B. Seeber et al. / Contraception 86 (2012) 345–349
Mirena® is approved for up to 5 years of use. However, there is evidence of effectiveness beyond this period. Even after 7 years of use, the cumulative pregnancy rate is as low as 1.1% [8]. Levonorgestrel is the biologically active form of norgestrel (D-norgestrel) and is a synthetic second-generation gestagen. As part of the LNG-IUS, the hormone exerts its effect primarily locally within the uterine cavity. It exerts an antiproliferative effect on the endometrium leading to endometrial decidualization and atrophy and a thickening of cervical mucus which prevents sperm penetration [9,10]. Side effects of the hormone-releasing IUS are generally rare and, in most cases, caused by the local effect of the IUS. Common causes for premature removal are excessive bleeding and spotting, intrauterine infections and pelvic pain [8,11]. Even though systemic side effects are generally infrequent, hormonal side effects like hirsutism, acne, headache and mood changes are also reasons for elective LNG-IUS removal in the first 36 months of use [12]. LNG-IUS users have more discontinuations because of hair and skin changes and headache than users of the copper intrauterine device (IUD) [13]. In the first 5 years of use, approximately 12% of women prematurely discontinue the LNG-IUS because of hormone-attributable complaints [14]. Hormonal side effects could possibly be explained by systemic LNG distribution. The aim of this study was to determine the extent of and time course of change in LNG plasma distribution in patients with in-dwelling LNG-IUS.
2. Materials and methods This two-center study was performed at the University Hospital in Innsbruck, Austria, and at the General Hospital in Bregenz, Austria. Women were eligible to participate if they currently had a levonorgestrel-containing IUS with known date of insertion. An informed consent form, previously approved by the Innsbruck Medical University Ethical Board (IRB), was obtained to store aliquots of blood samples taken for routine medical reasons and to use them for research purposes. Evaluated were 113 patients with LNG-IUS at different time points of use. Three women were excluded from analyses because of unknown date of insertion (n=1) and unknown date of blood sampling (n=2). In total, 138 LNG plasma level concentrations were measured during the study time period in this retrospective study. If multiple blood samples were obtained from the same patient (n=28), the most recent sample was used for further analysis. Consequently, 110 patients, each with a single LNG plasma concentration, were included in the data analysis. There were 75 women who had used the LNG-IUS within the approved time of use of ≤5 years and 35 women with prolonged use of ≥6 years. In this study population, time of IUS use ranged from 20 days to 11.1 years.
A validated assay was used for LNG plasma concentration measurement. A selective and sensitive liquid chromatography–tandem mass spectrometry method for the determination of LNG plasma levels was developed [15]. An API 4000 tandem mass spectrometer (AB Sciex, Foster City, CA, USA) was used as detector. Lower limit of quantification was shown to be as low as 0.05 ng/mL. This assay was found to be valid in the range from 0.05 to 1.50 ng/mL. At this time, all data resulting from this assay may only be used for research purposes. Statistical analysis of the data was performed using SPSS version 18 (SPSS, Chicago, IL, USA). All concentrations were specified as mean value±SD, including minimum and maximum values. We analyzed the data for all study participants and also performed subgroup analyses according to duration of LNGIUS use and to body mass index (BMI). Since the data were not normally distributed, we used nonparametric tests, namely, Spearman correlation coefficient for testing bivariate correlations, and the Mann– Whitney U test and the Kruskal–Wallis H test, where appropriate. Linear regression was used to test the predictive value of single variables, accounting for possible confounders. A cut-off point of p≤.05 was set as significant and p≤.01 as highly significant.
3. Results 3.1. Demographic data The mean age±SD of the 110 patients was 42.3±8.2 years (range 18.3–55.4 years). Mean duration±SD of LNG-IUS use was 1380±971 days (range 20–4045 days). The mean BMI±SD of the patients was 24.3±4.3 (range 17.8–38.5). The most common bleeding pattern in the population was secondary amenorrhea (42.9%), followed by spotting, defined as very light vaginal bleeding, with 37.7%. Regular menstrual cycles were reported by 6.5% and 11.7% reported irregular menstrual cycles under LNG-IUS. Occasional continuous bleeding lasting N7 days was reported by 1.2% of the patients. 3.2. Relationship between LNG plasma levels and duration of LNG-IUS use The mean LNG plasma level±SD in all patients was 147± 59 pg/mL (range 63–361 pg/mL). There was a highly significant negative correlation between the LNG plasma level and the duration of LNG-IUS use: the longer the period of LNG-IUS use, the lower the LNG plasma level as shown in Fig. 1 (Spearman r=−0.353, pb.001). Fig. 2 shows the average LNG concentrations for subjects grouped according to number of years of use, with the groups differing significantly from each other (p=.029). Pair-wise comparisons showed significant correlations between the first year and third, fifth, seventh and eighth year of LNG-
347
Mean LNG plasma level [pg/mL]
B. Seeber et al. / Contraception 86 (2012) 345–349
Duration of use (days)
Fig. 1. LNG plasma levels according to duration of LNG-IUS use (in days) at the time when blood samples were obtained. There was a slight but significant decline over time as indicated by the regression line of mean values (Spearman rank correlation coefficient −0.353, pb.001).
