A multicentre efficacy and safety study of the single contraceptive implant Implanon ®

Horacio B. Croxatto1, János Urbancsek2, Rebeca Massai1, Herjan Coelingh Bennink3, Agaath van Beek3,4 and the Implanon® Study Group*

1 Chilean Institute of Reproductive Medicine, Santiago, Chile and the 2 Department of Obstetrics and Gynaecology, Semmelweis University, Budapest, Hungary and 3 Scientific Development Group, NV Organon, Oss, The Netherlands


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
An open, multicentre study was performed to assess efficacy, safety and acceptability of the single-rod contraceptive implant Implanon®. The study involved 635 young healthy women, who were sexually active and of childbearing potential. The women were followed up every 3 months over the entire study period. Originally the study was designed for 2 years, but was extended to 3 years in a group of 147 women from two centres. Altogether, 21 centres in nine different countries participated. The average age of the women was 29 years (range 18–42 years), of whom 83.5% had been pregnant in the past. No pregnancy occurred during treatment with Implanon®, resulting in a Pearl Index of 0 (95% confidence interval: 0.0–0.2). In the first 2 years, 31% had discontinued the treatment. Of the 147 women in the study extension, nine discontinued (6%) treatment. Bleeding irregularities was the main reason for discontinuation during the first 2 years of use (17.2%) and adverse experiences in the third year (3.4%). Implant insertion and removal were fast and uncomplicated in the vast majority (97%) of cases. Return of fertility was prompt. In conclusion, Implanon® has excellent contraceptive action during its lifetime of 3 years. The safety profile is acceptable and not essentially different from progestogens in general.

Key words: contraceptive implant/efficacy/Implanon®?/safety/vaginal bleeding


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Implanon® is a single-rod implant (NV Organon, Oss, The Netherlands) made of an ethylene vinyl acetate copolymer (EVA) with a core containing 68 mg of etonogestrel (3-ketodesogestrel). The implant has a length of 40 mm and a diameter of 2 mm and is provided in a sterile disposable inserter for subdermal application. Contraceptive action is mainly by inhibition of ovulation and lasts for 3 years (Mäkäräinen et al., 1998Go).

The study was designed to determine contraceptive reliability by means of a pregnancy rate, safety by means of regular medical examinations and assessment of adverse experiences, and acceptability by means of vaginal bleeding patterns. Data were collected in 21 centres situated in nine different countries. The number of women required to provide at least 10 000 cycles of 28 days was calculated to be ~600.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
The study was designed to include healthy women, who were sexually active, of childbearing potential and requesting contraception. Their age had to be between 18 and 40 years. The volunteers were included if they had menstrual cycles with a length of 24–35 days with an intra-individual variation of at most ±3 days. Pregnancy and breastfeeding were exclusion criteria and so was weight outside 80–130% of ideal (Metropolitan Height and Weight Tables, 1983Go). In general, we used the same recommendations as published in `Labeling guidance text for progestin-only oral contraceptives' (Corfman, 1995Go). Liver enzyme inducing medication was not allowed during the study. All volunteers gave their consent in writing after receiving full information on Implanon® and the requirements of the study.

The implant was inserted in the inner aspect of the non-dominant upper arm. The woman had to be on or between days 1–5 of a spontaneous menses. The protocol for the study was submitted to National Health Authorities and to Ethics Committees of the respective sites. The study started in November 1991 and was completed in December 1996. Originally the study was designed to last for 2 years, which was later extended to 3 years in two centres. The decision was based upon data obtained in-vitro, ex-vivo and non-human data. Human data, obtained in sterilized women exposed to an extracted implant, revealed that ovulation inhibition was sustained for longer than 2 years (Davies et al., 1993Go).

