Contraception xx (2015) xxx xxx

Original research article

Relationship between etonogestrel level and BMI in women using the

contraceptive implant for more than 1 year
Kathleen M. Morrella, , Serge Cremersb, Carolyn L. Westhoffc, Anne R. Davisa
a
Department of Obstetrics and Gynecology, Columbia University Medical Center, 622 W 168th Street, PH 16-69, New York, NY, 10032, USA
b
Biomarkers Core Laboratory, Irving Institute for Clinical and Translational Research, Columbia University Medical Center, 630 W 168th Street, PH 10-
105b, New York, NY, 10027, USA
c
Department of Obstetrics and Gynecology and Epidemiology, Columbia University Medical Center, 622 W 168th Street, PH 16-69, New York, NY, USA
Received 22 May 2015; revised 9 September 2015; accepted 9 November 2015

Abstract

Progestin-only contraceptive methods, including the 3-year, single-rod etonogestrel (ENG) implant, may be preferred for obese women
to avoid additional estrogen-related thrombosis risk; however, whether obese women receive an ENG sufficient dose to suppress ovulation
is understudied. Our analysis expands on the limited information currently available by studying ENG levels related to body mass index
(BMI) in a community sample of primarily Hispanic women. This cross-sectional, descriptive study of 52 long-term implant users found
comparable ENG levels across a wide BMI range (p=.1). These results further support that ENG levels are independent of BMI through 3
years of implant use and are thus reassuring that ENG implants will be effective for women of all BMIs. 2015 Elsevier Inc. All rights
reserved.

Keywords: Contraception; Obesity; Implant; Etonogestrel

1. Introduction series of 3 women before and after gastric bypass surgery

Two thirds of reproductive-age women in the United States
are either overweight or obese [1]. Progestin-only methods of
contraception such as the 3-year, single-rod etonogestrel
(ENG) implant (Nexplanon, Merck & Co., Inc., Whitehouse
Station, NJ, USA) may be preferred for obese women to avoid
additional estrogen-related thrombosis risk with combined
methods. The ENG implant releases 3545 mcg per day of
ENG at 1 year of use with declining levels thereafter and
prevents pregnancy by suppressing ovulation and increasing
the viscosity of cervical mucus [2,3].
Body composition may affect contraceptive hormone
metabolism because of the increased volume of distribution
and altered plasma clearance in obese individuals [4]. Three
studies describe ENG levels in obese implant users: a case
[5], a pharmacokinetic study comparing 13 obese women
to 4 normal-weight controls [6] and a subanalysis of the
CHOICE Project examining levels in the third and fourth
year of use [7].
Our analysis expands on the limited information available
thus far by studying ENG levels related to body mass index
(BMI) in a community sample of mainly Hispanic women.
Our primary objective was to quantify ENG levels in implant
users after at least 1 year of use in women across BMI
categories. A secondary objective was to describe what
proportion, if any, fell below a suggested level for ovulation
suppression (90 pg/mL) [8]. We hypothesized that ENG levels
would be lower in women of higher BMI categories and that
none of the women would exhibit an ENG level below the
suggested threshold for ovulation suppression.

ClinicalTrials.gov: NCT01855620.
2. Material and methods
Corresponding author.
E-mail addresses: kmm2112@columbia.edu (K.M. Morrell),
We conducted a cross-sectional, descriptive study of a
sc2752@columbia.edu (S. Cremers), clw3@columbia.edu (C.L. convenience sample of normal-weight (BMI b 25 kg/m2),
Westhoff), ard4@columbia.edu (A.R. Davis). overweight (BMI 25 and b 30 kg/m2) and obese women
http://dx.doi.org/10.1016/j.contraception.2015.11.005
0010-7824/ 2015 Elsevier Inc. All rights reserved.
2 K.M. Morrell et al. / Contraception xx (2015) xxxxxx

