- Email: [email protected]
Evaluation of a 1-year levonorgestrel-releasing contraceptive implant: side effects, release rates, and biodegradability*
FERTILITY AND STERILITY
Vol. 58, No.1, July 1992
Copyright It! 1992 The American Fertility Society
Printed on acid-free paper in US.A.
Evaluation of a l-year levonorgestrel-releasing contraceptive implant: side effects, release rates, and biodegradability*
Philip D. Darney, M.D.t:I: Cynthia M. Klaisle, M.S.N.t Scott E. Monroe, M.D.t
C. Edgar Cook, Ph.D.§ Nancy R. Phillips, Ph.D. II Anton Schindler, Ph.D.§
University of California, San Francisco, California, and Research Triangle Institute, Research Triangle Park, North Carolina
Objective: To determine the release rates, effects on ovulation, and side effects of two lengths of a biodegradable, subdermal contraceptive implant containing levonorgestrel in a caprolactone capsule. Design: Phase II randomized clinical trial. Setting: Public family planning clinic at an urban general hospital. Participants: Forty-eight healthy, parous, ovulating volunteers. Intervention: Subjects were randomly assigned either a 2.5- or a 4.0-cm contraceptive capsule that was worn under the skin of the upper arm for 1 year if not removed earlier for other reasons. Main Outcome Measures: Serum concentrations of levonorgestrel, progesterone, estradiol, and lipoproteins were measured as were metabolic parameters. Vaginal bleeding and other side effects were recorded. After implant removal, remaining levonorgestrel, capsule viscosity, and molecular weight were measured. Results: The 4-cm implant provided serum concentrations of levonorgestrel ranging from 0.65 ng/mL shortly after insertion to 0.20 ng/mL at 12 months, but the 2.5-cm implant resulted in levels too low for contraception. The 4-cm implant suppressed ovulation in approximately 80% of cycles over 1 year of use, but the 2.5-cm implant failed to suppress ovulation. The implants were rapidly and easily inserted and removed. They retained structural integrity through 1 year of use. Of 48 subjects, 32 had abnormal bleeding patterns; the mean number of days of bleeding per month was 7 with 23 days between episodes. Women using capronor maintained normal metabolic parameters except that low-density lipoprotein decreased slightly. Conclusions: The 4.0-cm implant is a promising contraceptive. There were no important metabolic effects, but there were bothersome minor side effects typical of progestin-only contraception. Fertil Steril 1992;58:137-43 Key Words: Contraception, implant, levonorgestrel, biodegradable
Sustained release implant contraceptives provide a new approach to birth control (1). More than
Received October 30, 1991; revised and accepted March 3, 1992.
* Supported by contracts NOI-HD-6-2829 and NOI-HD-82916 from the National Institutes of Health, Bethesda, Maryland. t Department of Obstetrics, Gynecology and Reproductive Sciences, University of California. :\: Reprint requests: Philip D. Darney, M.D., Department of Obstetrics, Gynecology and Reproductive Sciences, San Francisco General Hospital, 1001 Potrero Avenue, Ward 6D, San Francisco, California 94110. § Research Triangle Institute. II Cardiovascular Research Institute, University of California. Vol. 58, No.1, July 1992
1 X 106 women around the world currently use the contraceptive implant, Norplant (Wyeth-Ayerst Laboratories, Philadelphia, PA), which was approved for use on December 10, 1990. More than 55,000 women have participated in Norplant clinical trials and have demonstrated that this approach to contraception is effective, safe, and acceptable (25). The utility and cost of implant contraception could be reduced by a simpler system requiring only a single implant that is easier to insert and, if necessary, remove than are the six silastic implants of Norplant. In addition, an implant contraceptive that prevents conception for an interval shorter than Darney et al.
Capronor contraceptive
137
Norplant's 5 years could be an attractive and less expensive alternative for women who may want pregnancy sooner. The following account describes a clinical evaluation of a single implant contraceptive that can be removed on request but is intended to provide 1 year of contraception and then degrade spontaneously. The implant consists of a poly(e-caprolactone) capsule, measuring either 2.5 or 4.0 cm in length, that is filled with 12.0 mg or 21.6 mg, respectively, of levonorgestrel in an oleate base. These capsules release levonorgestrel at a rate approximately 10 times faster than silastic so that a smaller surface area is required to maintain contraceptive levels (6). The capsules slowly degrade to e-hydroxycaproic acid and, finally, CO 2 and H 20 but retain structural integrity during the period of levonorgestrel release so that removal, if required, is easily accomplished. MATERIALS AND METHODS
Forty-eight women between the ages of 18 and 35 years who had regular menstrual cycles (25 to 35 days), were sexually active (intercourse more than once per month), had delivered at least one live child, and wanted continuous contraception for 12 months received capronor implants. They were in good health as determined by medical history, complete physical examination, and laboratory determinations of blood count, liver, thyroid, renal, and adrenal functions. Each subject signed an informed consent approved by the University of California, San Francisco, Committee on Human Research. Subjects had not used hormonal contraception in the month preceding enrollment. The original study design hypothesized that there would be no difference in frequency of ovulation between the two implants. Sixty subjects were assigned to either a 4.0- or a 2.5-cm implant using a table of random numbers. Enrollment in the 2.5-cm group was limited to 16 subjects after serum levonorgestrellevels among users of this device declined to levels considered inadequate for contraception. Thirty-two women received 4.0-cm implants; the last 16 were not randomized. Insertions and removals were performed by techniques previously described (5, 7). All subjects returned monthly for review of symptoms and menstrual calendars on which were recorded any blood loss without distinction between bleeding or spotting. Serum samples were obtained for levonorgestrel, progesterone (P), estradiol (E 2), and human chorionic gonadotropin determinations. A complete physical exam and laboratory determinations for blood count, liver, thyroid, renal, and adrenal func138
Darney et al.