The assessment schedule was as follows: at baseline medical history and gynaecological history were taken and a physical and pelvic examination including a cervical smear according to Papanicolaou (Pap) were done. The Pap smear at baseline was omitted if results Pap I or II were obtained <1 year previously. The medical examinations were repeated once every year or at the time of implant removal. Body height was recorded at baseline and so were weight and blood pressure (BP). The latter two variables were also recorded at months 3, 6, 12, 18, 24, 27, 30, 33 and 36. Status at implant site was checked once every 3 months and so was the bleeding card to see whether a pregnancy test was indicated in case of amenorrhoea. Women were asked to communicate unwanted effects at each visit and were instructed in further proceedings of the study.

Statistical methods
In the calculation of the percentages of discontinuation, the denominator excluded women who were lost to follow-up. The analysis of the bleeding pattern was done using 90-day reference periods (Rodriguez et al., 1976Go; Belsey et al., 1986; Belsey and Farley, 1988Go). The first reference period of 90 days started on the day of implant placement. For the analysis of the bleeding pattern, the following definitions were used: amenorrhoea was defined as no bleeding or spotting throughout the reference period. Infrequent bleeding was defined as fewer than three bleeding-spotting episodes (B-S) starting within a reference period, excluding amenorrhoea. Frequent bleeding was defined as more than five B-S episodes starting within a reference period and prolonged bleeding was defined by at least one B-S episode starting within a reference period and lasting >14 days. A B-S episode was defined as any set of >=1 consecutive bleeding or spotting days bounded at each end by at least 1 bleeding-free day. Coding of adverse experiences was done using the WHO adverse reactions terminology (WHO Adverse Reaction Dictionary, 1994).


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
In 635 volunteers, the implant was inserted. Not all women fulfilled the inclusion/exclusion criteria. Most importantly, four women were pregnant at the start of the study. They received an implant as the nidation bleeding was mistaken for a menses. One of these pregnancies was ectopic and a salpingectomy was indicated. Two women had an induced abortion and one gave birth to a healthy baby girl. In the latter woman, the implant was removed 30 days after it was placed. The other three women continued with their implants. Other protocol violations were: age >40 years, weight >130% of ideal and irregular menstrual cycles. Once a woman was included in the study, her data were included in the results.

Centres varied in size between six and 114 volunteers. The centres in Budapest (n = 114) and Santiago (n = 107) contributed substantially more volunteers than the other centres. A total of 147 volunteers from these two centres entered the third year of study. Safety and bleeding data from Chile (n = 107), Hungary with three centres (n = 195) and the remaining centres (n = 333) were compared to detect possible centre effects. Regarding bleeding data, no major differences were observed. Safety reporting, however, showed a difference which is the reason for presenting the adverse experiences by centre (region).

Relevant characteristics of the study population are presented in Table IGo. The average age at baseline of the total study population was 29 years and 41% of the women were smokers. Baseline characteristics of those who continued for a third year are also presented. Although those who continued were slightly younger at baseline, they were also of slightly higher parity. Of the total study population, 9% were >75 kg in weight at the start of the study, with one woman weighing as much as 110 kg. Those who continued with the implant for a third year were, in 3.4% of cases, >75 kg in weight at the start of the study.


View this table:
[in this window]
[in a new window]
 
Table I. Characteristics of the Implanon® study population at baseline
 
The total exposure to Implanon® was 1200 woman-years or 15 653 28-day cycles. No pregnancies occurred during treatment, resulting in a Pearl Index of 0 (95% confidence interval 0.0–0.2). Altogether, 436 women completed 2 years of treatment (out of 635). In all, 162 women were given the option to participate for a third treatment year. From these, 147 decided to continue, of whom 137 completed. Cycles provided by those who discontinued treatment are included in the total of 15 653.

Discontinuation was as follows: 10% (n = 64) after 6 months. At 12 months, another 62 women had discontinued treatment, resulting in a total of 126 (20%). After 24 months the total number of women who discontinued was 196 (31%). In the third year, 6% (n = 9) discontinued. For calculation of the percentage of discontinuation at 36 months, a different denominator (n = 147) is required due to extension of the study in a sub-group of women only. The reasons for discontinuation are presented in Table IIGo, together with numbers and percentages.