IUS use and between the second year and third year of use as shown in the figure. The mean LNG level±SD in the first year of LNG-IUS use was 191±71 pg/mL as shown in Fig. 2. In the second year of LNG-IUS use, the mean LNG level±SD was 157± 68 pg/mL and decreased to 134±41 pg/mL in the third year of LNG-IUS use (Fig. 2). In the fourth and fifth year, the LNG level±SD remained approximately stable at 150± 47 and 141±59 pg/mL, respectively. Even after the recommended LNG-IUS use time of 5 years is surpassed, plasma LNG is detectable. LNG plasma levels in the sixth and seventh year do not significantly differ from plasma levels in the second to fifth year. In the group of women who still had their LNG-IUS in the eighth year, the average LNG concentration±SD was measured at its minimum (117±59 pg/mL) as shown in Fig. 2. We further subdivided the study population into those who had used LNG-IUS for fewer than 5 years (n=75) and
those who had the IUS in place for 6 years and longer (n=35) to compare LNG concentrations. Mean LNG level±SD in the group with ≤5 years of application time was 157±62 pg/mL. Mean LNG level±SD in the group with extended application time was significantly lower at 126±44 pg/mL (p=.014). However, no one in either group experienced a pregnancy with a LNG-IUS. 3.3. Relationship between LNG plasma levels and BMI BMI, as determined at the time when blood sample was taken for analysis, showed a significant negative correlation with LNG plasma levels (Spearman rank correlation coefficient −0.239, p=.012). The mean LNG was significantly lower in obese patients with BMI ≥25 compared to patients with BMI b25 (p=.045). The mean LNG level±SD in patients with BMI b20 (n=13) was 165±57 pg/mL and in those with BMI 20–24.9 (n=57), it was 152±59 pg/mL. In
Fig. 2. Time-dependent LNG plasma concentrations. LNG plasma concentrations in all patients grouped according to the year after LNG-IUS insertion. Note the decline from first to third year and the plateau attained thereafter. Data represent mean; the error bars show the 95% confidence interval. Between-group differences are shown within the graph where pb.05 (⁎), pb.01 (⁎⁎) and pb.001 (⁎⁎⁎).
348
B. Seeber et al. / Contraception 86 (2012) 345–349
Fig. 3. LNG plasma levels grouped according to BMI, independent of duration of use. Data represent median and upper and lower quartiles (horizontal lines) with maximum and minimum values shown with whiskers (vertical lines). Outliers are represented by individual open circles. The SPSS output is calculated as follows: the box encompasses the interquartile range; the vertical lines with whiskers represent the values 1.5×the interquartile range; outliers — as defined by SPSS — are outside this range and are indicated as circles. The vertical lines are not of equal length, since they end with the last value within this range.
patients with BMI 25–29.9 (n=28) and BMI ≥30 (n=11), LNG levels were seen to be lower at 141±64 and 119±43 pg/mL, respectively. Data are summarized in Fig. 3. These differences in mean LNG levels between BMI groups were not statistically significant, likely due to the relatively few cases in the extreme weight groups. However both absolute BMI and duration of Mirena® use were independent predictors of LNG plasma concentration. 3.4. Relationship between LNG plasma levels and bleeding patterns No significant difference was found between LNG plasma levels and vaginal bleeding patterns (p=.684). 4. Discussion The present data show that plasma concentrations of LNG are detectable in all patients with LNG-IUS — even after the recommended application period of 5 years. This systemic hormone distribution could be a possible explanation for systemic side effects caused by the LNG IUS like acne, hirsutism, headache and mood changes. However, plasma levels were seen to be much lower than in other application forms of LNG. LNG concentrations are 5 to 10 times higher after oral ingestion (1500 to 2000 pg/mL) and higher with LNG implant/rod (350 pg/mL) for contraceptive use [16,17] and even higher in LNG dosage for emergency contraception (14,600 pg/mL) [18] compared with concentration seen with LNG-IUS (100–200 pg/mL). LNG plasma levels are also influenced by the patients' BMI as determined at the time of blood sampling. This relationship can be explained by the pharmacokinetics of LNG. Like other steroidal hormones, LNG is a lipophilic molecule and partitions into fat tissue [19]. In vitro experi-
ments with LNG concentrations typically seen after LNGIUS insertion showed a rapid uptake of levonorgestrel and an unsaturability of the fat tissue by the steroid [10]. A large volume of adipose tissue increases the volume of distribution for lipophilic substances like LNG and, as a consequence, decreases levels in plasma [20]. Therefore, obese people generally show lower LNG plasma levels. In this study, no difference in LNG plasma levels in patients with and without vaginal bleeding was found. This underscores the fact that changes in bleeding patterns like amenorrhea are caused by local effects of the LNG-IUS and do not depend on plasma levels. Although patients with LNG-IUS use of N6 years continue to have detectable LNG plasma levels, the concentrations are significantly lower than in those who have used the IUS for the recommended time of use of up to 5 years. The contraceptive efficacy of LNG-IUS should not be attributed to the LNG plasma levels because the contraceptive effect of the LNG-IUS is primarily local. Therefore, we still recommend changing the IUS after the recommended 5 years of use. However, we agree with other authors that there is a certain time window for IUS change and, therefore, the 5-year lifespan of the LNG-IUS can be slightly exceeded if necessary for important reasons [21]. Furthermore, in perimenopausal patients who no longer desire to change their IUS and who do not need any further contraception, prolonged use of LNG-IUS could be offered as a convenient and bleeding-free transition into the menopause [22].