Capronor contraceptive
tions were performed at the 2, 5, 8, and 11-month visits. A serum sample was obtained after a 12-hour fast at months 0, 3, 9, and 13 for lipoprotein analysis. A questionnaire assessed user perceptions, experiences' and acceptance of this new contraceptive before and after insertion and removal of capronor. Fifteen women who elected to participate in intensive studies of ovulatory patterns underwent venipuncture three times per week during months 0, 1,6, 12, and after removal until the onset of menses and one time per week during months 2, 3, 4, 5, 7, 8,9, 10, and 11. Four women who received the 2.5cm implant and 11 who received the 4.0-cm implant were included in this subset. Each serum specimen was assayed for levonorgestrel, P, and E2 levels. The complete and differential blood count, thyroid function studies, cortisol (F), and blood chemistries were performed by Smith-Kline Bioscience Laboratory (San Francisco, CA) using standard techniques. The concentrations of cholesterol and triglycerides were measured in the whole serum in the three major lipoprotein fractions: very low density (VLDL), low-density (LDL), and high-density (HDL), and in the HDL3 subfraction. The lipoprotein fractions were separated by preparative ultracentrifugation (8); the lipid determinations were referenced to a calibrator serum pool supplied by the Clinical Chemistry Standardization Section, Centers for Disease Control (Atlanta, GA). The lipid level in the HDL2 sub fraction was computed as the difference between the levels in the HDL and HDL3 fractions. These measurements were made on the fresh serum at the Cardiovascular Research Institute at the University of California (San Francisco, CA). Approximately 10 mL of blood was collected in serum separator tubes for levonorgestrel, P, and E2 determinations. Reagents for the levonorgestrel radioimmunoassay (RIA) were obtained from the World Health Organization (Geneva, Switzerland). The minimal detectable level of levonorgestrel that could be quantified was approximately 0.02 ng/mL. At the highest levels found in the serum of women receiving the 4.0-cm capronor implant, levonorgestrel did not cross-react with P in the assay; consequently, serum samples were not subjected to either extraction or chromatography before quantification by RIA. The capsules retrieved from subjects at the conclusion of their participation in the trial were stored frozen until the study ended. Six of these capsules were analyzed to determine release rates in vitro before the measurement of the residuallevonorgestrel contents of all capsules. Another batch of six capsules was selected to characterize the capsule Fertility and Sterility
walls with respect to changes in crystallinity and molecular weight (MW). The amount of levonorgestrel released from each capsule was calculated as the difference between the analytically determined residual drug contents and the average initial drug content of 2.5- and 4.0-cm capsules.
100
fB
§ il: w "0
a: w
CD ~
:;)
z
RESULTS 20
10
25
30
35
40
Insertion and Removal of Implants DAYS per EPISODE
Insertions were accomplished in a mean time of 1.5 minutes. There were no infections, expulsions, bleeding, or adverse reactions related to the insertion procedure. Capronor insertions were well tolerated. Perceived insertion discomfort matched that expected for 13% (n = 6), was less than expected for 73% (n = 35), and more than expected for only 6% (n = 3) ofthe subjects. After insertion, 92% (n = 44) of the subjects returned immediately to their daily routine. Mean time for removal was 3 minutes. Although 50% (n = 21) of subjects expressed some fear of pain during removal, 63% (n = 30) stated that the actual procedure was without pain. Removal discomfort matched subjects' expectations for 33% (n = 14) of subjects, was less than expected for 48% (n = 20), and was more than expected for 14% (n = 6).
Figure 1 Number of bleeding episodes by number of days per episode during 422 months of capronor subdermal contraceptive implant use by 48 women.
Ovarian Cysts and Abdominal Pain
Three subjects complained of lower abdominal pain, and five asymptomatic women were found to have ovarian cysts on pelvic examination. These were confirmed by ultrasonography. Seven of the eight ovarian cysts resolved spontaneously, and one 3-cm cyst was drained at the time of laparoscopy, performed to determine the origin of persistent pelvic pain. Three other subjects presented with abdominal pain. One had a hydrosalpinx that was laparoscopically diagnosed, and the other two had normal pelvic examinations.
Bleeding Patterns
Local Effects
Thirty-two (67%) of the 48 subjects reported irregular or prolonged bleeding, whereas seven subjects complained of amenorrhea during use of capronor. Analysis of bleeding patterns obtained from menstrual calendars showed that of 13,829 total days of capronor use, 2,909 (21 %) were marked by vaginal bleeding. The mean number of days of bleeding per month per subject was 7 with an average interval of 23 days between episodes. Two thirds of bleeding episodes were of 4 to 8 days duration, but 6% of episodes were more than 15 days in length (Fig. 1). Absence of bleeding was reported less frequently than prolonged bleeding. There were only 36 instances of amenorrhea lasting >40 days in 35 woman-years of capronor use (Fig. 2). Correlation of P levels with individual bleeding histories demonstrated that women who ovulated regularly also had regular menses. Subjects experiencing suppression of ovulation were more likely to have irregular bleeding and were also more likely to be amenorrheic than those who ovulated regularly. The lower the concentrations of E2 the more likely a subject was to report bleeding.
Eight subjects experienced downward migration of the implant from the upper arm insertion site during the 1st through 5th months of use. Downward migration ranged from 1 to 6 cm from the original placement site and did not prompt complaints or make removal of the implants more difficult. Twenty subjects (41 %) reported pruritis directly over the implant. Sixteen of the subjects had a single
Vol. 58, No.1, July 1992
40
10
50
60
70
80
90
100
DAYS per INTERVAL
Figure 2 Number of intervals between bleeding episodes by number of days per interval for 422 months of use of capronor subdermal implant contraceptive by 48 women. (One subject was amenorrheic for 206 days.) Darney et al.
Capronor contraceptive
139
episode soon after insertion, whereas four complained of intermittent pruritis throughout implant use. One of the subjects with a single episode of pruritis and two subjects with intermittent pruritis were treated with a topical corticosteroid with resolution of symptoms. Nine subjects complained of an erythematous rash within several days of insertion. Six of these subjects received no treatment, and three applied a topical corticosteroid. All experienced complete resolution of symptoms within 3 to 5 days.
(Bars show standard error of the means.)
i
J5
-'
~
~
2S
.L
IS
>
~
05
Pregnancies
10
Three pregnancies occurred among women using capronor. The first was in a 2.5-cm implant user during her 8th month of use when levonorgestrel levels ranged from 0.08 to 0.14 ng/mL and the subject demonstrated bleeding patterns consistent with ovulation. The second pregnancy occurred 9 months after the insertion of a 4.0-cm implant when levonorgestrel ranged from 0.08 to 0.14 ng/mL. The third pregnancy occurred after 11 months of a 4.0-cm implant use at a levonorgestrel of 0.12 ng/mL. The Pearl Index for the 2.5-cm implant was 18.2 per 100 woman-years; for the 4.0-cm implant it was 4.5. Continuation Rates
The continuation rate for all users of capronor (excluding 6 who terminated use early because the study ended) was 43% at 12 months. Concern about decreased effectiveness when informed about low levonorgestrellevels prompted 6 (14%) subjects to discontinue use. Six subjects (14%) discontinued for reasons not related to the method: 3 for failure to keep appointments and 3 more because they were moving out of town. Twelve women (29%) discontinued capronor use for method-related reasons, with prolonged menses (n = 4) and pregnancy (n = 3) being the most common. When asked about their satisfaction with the method, 83% (35) stated they were satisfied with the implants, whereas only 10% stated they were dissatisfied. When asked whether they would use capronor again, 70% said yes. Serum Levels of Levonorgestrel
Serum levonorgestrel concentrations among users of the two capronor implants ranged from >0.8 ng/mL to <0.1 ng/mL. Determinants of levonorgestrel serum concentrations among subjects in the comparison of 2.5-cm and 4.0-cm capronor implants were the length of the implant and duration of its use (Fig. 3). Mean levonorgestrel values for women 140
Darney et al.