View this table:
[in this window]
[in a new window]
 
Table II. Reasons for discontinuation
 
The majority of women who discontinued the method, 17.2% in the first 2 years and 0.7% in the third year, did so because of an unacceptable bleeding pattern. Women who dropped out for `other reasons' (3.5 and 2.0% in 2 years and the third year respectively) included those who wanted to become pregnant, those who were no longer in need of a contraceptive method, or those moving away from the study centre.

Of the total study population, 603 women contributed at least one reference period for analysis. If a woman had taken sex steroids or liver enzyme inducing drugs during the study, which was not allowed according to protocol, her data were (partly) excluded from the bleeding analysis. The reasons for discontinuation due to an unacceptable bleeding pattern are categorized and presented in Table IIIGo. The clinically most important bleeding indices, together with the numbers of women contributing, are presented in Table IVGo. It must be borne in mind that the high percentages reported for `infrequent bleeding' do not include those women having amenorrhoea. The bleeding categories mentioned in Table IVGo are distinctly different from each other, except for `prolonged bleeding' which may occur together with `infrequent bleeding' as well as with `frequent bleeding'. From reference period 9 onwards the study population is different. Only Budapest and Santiago contributed to reference periods 9–12. The declining incidences of amenorrhoea over the first eight reference periods subsequently show an increase at reference period 9 due to the different population involved.


View this table:
[in this window]
[in a new window]
 
Table III. Discontinuation in the first 2 years due to bleeding irregularities or amenorrhoea
 

View this table:
[in this window]
[in a new window]
 
Table IV. Bleeding pattern indices
 
Safety was monitored by recording of adverse experiences other than bleeding problems, including serious adverse experiences. Serious adverse experiences were those fulfilling the definition: `any experience that is either fatal or life-threatening, permanently disabling, requiring in-patient hospitalization or prolongs hospitalization, or is a congenital anomaly, cancer or an overdose'. Adverse experiences were recorded as those symptoms that either newly occurred or increased in frequency and/or severity, when compared to baseline.

The severity of serious adverse experiences as well as of adverse experiences was classified by the investigator as mild, moderate or severe. In Table VGo, an overall assessment is presented of adverse experiences during 3 years of treatment. Those adverse experiences that were in the investigator's opinion possibly, probably or definitely drug-related are presented as drug-related.


View this table:
[in this window]
[in a new window]
 
Table V. Overall assessment of adverse experiences
 
Of the 50 serious adverse experiences, the following six cases were judged by the investigators to be possibly related to treatment with Implanon®: uterine fibroid, ovarian cyst, intraductal papilloma with fibrocystic mastopathy, headache, transient ischaemic attack (TIA) and `raised heart rate with pressure over chest'. The woman with the TIA had no pathological findings, i.e. no signs of demyelination on magnetic resonance imaging and the computerized tomography scans were normal. The preliminary diagnosis TIA could not be substantiated but remained. The headache, which was mild, was accompanied by fever. It developed after 187 days on treatment and the woman was withdrawn from the study. The woman who developed `cardiac' symptoms first noticed these half a day after implant insertion. The symptoms worsened over the next 3 days and the implant was therefore removed. Symptoms remained for at least another week. The half-life of elimination of etonogestrel is ~24 h and a drug relationship therefore appears doubtful. Ovarian cysts were diagnosed by gynaecological examination and followed up by ultrasound. All were found to disappear spontaneously.

Adverse experiences that were reported by >5% of the volunteers are presented in Table VIGo. Reporting in Santiago (Chile) is substantially higher than in the other centres, due to a different method of recording. Hungary consisted of three centres, one of which was Budapest in which the study extension took place. Adverse experiences reported in the third year are shown separately. These are newly reported, i.e. first occurrences in the third year. There were no remarkable changes in physical and pelvic examination findings.


View this table:
[in this window]
[in a new window]
 
Table VI. Frequently reported (>5% of all volunteers) adverse experiences per region
 
At baseline the Pap smear had to be class I or II. During treatment and at implant removal changes were recorded. Pap class IV or V was not recorded at any time. In two women a conization was done, indicated by Pap smear findings. In other cases where Pap III was seen, this either resolved spontaneously or upon treatment with antibiotics.