(BMI 30 kg/m2) using the single-rod contraceptive ENG the last 6 months. Table 1 presents participants
implant for at least 1 year. We obtained approval for this study characteristics and ENG levels.
from the local Institutional Review Board. We offered ENG levels were comparable across BMI categories
enrollment to all English- or Spanish-speaking women aged (Table 1, Fig. 1, p=.1) and no participant fell below the
1845 years at a Title X Family Planning Clinic and private suggested threshold for ovulation of 90 pg/mL. Consistent
practice using the ENG implant for 12 months. with an expected slow decline, there was a trend of
After informed consent, participants completed a ques- decreasing ENG level as time since implant insertion
tionnaire to collect demographic information and relevant increased (p=.08). After controlling for time, ENG levels
medical history. A research assistant measured the partici- continued to be similar across the range of BMI included in
pants height and weight and collected a blood sample from this study (p=.13). In all comparisons, the overweight
the arm contralateral to the implant. The time from insertion women had slightly higher ENG levels than the normal-
of the implant to blood draw, in months, was ascertained from weight or obese women whose levels were similar.
the participants medical record or her implant identification
card. Participants received $25 compensation. Given resource
constraints, we planned to enroll 50 women and to oversample 4. Discussion
participants in the higher BMI categories. This sample size
allowed more precise estimates in the groups of interest than This study of primarily Hispanic women who had ENG
published research at that time [5,6]. implants placed in a community setting found therapeutic
Blood samples were allowed to clot for at least 10 min at levels in all women after 1 year of use regardless of BMI.
room temperature and then separated at 3400 rpm by These results are consistent with McNicholas results among
centrifuge. The serum was then transferred and stored in primarily African-American women who received their
aliquots at 80C until analysis. The analysis was performed implant in the framework of a study protocol [7]. These results
in a single batch with a validated ENG assay using liquid also support the observed high efficacy rates in obese ENG
chromatographymass spectrometry (LCMS/MS) at the implant users (0.23 per 100 woman-years) [11].
Biomarkers Core Laboratory of the Irving Institute of Clinical The suggested ENG threshold for suppression of ovulation (90
and Translational Research at Columbia University Medical pg/mL) is based on a radioimmunoassay (RIA) and not LC
Center. The laboratory measured ENG in serum by LC MS/MS, which was used in this study and for which such a
MS/MS after liquid/liquid extraction using D-8 Proges-terone threshold has not been determined [8,9]. While we lack data
as the internal standard. The method was linear between 1 and directly comparing LCMS/MS with RIA results for ENG, the
100 pg/mL (Limit of Quantification: 50 pg/mL). The intra- ENG steady-state levels measured during contraceptive ring use
and interassay coefficients of variation were b 6% and b 13%, among normal-weight women in two studies, one via RIA and the
respectively. Validation was carried out on three subjects other via LCMS/MS, were superimposable, suggesting that
during a study of the contraceptive ring where ENG levels at these assays give similar results at those serum levels [9,10,12].
steady state were 1500 pg/mL [9]. The LCMS/MS levels of Because ENG levels during ring use are much
ENG in normal-weight women were consistent with those
reported in an earlier study that used a radioimmuno-assay
Table 1
[9,10]. McNicholas and colleagues [7] used the same assay Participant characteristics by BMI category.
conducted by the same laboratory; thus, those results are
Normal Overweight Obese
commensurate with the present results.
(n= 10) (n= 19) (n= 22)
In our analysis, we used descriptive statistics to compare
Age
demographic variables. We analyzed the effect of BMI
Median 27 {2132} 29 {2043} 27 {1844}
category and time on ENG level using the KruskalWallis Race
test. We conducted multivariable analyses with linear Non-Hispanic white 0 3 1
regression using time and BMI category with ENG level as Hispanic 8 16 21
the dependent variable with SPSS (IBM Corp. released 2013. African-American 2 0 1
Gravidity
IBM SPSS Statistics, Version 22.0. Armonk, NY, USA).
Median {range} 4 {17} 3 {015} 3 {011}
Parity
Median {range} 2 {04} 1 {05} 1 {06}
3. Results Time since insertion (months)
Median 38 35 34
Range {1754} {1442} {1339}
We enrolled 52 women from December 2012 through October
BMI
2013: normal weight (n= 10), overweight (n= 19) and obese Median 23 27 35