Capronor contraceptive
12
I.
MONnlS
Figure 3 Levonorgestrel concentrations (ng/mL) in women using 2.5-cm (n = 16) or 4.0-cm (n = 32) capronor subdermal contraceptive implants.
using the 4.0-cm implant were significantly higher than in women using the 2.5-cm device, although a more rapid decline in serum concentration during the 1st month was observed among users of the 4.0cm implant. By the 8th month of use, there were too few 2.5-cm users to make valid comparisons between the 4.0- and the 2.5-cm groups. Serum levonorgestrel concentrations in the 4.0-cm group consistently averaged >0.20 ng/mL for 1 year of use. In the 2.5-cm group, however, mean levels fell <0.20 ng/mL, a level judged inadequate for contraception by the 4th month of use. The nine women who elected insertion of two 2.5cm implants received the second capsule after 110 to 275 days of use of the single 2.5-cm implant. Insertion of a second 2.5-cm implant for these women delayed the decline of their mean levonorgestrel serum concentrations but did not sustain levonorgestrel levels >0.20 ng/mL for longer than women wearing the 4.0-cm implant. Effects on Ovulation
Eleven of 13 subjects who had frequent blood sampling to study ovulation received the 4.0-cm capronor implant (Table 1). In this group, only 19% of 104 cycles studied were ovulatory, and no ovulations occurred until after the 12th week of capronor use. By this time, levonorgestrellevels in this group had fallen to a mean of <0.35 ng/mL. Figure 4 shows that ovulation rarely occurred when levonorgestrel concentrations averaged >0.35 ng/mL. Three users of the 4.0-cm implant did not ovulate even when levonorgestrel concentrations fell below 0.25 ng/mL. For the two women who were studied Fertility and Sterility
Table 1 Cycles Monitored* and Number Ovulatoryt by Type of Capronor Device Cycle
2.5-cm
Ovulatory Nonovulatory Total cycles
Two 2.5-cm
4.0-cm
4 (67) 2 (33) 6 (100)
20 (19) 84 (81) 104 (100)
21 (95)+ 1 (5)
22 (100)
and the 2.5-cm implants. Even women who ovulated infrequently maintained E2 concentrations within the limits of normal. Levels of E2 were also within normal limits during the month immediately after insertion of capronor when levonorgestrel levels were at their peak (0.5 to 0.8 ng/mL).
* One or more P determination per week. t Progesterone :2: 3.0 ng/mL. Values in parentheses are percents.
Release Rates
+
The mean levonorgestrel release rate for subjects who used the 4.0-cm capronor implant was 33.0 ± 10.34 ~g/d with a range from a low of 20 ~g to a high of 65 ~g. Figure 3 shows that the average serum levonorgestrellevels of 32 women using the 4.0-cm capsule decreased from an initial value of 0.50 ng/mL to a final value of 0.20 ng/mL with a mean value of 0.23 ng/mL at 10 months. The observed decrease in the serum level of levonorgestrel was a consequence of a steadily decreasing rate of release from the capsules. In Figure 5, the amounts of levonorgestrel released from 4.0-cm capsules are plotted against the time the capsules remained in place. The average initial daily release rate was 51.0 ~g/d, which compared well with reported average in vitro release rates for 4.0-cm capsules of 54.0 ± 3.1 ~g/d (6). After
while they wore the 2.5-cm implant, all but 1 of 22 cycles were ovulatory. Of 30 ovulatory cycles occurring among the 11 frequently sampled users of the 4.0-cm implant, only 1 resulted in pregnancy. Both of the pregnancies that occurred during a total of 276 woman-months of 4.0-cm implant device use occurred in subjects who had levonorgestrel concentrations far below the group mean for their duration of use. Effects on E2
Mean E2 concentrations were within the limits for ovulating women among users of both the 4.022.5
20
o
o
o o
o
o o 00
o
o o
o
0
o
o
o o
o
o o
o
0
o 2.5
o 0
o
o
o
0 0 0
o
0
o 00. "oBtllfoi i.z, J.. Ol+-____~o~--~~~~~~==~~~~ o
o
__ __
" \ 0 00
.1
.2
.3
.4
000· 0
~~ .5
o 0
0
00
0
~r_~_ _~--~--;_----~~~--, .8
.7
.8
.9
MEAN LEVONORGESTREL LEVEL I CYCLE (NGIM.)
Figure 4 Mean serum levonorgestrel concentrations (ng/mL) per cycle and maximum serum P (ng/mL) per cycle in 11 women using 4.0-cm capronor subdermal contraceptive implant and having blood sampled at least once weekly. Vol. 58, No.1, July 1992
Darney et al.
Capronor contraceptive
141
,. ~
•
12
•
10
"
C
w
m a: irl
•
8
ii 0
0
/
/. 50
of 46.8% to a final average value of 60.0%, showing maintenance of their structural integrity .
•
Effects on Serum Lipoproteins
•
/ • 100
150
200
250
300
400
350
DAYS IMPLANTED
Figure 5 Amounts of levonorgestrel (mg) released from 4.0cm capronor subdermal contraceptive implants (n = 25) during various periods of implantation in women.