At the start of the study, 36% of the women had a history of dysmenorrhoea. At the end of the study dysmenorrhoea had improved in 87% of the women who had dysmenorrhoea before they received the implant. In 4% this symptom was reported as a new occurrence or as worsening of existing dysmenorrhoea.

Acne, which is one of the most frequently reported adverse experiences, was also recorded at baseline (history). At the end of the study, women were asked about this adverse experience. In 12.8% of the women, acne had improved during treatment with Implanon®, whilst in 12.6% this symptom was new or had worsened.

The BP was measured frequently during the period of study. A systolic BP recording that was >140 mmHg and had increased >20 mmHg from baseline during at least two assessments or at the last assessment (single recording) was considered clinically significant. A clinically significant diastolic BP had to read >90 mmHg and increase >10 mmHg from baseline during at least two assessments or at last assessment. In 10 women clinically significant BP readings occurred; five (0.8%) had a clinically significant systolic reading and seven (1.1%) had a clinically significant diastolic reading during or at the end of the study. The mean systolic as well as diastolic BP showed a small decrease over time.

The body mass index (BMI) showed an increase from baseline >10% once or over several measurements in 20.2% of the women. There was a gradual mean increase in BMI over the full study period. The mean change was 0.80 kg/m2 (SD 1.59) which also included women who discontinued due to weight gain. The mean percent increase in BMI was 3.5%. In 15 (2.4%) women the weight gain was the reason given for discontinuing treatment.

Swelling, redness, pain and/or haematoma at implant site were rarely reported. A combination of symptoms may have occurred in an individual. In that case each symptom was counted separately and percentages shown indicate not only each woman but also each symptom reported. Incidences are shown in Table VIIGo. Insertion of the implant took on average 2.2 min (SD 2.1) with a minimum of 0.03 and a maximum of 10 min. In eight cases (1.3%), some complication related to the insertion occurred (e.g. tip of implant visible, some blood loss from injection site etc.). Average time needed for removal was 5.4 min (SD 5.4). The minimum time required was 0.33 and the maximum 45 min. Removal was complicated in 19 cases (3.0%), usually caused by too deep insertion.


View this table:
[in this window]
[in a new window]
 
Table VII. Incidence of symptoms at implant site
 
After removal of the implant, women were followed-up for 3 months. Of those who chose a non-hormonal method of contraception or no method at all, menses returned to normal in 90.9% within 3 months. This was not influenced by the length of time that the implant was in place. Of the post-treatment pregnancies that were reported, the estimated date of conception was within 90 days in 20 women out of 145 (13.8%) who used `no contraceptive method'. Some of those pregnant reported to have used condoms or rhythm method. The percentage of women using `no method', therefore, most probably was lower than stated. It is not known how many actively tried to become pregnant.


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
This is the first report on a single contraceptive implant with an effective life span of 3 years. The results show that Implanon® provides excellent contraceptive cover for the full period of 3 years, to women of a wide range of ages, weights and cultural backgrounds. There were no pregnancies in a total of more than 15 000 cycles of exposure, 2000 of which were in the third year of use. Total protection in studies of this size is rarely observed, even among the most effective methods, including tubal ligation. This fact allows us to forecast the new implant's high efficacy in the youngest, most fertile women, as well as in those who were overweight, for whom the dose delivered by Norplant® near the end of its effective life has been shown to be insufficient in some cases (Sivin, 1994Go, 1998Go).