(n=23). Demographic characteristics were similar across BMI Range {2125} {2529} {3156}
groups. The median time since implant insertion was 35 months ENG (pg/mL)
on the day of blood sampling (range, 1354 months). No Median 216 288 225
Range {117523} {134558} {128511}
participants had taken an enzyme-inducing medication within
 K.M. Morrell et al. / Contraception xx (2015) xxxxxx
Fig. 1. (A) ENG levels by BMI group (B) ENG level vs. time by BMI group
higher, we cannot, however, say with any certainty that
RIA and LCMS/MS ENG levels would be comparable at
the much lower exposure that occurs during implant use.
The first studies of obese women using this implant were
small and measured levels at less than 1 year of use [5,6]. A
recent larger study showed no difference in ENG levels between
women of different BMI categories at 3 (n= 92) and 4 (n=47)
years of use [7]. At that longer duration of use, some ENG levels
among overweight and obese women fell below 90 pg/mL, which
was presumed to be the threshold for ovulation suppression. It is
unclear if the high efficacy of the ENG implant seen in that study
(0 per 100 woman-years) is due purely to ovulation suppression
or to cervical mucus effects even if ovulation occasionally occurs.
A larger pharmacody-namic study using a well-validated LC
MS/MS assay for ENG would have to be performed to identify a
valid threshold value.
This study has limited power due to great ENG variability
and the small sample size; however, these results further
support no effect of increasing BMI on ENG levels through 3
3
years of implant use and are reassuring that the implant is
an effective option for women of all BMIs. Further
pharmacodynamic studies to define ENG levels (as
measured by LCMS/MS) that are associated with
complete ovulation suppression are needed.
Acknowledgments
The Society of Family Planning funded this study but
had no role in study design, collection, analysis or
interpretation of the data, writing of the report or the
decision to submit the article for publication.
This publication was supported by the National Center
for Advancing Translational Sciences, National Institutes
of Health, through Grant Number UL1 TR000040. The
content is solely the responsibility of the authors and does
not necessarily represent the official views of the NIH.
References
[1] Ogden CL, Carroll MD, Kit BK, Flegal KM. Prevalence of
childhood and adult obesity in the United States, 20112012. JAMA
2014;311(8):80614, http://dx.doi.org/10.1001/jama.2014.732.
[2] Merck & Co., Inc. Nexplanon prescribing information. Accessed
September 3, 2015 at, http://www.merck.com/product/usa/pi_circulars/n/
nexplanon/nexplanon_pi.pdf2015.
[3] Croxatto HB, Mkrinen L. The pharmacodynamics and efficacy of
Implanon. Contraception 1998;58:91S7S.
[4] Cheymol G. Effects of obesity on pharmacokinetics: implications for
drug therapy. Clin Pharmacokinet 2000;39(3):21531.
[5] Ciangura C, Corigliano N, Basdevant A, Mouly S, Declves X,
Touraine P, et al. Etonogestrel concentrations in morbidly obese
women following Roux-en-Y gastric bypass surgery: three case
reports. Contraception 2011;84:64951, http://dx.doi.org/10.1016/
j.contraception.2011.03.015.
[6] Mornar S, Chan LN, Mistretta S, Neustadt A, Martins S, Gilliam M.
Pharmacokinetics of the etonogestrel contraceptive implant in obese
women. Am J Obstet Gynecol 2012;207(2):110.e16,
http://dx.doi.org/ 10.1016/j.ajog.2012.05.002.
[7] McNicholas C, Maddipati R, Zhao Q, Swor E, Peipert JF. Use of the
etonogestrel implant and levonorgestrel intrauterine device beyond the U.S.
Food and Drug Administration-approved duration. Obstet Gynecol
2015;125(3):599604, http://dx.doi.org/10.1097/AOG.0000000000000690.
[8] Daz S, Pavez M, Moo-Young AJ, Bardin CW, Croxatto HB.
Clinical trial with 3-keto-desogestrel subdermal implants.
Contraception 1991;44(4):393408.
[9] Thomas T, Petrie K, Shim J, Abildskov KM, Westhoff CL, Cremers
S. A UPLCMS/MS method for therapeutic drug monitoring of
etonogestrel. Ther Drug Monit 2013;35(6):8448, http://dx.doi.org/
10.1097/FTD.0b013e31829a10fa.
[10] Timmer CJ, Mulders TM. Pharmacokinetics of etonogestrel and
ethinyl estradiol released from a combined contraceptive vaginal
ring. Clin Pharmacokinet 2000;39:23342.
[11] Xu H, Wade JA, Peipert JF, Zhao Q, Madden T, Secura GM.
Contraceptive failure rates of etonogestrel subdermal implants in
overweight and obese women. Obstet Gynecol 2012;120(1):216,
http://dx.doi.org/10.1097/AOG.0b013e318259565a.
[12] Westhoff CL, Torgal AH, Mayeda ER, Petrie K, Thomas T,
Dragoman M, et al. Pharmacokinetics and ovarian suppression
during use of a contraceptive vaginal ring in normal-weight and
obese women. Am J Obstet Gynecol 2012;207(1):39.e16,
http://dx.doi.org/10.1016/ j.ajog.2012.04.022.