20 days, the average release rate of the capsules was 28.8 ± 2.4 ILgjd per capsule, approximately half the initial release rate. After 365 days, the calculated release rate decreased to 8.6 ILgjd. Degradation of Implanted Capsules
Six 4.0-cm capsules were analyzed for MWsand crystallinity contents to assess the degree to which they had degraded during use. All of the selected capsules had been implanted for approximately 1 year (344 to 373 days). Intrinsic viscosities in toluene were determined for all 32 of the retrieved 4-cm capronor capsules revealing the expected first order decline during implantation. The half-life time for the MW was found to be 10.6 months-in good agreement with an observed in vitro value of 10.0 months. The crystallinity of capsules implanted for approximately 1 year increased slightly from an initial value
As previously reported, there was no evidence that the use of capronor levonorgestrel implants caused adverse effects in blood lipids (Table 2). Although the HDL subfraction (HDL 2 cholesterol) tended to decrease slightly after capronor placement, the observed decrease was both statistically insignificant and proportionally smaller than the concurrent decrease in LDL cholesterol. The statistically significant decreases in LDL cholesterol and total triglycerides after placement of capronor were comparable in magnitude with those previously reported in users of silastic levonorgestrel implants (2, 9-11). One month after removal of capronor, LDL cholesterol and total triglycerides remained below preinsertion levels. Effects on Other Laboratory Values
There were no significant changes in serum F or blood chemistry levels over preinsertion values. One subject developed abnormalities in thyroid function and was discovered to have thyroiditis causing hypothyroidism. Five subjects developed mild anemia during study participation and one during the month after removal. None of the cases of anemia was associated with a history of heavy, prolonged, or irregular vaginal bleeding. Four of these five subjects were participants in the study of ovulatory patterns requiring 60 mL of blood sampling per week for up to 6 consecutive weeks. DISCUSSION
This single implant contraceptive system was easy to place under the skin and simple to remove through
Table 2 Differences Between Repeated Lipid Measurements Taken Before, During, and After Use of Capronor Subdermal Levonorgestrel Contraceptive Implants 3-mo postimplant versus baseline (n = 44)
Measured lipid Triglycerides (mg/dL) In whole serum In VLDL fraction Cholesterol (mg/dL) In whole serum In LDL fraction In HDL fraction In HDL2 subfraction
*t 142
=
Postremoval versus baseline (n = 23)
Mean difference
t*
Mean difference
t
Mean difference
t
-16.8 -6.2
-4.4t -2.3
-1.7 -0.2
-0.5 -0.1
-12.9 -2.5
-4.0t -1.1
-13.8 -8.0 -1.5 -1.3
-3.9t -2.7 -1.1 -0.83
-11.8 -6.0 -1.9 -0.4
-3.It -1.8 -1.1 -0.2
-18.6 -7.5 -0.8 +3.9
-5.7t -2.8 -0.5 +2.7t
difference/standard error of difference. Darney et al.
9-mo postimplant versus 3 mo (n = 19)
Capronor contraceptive
t Contrast statistically significant at P ,,; 0.05.
Fertility and Sterility
a 2 to 3-mm incision using finger pressure without instruments. The implants were difficult to grasp with mosquito clamps, so finger pressure was used to direct the implant tip into the incision from which it could easily be expressed. Experience with digital extrusion of capronor prompted our use of this technique with Norplant. Digital extrusion allowed for a smaller skin incision and less trauma and pain with decreased risk of implant fracture during removal of N orplant as well as capronor, but the latter implant was much faster and easier to remove (3minute mean for capronor versus our 22-minute mean for Norplant). In all but 2 of 32 subjects, the 4.0-cm implant maintained levels of levonorgestrel over the course of 1 year that were judged adequate to provide effective contraception (>0.25 ng/mL), but the 2.5cm implant did not maintain adequate serum levonorgestrel levels. The three subjects who became pregnant during their participation in the study had serum levonorgestrel levels considerably below the mean values for their groups. Compared with levonorgestrel-releasing silastic implants (N orplant), serum levonorgestrel levels were initially high and then declined more rapidly so that after 12 months, mean levels were generally too low to provide effective contraception, confirming capronor's I-year limit. A minority of subjects reported bothersome minor side effects typical for women using hormonal contraception such as weight gain or loss, acne, mastalgia, and mood changes, in order of frequency. Only aberrations of menstrual pattern were clearly associated with varying serum levonorgestrel levels and with ovulation. All women maintained E2 levels within normal limits, an important consideration in hormonal contraceptives that do not replace estrogen (12). Implants left in place for up to 1 year maintained structural integrity sufficient to permit easy removal. Because none ofthe implants were allowed to remain in place beyond 1 year, we cannot specify the time after insertion at which the implants completely degrade or what the levonorgestrel content of such implants might be, but release rates were low after 1 year. Analysis of the retrieved implants revealed a small increase in their crystallinity and a first order decrease of their MWs. The implants remained strong enough, however, for easy removal at 1 year. A symptom clearly related to capronor use was local irritation ofthe skin over the implant. It seems likely that a residue of manufacture that remained on the implants caused this irritation. It was selflimited and mild, but its occurrence suggests that Vol. 58, No.1, July 1992
after manufacture a thorough washing of implants is required. If capronor is to provide a useful contraceptive option, the implant must maintain serum levonorgestrel levels above approximately 0.25 ng/mL throughout a year of use. The 2.5-cm implant does not do so and will not be further studied. The 4.0cm implant can probably be altered to maintain this serum levonorgestrellevel by eliminating the oleate excipient to increase slightly the levonorgestrel release rate. The ease of insertion and removal of this I-year, single implant makes it a promising possible addition to the sustained-release progestin contraceptive systems introduced by Norplant. Its biodegradability is another attractive characteristic because removal may not be necessary. The caproic acid capsule may also be a satisfactory vehicle for other contraceptive progestins. REFERENCES 1. Segal S. A new delivery system for contraceptive steroids. Am J Obstet Gynecol 1987;157:1090-2. 2. Roy S, Mishell D, Robertson DN, Kraus RM, Lacarra M, Duda MJ. Long-term reversible contraception with levonorgestrel-releasing silastic rods. Am J Obstet Gynecol 1984;148:1006-13. 3. Diaz S, Pavez M, Miranda P, Robertson DN, Sivin I, Croxatto HE. A five-year clinical trial oflevonorgestrel silastic implants (NorplaneM ). Contraception 1982;25:447-58. 4. Sivin I. International experience with Norplant and Norplant2 contraceptives. Stud Fam Plann 1988;19:81-94. 5. Darney PD, Klaisle CM, Tanner ST, Alvarado AM. Sustained release contraceptives. Curr Probl Obstet Gynecol Fertil 1990;13:87-125. 6. Pitt CG, Gratzl MM, Jeffcoat RA, Zwidinger R, Schindler A. Sustained drug delivery system. II. Factors affecting rates of release from poly (E-caprolactone) and related biodegradable polyesters. J Pharm Sci 1979;68:1534-8. 7. Darney PD, Monroe SE, Klaisle CM, Alvarado AM. Clinical evaluation of the Capronor contraceptive implant: preliminary report. Am J Obstet Gynecol 1989;160:1292-5. 8. Myers LH, Phillips NR, Havel RJ. Mathematical evaluation of methods for estimation of the concentration of the major lipid components of human serum lipoproteins. J Lab Clin Med 1976;88:491-505. 9. Shaaban M, Sawaan I, Elwan S, Darwish HA. Effect of subdermallevonorgestrel contraceptive implants, Norplant, on serum lipids. Contraception 1984;35:413-9. 10. Affandi B, Suherman S, Djajalelana J, Hadisputra W, Moeloek FA. Serum lipids in Norplant implant users: a cross sectional study. Contraception 1987;36:429-34. 11. Viegas 0, Singh K, Liew D, Singh P, Ratnam SS. The effects of Norplant on clinical chemistry in Singaporean acceptors after one year of use: metabolic changes. Contraception 1988;38:79-89. 12. Ory SJ, Hammond CB, Yancy SG, Hendren RW, Pitt LG. The effect of a biodegradable contraceptive capsule (Capronor) containing levonorgestrel on gonadotropin, estrogen, and progesterone levels. Am J Obstet Gynecol 1983;145: 600-4. Darney et al.