In the absence of a control group, comparisons between methods have intrinsic limitations due to the multiplicity of factors that affect their performance. In the case of contraceptive implants, the only objective measure of clinical performance is the pregnancy rate, which has been shown to be affected by factors such as age and body weight (Sivin, 1988Go, 1994Go). With due consideration to these caveats, at the present time, the performance of Implanon® can only be compared with the performance of other contraceptive implants, using literature reports. This exercise should include Norplant® (six capsule levonorgestrel system), Norplant®-II (two rod levonorgestrel system), Uniplant (nomegestrol acetate) and Nestorone® implants. Literature reports of the latter two implant systems remains very limited, whereas there is a vast literature on Norplant®. Notwithstanding, only recent large studies deal with the present soft tubing formulation. Annual pregnancy rates during 3 years of use ranged from 0.0 to 0.4 per 100 women in a trial that encompassed 1530 users of Norplant® (Sivin, 1994Go). No pregnancies were reported in another trial encompassing 598 users of the six-capsule system (Norplant®) and 600 users of the two-rod system (Norplant®-II) over 3 years (Sivin et al., 1997Go). Most recently, results were reported of the six-capsule system in 511 women, in the United States. The annual pregnancy rate ranged from 0.0 to 0.3 per 100 women over 3 years which rose to 1.0 per 100 women over 5 years (Sivin et al., 1998Go). Therefore, as far as contraceptive efficacy is concerned, Implanon® is as good or better than the only marketed implant.

As for the bleeding pattern, this study confirms the well established concept that the use of continuous progestagen-only methods is associated with a disturbed pattern in a high proportion of users and that this is the main reason for discontinuation of the method, as is the case with Norplant® (Sivin et al., 1997Go).

The bleeding pattern reported here, shows a shift towards amenorrhoea and infrequent bleeding. Although the incidence of amenorrhoea was just below 20% during most of the time, it was the reason for discontinuation in only 1.7% of the women. An objective study of the bleeding cards indicated that the acceptance of an irregular bleeding pattern was greatly at variance. Some women accepted very little in the way of an irregular bleeding pattern, while others accepted `frequent irregular patterns' for long periods of time. Discontinuation rates, therefore, are in the first place a reflection of tolerance, rather than safety.

The percentages of discontinuation due to adverse experiences other than bleeding were 8.5 and 3.4 respectively in the first 2 years and in year 3, whereby weight increase was the most frequently reported reason for discontinuation (2.4%).

Although reporting of adverse experiences was substantially more frequent in Chile, the discontinuation rate was generally lower than in other centres. This most probably was the result of good counselling. The reporting of adverse experiences may be subject to methodological differences, which was also the case in the study reported here. In Chile, women were given ample opportunity to express their complaints freely, which were recorded each time they were expressed. In a study with Norplant® in the same clinic, a symptom was only recorded when it was reported in at least three follow-up visits. Moreover, there had to be an increase from the baseline situation. Using this method of assessment, headache occurred with an incidence of 32.6% (Diaz et al., 1979Go). In the present study the incidence of headache was 69.2% in Chile, which was in contrast to the average incidence of 6.5% in the other centres.

The low incidence of adverse experiences at the site of implant placement, together with the fast insertion and removal, stand out as practical advantages over multiple unit implants.

Results from this study show that Implanon® has excellent contraceptive action. The bleeding pattern generally was irregular whereby the main tendency was towards amenorrhoea and infrequent bleeding. Weight increase was a side effect that was less well tolerated. There was no untoward effect on BP as reflected in a small mean decrease in systolic as well as in diastolic BP. Implant insertion and removal were fast and uncomplicated in the vast majority (97%) of cases.


    Acknowledgments
 
Drs Gonnie van Osta and Ad Theeuwes are gratefully acknowledged for statistical support and for critically reading the manuscript. Financial support of the study was provided by NV Organon, Oss, The Netherlands.