References [1] Thonneau PF, Almont T. Contraceptive efficacy of intrauterine devices. Am J Obstet Gynecol 2008;198:248–53.
B. Seeber et al. / Contraception 86 (2012) 345–349 [2] Concin H, Bosch H, Hintermuller P, et al. Use of the levonorgestrelreleasing intrauterine system: an Austrian perspective. Curr Opin Obstet Gynecol 2009;21(Suppl 1):S1–9. [3] ACOG committee opinion. No. 337: Noncontraceptive uses of the levonorgestrel intrauterine system. Obstet Gynecol 2006;107: 1479–82. [4] Varila E, Wahlstrom T, Rauramo I. A 5-year follow-up study on the use of a levonorgestrel intrauterine system in women receiving hormone replacement therapy. Fertil Steril 2001;76:969–73. [5] Suvanto-Luukkonen E, Kauppila A. The levonorgestrel intrauterine system in menopausal hormone replacement therapy: five-year experience. Fertil Steril 1999;72:161–3. [6] Luukkainen T, Lahteenmaki P, Toivonen J. Levonorgestrel-releasing intrauterine device. Ann Med 1990;22:85–90. [7] Bayer Healthcare Pharmaceuticals Inc. Mirena prescribing information. Wayne, NJ, USA; 2009. [8] 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;61:70–7. [9] Xiao BL, Zhou LY, Zhang XL, Jia MC, Luukkainen T, Allonen H. Pharmacokinetic and pharmacodynamic studies of levonorgestrelreleasing intrauterine device. Contraception 1990;41:353–62. [10] Nilsson CG, Haukkamaa M, Vierola H, Luukkainen T. Tissue concentrations of levonorgestrel in women using a levonorgestrelreleasing IUD. Clin Endocrinol (Oxf) 1982;17:529–36. [11] Backman T, Huhtala S, Blom T, Luoto R, Rauramo I, Koskenvuo M. Length of use and symptoms associated with premature removal of the levonorgestrel intrauterine system: a nation-wide study of 17,360 users. BJOG 2000;107:335–9. [12] Toivonen J, Luukkainen T, Allonen H. Protective effect of intrauterine release of levonorgestrel on pelvic infection: three years' comparative
[13]
[14]
[15]
[16] [17]
[18]
[19] [20] [21]
[22]
349
experience of levonorgestrel- and copper-releasing intrauterine devices. Obstet Gynecol 1991;77:261–4. Sivin I, Stern J, Coutinho E, et al. Prolonged intrauterine contraception: a seven-year randomized study of the levonorgestrel 20 mcg/day (LNg 20) and the copper T380 Ag IUDS. Contraception 1991;44:473–80. Andersson K, Odlind V, Rybo G. Levonorgestrel-releasing and copper-releasing (Nova T) IUDs during five years of use: a randomized comparative trial. Contraception 1994;49:56–72. Moser C, Gschliesser A, Mattle V, Wildt L, Griesmacher A, Seger C. An ultra-sensitive online SPE-LC-MS/MS method for the quantification of levonorgestrel released from intrauterine devices. Anal Bioanal Chem 2011;400:2655–62. Orme ML, Back DJ, Breckenridge AM. Clinical pharmacokinetics of oral-contraceptive steroids. Clin Pharmacokinet 1983;8:95–136. Sivin I, Lahteenmaki P, Ranta S, et al. Levonorgestrel concentrations during use of levonorgestrel rod (LNG ROD) implants. Contraception 1997;55:81–5. Kives S, Hahn PM, White E, Stanczyk FZ, Reid RL. Bioavailability of the Yuzpe and levonorgestrel regimens of emergency contraception: vaginal vs. oral administration. Contraception 2005;71:197–201. Göretzlehner G. Praktische Hormontherapie in der Gynäkologie. 4th ed. Berlin: de Gruyter; 2003. Griffin JP. The textbook of pharmaceutical medicine. 6th ed. Oxford: Wiley-Blackwell; 2009. Hidalgo MM, Hidalgo-Regina C, Bahamondes MV, Monteiro I, Petta CA, Bahamondes L. Serum levonorgestrel levels and endometrial thickness during extended use of the levonorgestrel-releasing intrauterine system. Contraception 2009;80:84–9. Ronnerdag M, Odlind V. Health effects of long-term use of the intrauterine levonorgestrel-releasing system. A follow-up study over 12 years of continuous use. Acta Obstet Gynecol Scand 1999;78: 716–21.