Capronor contraceptive
143
Vol. 58, No.1, July 1992
Copyright It! 1992 The American Fertility Society
Printed on acid-free paper in US.A.
Evaluation of a l-year levonorgestrel-releasing contraceptive implant: side effects, release rates, and biodegradability*
Philip D. Darney, M.D.t:I: Cynthia M. Klaisle, M.S.N.t Scott E. Monroe, M.D.t
C. Edgar Cook, Ph.D.§ Nancy R. Phillips, Ph.D. II Anton Schindler, Ph.D.§
University of California, San Francisco, California, and Research Triangle Institute, Research Triangle Park, North Carolina
Objective: To determine the release rates, effects on ovulation, and side effects of two lengths of a biodegradable, subdermal contraceptive implant containing levonorgestrel in a caprolactone capsule. Design: Phase II randomized clinical trial. Setting: Public family planning clinic at an urban general hospital. Participants: Forty-eight healthy, parous, ovulating volunteers. Intervention: Subjects were randomly assigned either a 2.5- or a 4.0-cm contraceptive capsule that was worn under the skin of the upper arm for 1 year if not removed earlier for other reasons. Main Outcome Measures: Serum concentrations of levonorgestrel, progesterone, estradiol, and lipoproteins were measured as were metabolic parameters. Vaginal bleeding and other side effects were recorded. After implant removal, remaining levonorgestrel, capsule viscosity, and molecular weight were measured. Results: The 4-cm implant provided serum concentrations of levonorgestrel ranging from 0.65 ng/mL shortly after insertion to 0.20 ng/mL at 12 months, but the 2.5-cm implant resulted in levels too low for contraception. The 4-cm implant suppressed ovulation in approximately 80% of cycles over 1 year of use, but the 2.5-cm implant failed to suppress ovulation. The implants were rapidly and easily inserted and removed. They retained structural integrity through 1 year of use. Of 48 subjects, 32 had abnormal bleeding patterns; the mean number of days of bleeding per month was 7 with 23 days between episodes. Women using capronor maintained normal metabolic parameters except that low-density lipoprotein decreased slightly. Conclusions: The 4.0-cm implant is a promising contraceptive. There were no important metabolic effects, but there were bothersome minor side effects typical of progestin-only contraception. Fertil Steril 1992;58:137-43 Key Words: Contraception, implant, levonorgestrel, biodegradable
Sustained release implant contraceptives provide a new approach to birth control (1). More than
Received October 30, 1991; revised and accepted March 3, 1992.
* Supported by contracts NOI-HD-6-2829 and NOI-HD-82916 from the National Institutes of Health, Bethesda, Maryland. t Department of Obstetrics, Gynecology and Reproductive Sciences, University of California. :\: Reprint requests: Philip D. Darney, M.D., Department of Obstetrics, Gynecology and Reproductive Sciences, San Francisco General Hospital, 1001 Potrero Avenue, Ward 6D, San Francisco, California 94110. § Research Triangle Institute. II Cardiovascular Research Institute, University of California. Vol. 58, No.1, July 1992
1 X 106 women around the world currently use the contraceptive implant, Norplant (Wyeth-Ayerst Laboratories, Philadelphia, PA), which was approved for use on December 10, 1990. More than 55,000 women have participated in Norplant clinical trials and have demonstrated that this approach to contraception is effective, safe, and acceptable (25). The utility and cost of implant contraception could be reduced by a simpler system requiring only a single implant that is easier to insert and, if necessary, remove than are the six silastic implants of Norplant. In addition, an implant contraceptive that prevents conception for an interval shorter than Darney et al.
Capronor contraceptive
137
Norplant's 5 years could be an attractive and less expensive alternative for women who may want pregnancy sooner. The following account describes a clinical evaluation of a single implant contraceptive that can be removed on request but is intended to provide 1 year of contraception and then degrade spontaneously. The implant consists of a poly(e-caprolactone) capsule, measuring either 2.5 or 4.0 cm in length, that is filled with 12.0 mg or 21.6 mg, respectively, of levonorgestrel in an oleate base. These capsules release levonorgestrel at a rate approximately 10 times faster than silastic so that a smaller surface area is required to maintain contraceptive levels (6). The capsules slowly degrade to e-hydroxycaproic acid and, finally, CO 2 and H 20 but retain structural integrity during the period of levonorgestrel release so that removal, if required, is easily accomplished. MATERIALS AND METHODS
Forty-eight women between the ages of 18 and 35 years who had regular menstrual cycles (25 to 35 days), were sexually active (intercourse more than once per month), had delivered at least one live child, and wanted continuous contraception for 12 months received capronor implants. They were in good health as determined by medical history, complete physical examination, and laboratory determinations of blood count, liver, thyroid, renal, and adrenal functions. Each subject signed an informed consent approved by the University of California, San Francisco, Committee on Human Research. Subjects had not used hormonal contraception in the month preceding enrollment. The original study design hypothesized that there would be no difference in frequency of ovulation between the two implants. Sixty subjects were assigned to either a 4.0- or a 2.5-cm implant using a table of random numbers. Enrollment in the 2.5-cm group was limited to 16 subjects after serum levonorgestrellevels among users of this device declined to levels considered inadequate for contraception. Thirty-two women received 4.0-cm implants; the last 16 were not randomized. Insertions and removals were performed by techniques previously described (5, 7). All subjects returned monthly for review of symptoms and menstrual calendars on which were recorded any blood loss without distinction between bleeding or spotting. Serum samples were obtained for levonorgestrel, progesterone (P), estradiol (E 2), and human chorionic gonadotropin determinations. A complete physical exam and laboratory determinations for blood count, liver, thyroid, renal, and adrenal func138
Darney et al.