    Notes
 
1 Study Group: Austria, Professor Dr J.Huber, I. UniversitaÐts Frauenklinik, Vienna; Belgium, Dr M.Vekemans/Dr A.Delvigne, HoÃpital Universitaire St Pierre, Brussels; Professor Dr J.Schwers, HoÃpital Erasme, Brussels; Chile, Dr H.B.Croxatto, Instituto Chileno de Medicina Reproductiva, Santiago; France, Professor P.Madelenat, Maternite HoÃpital Bichat, Paris; Dr D.Serfaty, HoÃpital Saint-Louis, Paris; Professor J.Colau, Centre Medico-Chirurgical Foch, Suresnes; Germany, Dr med E.Brandl/Dr med P.Tykal, General Practice, NoÐrdlingen; Dr med G.MaÐlzer, StaÐdt. Frauenklinik Leipzig; Dr med habil H.Stietzel, General Practice, Delitzsch; Professor Dr med T.Rabe, UniversitaÐts Frauenklinik Heidelberg; Dr med K.Hoffmann, General Practice, Karlsruhe; Dr D.Ullrich, Frauenklinik des Bezirkskrankenhauses, Dessau; Hungary, Professor L.KovaÂcs, Albert Szent-GyoÐrgyi Medical University, Szeged; Professor L.LampeÂ, University Medical School of Debrecen, Debrecen; Dr. J.Urbancsek, Semmelweis University, Budapest; Netherlands, Dr S.van der Linden, Ziekenhuis Eemland, Amersfoort; Dr T.Ypma, Scheperziekenhuis, Emmen; Dr R.Beerthuizen, Streekziekenhuis Koningin Beatrix, Winterswijk; Sweden, Dr I.OÐstlund, Regionsjukhuset, OÐrebro; UK, Professor J.R.Newton, Birmingham Maternity Hospital, Birmingham. Back

4 To whom correspondence should be addressed at: Clinical Development Department, NV Organon, PO Box 20, 5340 BH Oss, The Netherlands Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Belsey, E.M., Machin, D. and d'Arcangues, C. (1986) The analysis of vaginal bleeding patterns induced by fertility regulating methods. Contraception, 34, 253–260.[ISI][Medline]

Belsey, E.M. and Farley, T.M.M. (1988) The analysis of menstrual bleeding patterns: a review. Contraception, 38, 129–156.[ISI][Medline]

Corfman, Ph. (1995) Labeling guidance text for progestin-only oral contraceptives. Contraception, 52, 71–76.

Davies, G.C., Li, X.F., Newton, J.R. et al. (1993) Release characteristics, ovarian activity and menstrual bleeding pattern with a single contraceptive implant releasing 3-ketodesogestrel. Contraception, 47, 251–261.[ISI][Medline]

Diaz, S., Pavez, M., Robertson, D.N. et al. (1979) A three-year clinical trial with levonorgestrel silastic implants. Contraception, 19, 557–573.[ISI][Medline]

Mäkäräinen, L., van Beek, A., Tuomivaara, L. et al. (1998) Ovarian function during the use of a single contraceptive implant: Implanon® compared with Norplant®. Fertil Steril, 69, 714–721.[ISI][Medline]

Metropolitan Height and Weight Tables (1983) Statistical Bulletin, Jan–Jun, 3–10.

Rodriguez, G., Faundes-Latham, A. and Atkinson, L.E. (1976) An approach to the analysis of menstrual patterns in the critical evaluation of contraceptives. Stud. Fam. Plan., 7, 42–51.[ISI]

Sivin, I. (1988) International experience with Norplant® and Norplant®-2 contraceptives. Stud. Fam. Plan., 19, 81–94.[ISI]

Sivin, I. (1994) Contraception with Norplant implants. Hum. Reprod., 9, 1818–1826.[Abstract]

Sivin, I., Viegas, O., Campodonico, I. et al. (1997) Clinical performance of a new two-rod levonorgestrel contraceptive implant: a three-year randomized study with Norplant® implants as controls. Contraception, 55, 73–80.[ISI][Medline]

Sivin, I., Mishell, D.R., Darney, Ph. et al. (1998) Levonorgestrel capsule implants in the United States: a 5-year study. Obstet. Gynecol., 92, 337–344.[Abstract/Free Full Text]

World Health Organization (1994) WHO Adverse Reaction Dictionary. WHO Collaborating Centre for International Drug Monitoring, Uppsala, Sweden.

Submitted on May 7, 1998; accepted on November 30, 1998.