Capronor contraceptive
tions were performed at the 2, 5, 8, and 11-month visits. A serum sample was obtained after a 12-hour fast at months 0, 3, 9, and 13 for lipoprotein analysis. A questionnaire assessed user perceptions, experiences' and acceptance of this new contraceptive before and after insertion and removal of capronor. Fifteen women who elected to participate in intensive studies of ovulatory patterns underwent venipuncture three times per week during months 0, 1,6, 12, and after removal until the onset of menses and one time per week during months 2, 3, 4, 5, 7, 8,9, 10, and 11. Four women who received the 2.5cm implant and 11 who received the 4.0-cm implant were included in this subset. Each serum specimen was assayed for levonorgestrel, P, and E2 levels. The complete and differential blood count, thyroid function studies, cortisol (F), and blood chemistries were performed by Smith-Kline Bioscience Laboratory (San Francisco, CA) using standard techniques. The concentrations of cholesterol and triglycerides were measured in the whole serum in the three major lipoprotein fractions: very low density (VLDL), low-density (LDL), and high-density (HDL), and in the HDL3 subfraction. The lipoprotein fractions were separated by preparative ultracentrifugation (8); the lipid determinations were referenced to a calibrator serum pool supplied by the Clinical Chemistry Standardization Section, Centers for Disease Control (Atlanta, GA). The lipid level in the HDL2 sub fraction was computed as the difference between the levels in the HDL and HDL3 fractions. These measurements were made on the fresh serum at the Cardiovascular Research Institute at the University of California (San Francisco, CA). Approximately 10 mL of blood was collected in serum separator tubes for levonorgestrel, P, and E2 determinations. Reagents for the levonorgestrel radioimmunoassay (RIA) were obtained from the World Health Organization (Geneva, Switzerland). The minimal detectable level of levonorgestrel that could be quantified was approximately 0.02 ng/mL. At the highest levels found in the serum of women receiving the 4.0-cm capronor implant, levonorgestrel did not cross-react with P in the assay; consequently, serum samples were not subjected to either extraction or chromatography before quantification by RIA. The capsules retrieved from subjects at the conclusion of their participation in the trial were stored frozen until the study ended. Six of these capsules were analyzed to determine release rates in vitro before the measurement of the residuallevonorgestrel contents of all capsules. Another batch of six capsules was selected to characterize the capsule Fertility and Sterility
walls with respect to changes in crystallinity and molecular weight (MW). The amount of levonorgestrel released from each capsule was calculated as the difference between the analytically determined residual drug contents and the average initial drug content of 2.5- and 4.0-cm capsules.
100
fB
§ il: w "0
a: w
CD ~
:;)
z
RESULTS 20
10
25
30
35
40
Insertion and Removal of Implants DAYS per EPISODE
Insertions were accomplished in a mean time of 1.5 minutes. There were no infections, expulsions, bleeding, or adverse reactions related to the insertion procedure. Capronor insertions were well tolerated. Perceived insertion discomfort matched that expected for 13% (n = 6), was less than expected for 73% (n = 35), and more than expected for only 6% (n = 3) ofthe subjects. After insertion, 92% (n = 44) of the subjects returned immediately to their daily routine. Mean time for removal was 3 minutes. Although 50% (n = 21) of subjects expressed some fear of pain during removal, 63% (n = 30) stated that the actual procedure was without pain. Removal discomfort matched subjects' expectations for 33% (n = 14) of subjects, was less than expected for 48% (n = 20), and was more than expected for 14% (n = 6).
Figure 1 Number of bleeding episodes by number of days per episode during 422 months of capronor subdermal contraceptive implant use by 48 women.
Ovarian Cysts and Abdominal Pain
Three subjects complained of lower abdominal pain, and five asymptomatic women were found to have ovarian cysts on pelvic examination. These were confirmed by ultrasonography. Seven of the eight ovarian cysts resolved spontaneously, and one 3-cm cyst was drained at the time of laparoscopy, performed to determine the origin of persistent pelvic pain. Three other subjects presented with abdominal pain. One had a hydrosalpinx that was laparoscopically diagnosed, and the other two had normal pelvic examinations.
Bleeding Patterns
Local Effects
Thirty-two (67%) of the 48 subjects reported irregular or prolonged bleeding, whereas seven subjects complained of amenorrhea during use of capronor. Analysis of bleeding patterns obtained from menstrual calendars showed that of 13,829 total days of capronor use, 2,909 (21 %) were marked by vaginal bleeding. The mean number of days of bleeding per month per subject was 7 with an average interval of 23 days between episodes. Two thirds of bleeding episodes were of 4 to 8 days duration, but 6% of episodes were more than 15 days in length (Fig. 1). Absence of bleeding was reported less frequently than prolonged bleeding. There were only 36 instances of amenorrhea lasting >40 days in 35 woman-years of capronor use (Fig. 2). Correlation of P levels with individual bleeding histories demonstrated that women who ovulated regularly also had regular menses. Subjects experiencing suppression of ovulation were more likely to have irregular bleeding and were also more likely to be amenorrheic than those who ovulated regularly. The lower the concentrations of E2 the more likely a subject was to report bleeding.
Eight subjects experienced downward migration of the implant from the upper arm insertion site during the 1st through 5th months of use. Downward migration ranged from 1 to 6 cm from the original placement site and did not prompt complaints or make removal of the implants more difficult. Twenty subjects (41 %) reported pruritis directly over the implant. Sixteen of the subjects had a single
Vol. 58, No.1, July 1992
40
10
50
60
70
80
90
100
DAYS per INTERVAL
Figure 2 Number of intervals between bleeding episodes by number of days per interval for 422 months of use of capronor subdermal implant contraceptive by 48 women. (One subject was amenorrheic for 206 days.) Darney et al.
Capronor contraceptive
139
episode soon after insertion, whereas four complained of intermittent pruritis throughout implant use. One of the subjects with a single episode of pruritis and two subjects with intermittent pruritis were treated with a topical corticosteroid with resolution of symptoms. Nine subjects complained of an erythematous rash within several days of insertion. Six of these subjects received no treatment, and three applied a topical corticosteroid. All experienced complete resolution of symptoms within 3 to 5 days.
(Bars show standard error of the means.)
i
J5
-'
~
~
2S
.L
IS
>
~
05
Pregnancies
10
Three pregnancies occurred among women using capronor. The first was in a 2.5-cm implant user during her 8th month of use when levonorgestrel levels ranged from 0.08 to 0.14 ng/mL and the subject demonstrated bleeding patterns consistent with ovulation. The second pregnancy occurred 9 months after the insertion of a 4.0-cm implant when levonorgestrel ranged from 0.08 to 0.14 ng/mL. The third pregnancy occurred after 11 months of a 4.0-cm implant use at a levonorgestrel of 0.12 ng/mL. The Pearl Index for the 2.5-cm implant was 18.2 per 100 woman-years; for the 4.0-cm implant it was 4.5. Continuation Rates
The continuation rate for all users of capronor (excluding 6 who terminated use early because the study ended) was 43% at 12 months. Concern about decreased effectiveness when informed about low levonorgestrellevels prompted 6 (14%) subjects to discontinue use. Six subjects (14%) discontinued for reasons not related to the method: 3 for failure to keep appointments and 3 more because they were moving out of town. Twelve women (29%) discontinued capronor use for method-related reasons, with prolonged menses (n = 4) and pregnancy (n = 3) being the most common. When asked about their satisfaction with the method, 83% (35) stated they were satisfied with the implants, whereas only 10% stated they were dissatisfied. When asked whether they would use capronor again, 70% said yes. Serum Levels of Levonorgestrel
Serum levonorgestrel concentrations among users of the two capronor implants ranged from >0.8 ng/mL to <0.1 ng/mL. Determinants of levonorgestrel serum concentrations among subjects in the comparison of 2.5-cm and 4.0-cm capronor implants were the length of the implant and duration of its use (Fig. 3). Mean levonorgestrel values for women 140
Darney et al.
Capronor contraceptive
12
I.
MONnlS
Figure 3 Levonorgestrel concentrations (ng/mL) in women using 2.5-cm (n = 16) or 4.0-cm (n = 32) capronor subdermal contraceptive implants.
using the 4.0-cm implant were significantly higher than in women using the 2.5-cm device, although a more rapid decline in serum concentration during the 1st month was observed among users of the 4.0cm implant. By the 8th month of use, there were too few 2.5-cm users to make valid comparisons between the 4.0- and the 2.5-cm groups. Serum levonorgestrel concentrations in the 4.0-cm group consistently averaged >0.20 ng/mL for 1 year of use. In the 2.5-cm group, however, mean levels fell <0.20 ng/mL, a level judged inadequate for contraception by the 4th month of use. The nine women who elected insertion of two 2.5cm implants received the second capsule after 110 to 275 days of use of the single 2.5-cm implant. Insertion of a second 2.5-cm implant for these women delayed the decline of their mean levonorgestrel serum concentrations but did not sustain levonorgestrel levels >0.20 ng/mL for longer than women wearing the 4.0-cm implant. Effects on Ovulation
Eleven of 13 subjects who had frequent blood sampling to study ovulation received the 4.0-cm capronor implant (Table 1). In this group, only 19% of 104 cycles studied were ovulatory, and no ovulations occurred until after the 12th week of capronor use. By this time, levonorgestrellevels in this group had fallen to a mean of <0.35 ng/mL. Figure 4 shows that ovulation rarely occurred when levonorgestrel concentrations averaged >0.35 ng/mL. Three users of the 4.0-cm implant did not ovulate even when levonorgestrel concentrations fell below 0.25 ng/mL. For the two women who were studied Fertility and Sterility
Table 1 Cycles Monitored* and Number Ovulatoryt by Type of Capronor Device Cycle
2.5-cm
Ovulatory Nonovulatory Total cycles
Two 2.5-cm
4.0-cm
4 (67) 2 (33) 6 (100)
20 (19) 84 (81) 104 (100)
21 (95)+ 1 (5)
22 (100)
and the 2.5-cm implants. Even women who ovulated infrequently maintained E2 concentrations within the limits of normal. Levels of E2 were also within normal limits during the month immediately after insertion of capronor when levonorgestrel levels were at their peak (0.5 to 0.8 ng/mL).
* One or more P determination per week. t Progesterone :2: 3.0 ng/mL. Values in parentheses are percents.
Release Rates
+
The mean levonorgestrel release rate for subjects who used the 4.0-cm capronor implant was 33.0 ± 10.34 ~g/d with a range from a low of 20 ~g to a high of 65 ~g. Figure 3 shows that the average serum levonorgestrellevels of 32 women using the 4.0-cm capsule decreased from an initial value of 0.50 ng/mL to a final value of 0.20 ng/mL with a mean value of 0.23 ng/mL at 10 months. The observed decrease in the serum level of levonorgestrel was a consequence of a steadily decreasing rate of release from the capsules. In Figure 5, the amounts of levonorgestrel released from 4.0-cm capsules are plotted against the time the capsules remained in place. The average initial daily release rate was 51.0 ~g/d, which compared well with reported average in vitro release rates for 4.0-cm capsules of 54.0 ± 3.1 ~g/d (6). After
while they wore the 2.5-cm implant, all but 1 of 22 cycles were ovulatory. Of 30 ovulatory cycles occurring among the 11 frequently sampled users of the 4.0-cm implant, only 1 resulted in pregnancy. Both of the pregnancies that occurred during a total of 276 woman-months of 4.0-cm implant device use occurred in subjects who had levonorgestrel concentrations far below the group mean for their duration of use. Effects on E2
Mean E2 concentrations were within the limits for ovulating women among users of both the 4.022.5
20
o
o
o o
o
o o 00
o
o o
o
0
o
o
o o
o
o o
o
0
o 2.5
o 0
o
o
o
0 0 0
o
0
o 00. "oBtllfoi i.z, J.. Ol+-____~o~--~~~~~~==~~~~ o
o
__ __
" \ 0 00
.1
.2
.3
.4
000· 0
~~ .5
o 0
0
00
0
~r_~_ _~--~--;_----~~~--, .8
.7
.8
.9
MEAN LEVONORGESTREL LEVEL I CYCLE (NGIM.)
Figure 4 Mean serum levonorgestrel concentrations (ng/mL) per cycle and maximum serum P (ng/mL) per cycle in 11 women using 4.0-cm capronor subdermal contraceptive implant and having blood sampled at least once weekly. Vol. 58, No.1, July 1992
Darney et al.
Capronor contraceptive
141
,. ~
•
12
•
10
"
C
w
m a: irl
•
8
ii 0
0
/
/. 50
of 46.8% to a final average value of 60.0%, showing maintenance of their structural integrity .
•
Effects on Serum Lipoproteins
•
/ • 100
150
200
250
300
400
350
DAYS IMPLANTED
Figure 5 Amounts of levonorgestrel (mg) released from 4.0cm capronor subdermal contraceptive implants (n = 25) during various periods of implantation in women.
20 days, the average release rate of the capsules was 28.8 ± 2.4 ILgjd per capsule, approximately half the initial release rate. After 365 days, the calculated release rate decreased to 8.6 ILgjd. Degradation of Implanted Capsules
Six 4.0-cm capsules were analyzed for MWsand crystallinity contents to assess the degree to which they had degraded during use. All of the selected capsules had been implanted for approximately 1 year (344 to 373 days). Intrinsic viscosities in toluene were determined for all 32 of the retrieved 4-cm capronor capsules revealing the expected first order decline during implantation. The half-life time for the MW was found to be 10.6 months-in good agreement with an observed in vitro value of 10.0 months. The crystallinity of capsules implanted for approximately 1 year increased slightly from an initial value
As previously reported, there was no evidence that the use of capronor levonorgestrel implants caused adverse effects in blood lipids (Table 2). Although the HDL subfraction (HDL 2 cholesterol) tended to decrease slightly after capronor placement, the observed decrease was both statistically insignificant and proportionally smaller than the concurrent decrease in LDL cholesterol. The statistically significant decreases in LDL cholesterol and total triglycerides after placement of capronor were comparable in magnitude with those previously reported in users of silastic levonorgestrel implants (2, 9-11). One month after removal of capronor, LDL cholesterol and total triglycerides remained below preinsertion levels. Effects on Other Laboratory Values
There were no significant changes in serum F or blood chemistry levels over preinsertion values. One subject developed abnormalities in thyroid function and was discovered to have thyroiditis causing hypothyroidism. Five subjects developed mild anemia during study participation and one during the month after removal. None of the cases of anemia was associated with a history of heavy, prolonged, or irregular vaginal bleeding. Four of these five subjects were participants in the study of ovulatory patterns requiring 60 mL of blood sampling per week for up to 6 consecutive weeks. DISCUSSION
This single implant contraceptive system was easy to place under the skin and simple to remove through
Table 2 Differences Between Repeated Lipid Measurements Taken Before, During, and After Use of Capronor Subdermal Levonorgestrel Contraceptive Implants 3-mo postimplant versus baseline (n = 44)
Measured lipid Triglycerides (mg/dL) In whole serum In VLDL fraction Cholesterol (mg/dL) In whole serum In LDL fraction In HDL fraction In HDL2 subfraction
*t 142
=
Postremoval versus baseline (n = 23)
Mean difference
t*
Mean difference
t
Mean difference
t
-16.8 -6.2
-4.4t -2.3
-1.7 -0.2
-0.5 -0.1
-12.9 -2.5
-4.0t -1.1
-13.8 -8.0 -1.5 -1.3
-3.9t -2.7 -1.1 -0.83
-11.8 -6.0 -1.9 -0.4
-3.It -1.8 -1.1 -0.2
-18.6 -7.5 -0.8 +3.9
-5.7t -2.8 -0.5 +2.7t
difference/standard error of difference. Darney et al.
9-mo postimplant versus 3 mo (n = 19)
Capronor contraceptive
t Contrast statistically significant at P ,,; 0.05.
Fertility and Sterility
a 2 to 3-mm incision using finger pressure without instruments. The implants were difficult to grasp with mosquito clamps, so finger pressure was used to direct the implant tip into the incision from which it could easily be expressed. Experience with digital extrusion of capronor prompted our use of this technique with Norplant. Digital extrusion allowed for a smaller skin incision and less trauma and pain with decreased risk of implant fracture during removal of N orplant as well as capronor, but the latter implant was much faster and easier to remove (3minute mean for capronor versus our 22-minute mean for Norplant). In all but 2 of 32 subjects, the 4.0-cm implant maintained levels of levonorgestrel over the course of 1 year that were judged adequate to provide effective contraception (>0.25 ng/mL), but the 2.5cm implant did not maintain adequate serum levonorgestrel levels. The three subjects who became pregnant during their participation in the study had serum levonorgestrel levels considerably below the mean values for their groups. Compared with levonorgestrel-releasing silastic implants (N orplant), serum levonorgestrel levels were initially high and then declined more rapidly so that after 12 months, mean levels were generally too low to provide effective contraception, confirming capronor's I-year limit. A minority of subjects reported bothersome minor side effects typical for women using hormonal contraception such as weight gain or loss, acne, mastalgia, and mood changes, in order of frequency. Only aberrations of menstrual pattern were clearly associated with varying serum levonorgestrel levels and with ovulation. All women maintained E2 levels within normal limits, an important consideration in hormonal contraceptives that do not replace estrogen (12). Implants left in place for up to 1 year maintained structural integrity sufficient to permit easy removal. Because none ofthe implants were allowed to remain in place beyond 1 year, we cannot specify the time after insertion at which the implants completely degrade or what the levonorgestrel content of such implants might be, but release rates were low after 1 year. Analysis of the retrieved implants revealed a small increase in their crystallinity and a first order decrease of their MWs. The implants remained strong enough, however, for easy removal at 1 year. A symptom clearly related to capronor use was local irritation ofthe skin over the implant. It seems likely that a residue of manufacture that remained on the implants caused this irritation. It was selflimited and mild, but its occurrence suggests that Vol. 58, No.1, July 1992
after manufacture a thorough washing of implants is required. If capronor is to provide a useful contraceptive option, the implant must maintain serum levonorgestrel levels above approximately 0.25 ng/mL throughout a year of use. The 2.5-cm implant does not do so and will not be further studied. The 4.0cm implant can probably be altered to maintain this serum levonorgestrellevel by eliminating the oleate excipient to increase slightly the levonorgestrel release rate. The ease of insertion and removal of this I-year, single implant makes it a promising possible addition to the sustained-release progestin contraceptive systems introduced by Norplant. Its biodegradability is another attractive characteristic because removal may not be necessary. The caproic acid capsule may also be a satisfactory vehicle for other contraceptive progestins. REFERENCES 1. Segal S. A new delivery system for contraceptive steroids. Am J Obstet Gynecol 1987;157:1090-2. 2. Roy S, Mishell D, Robertson DN, Kraus RM, Lacarra M, Duda MJ. Long-term reversible contraception with levonorgestrel-releasing silastic rods. Am J Obstet Gynecol 1984;148:1006-13. 3. Diaz S, Pavez M, Miranda P, Robertson DN, Sivin I, Croxatto HE. A five-year clinical trial oflevonorgestrel silastic implants (NorplaneM ). Contraception 1982;25:447-58. 4. Sivin I. International experience with Norplant and Norplant2 contraceptives. Stud Fam Plann 1988;19:81-94. 5. Darney PD, Klaisle CM, Tanner ST, Alvarado AM. Sustained release contraceptives. Curr Probl Obstet Gynecol Fertil 1990;13:87-125. 6. Pitt CG, Gratzl MM, Jeffcoat RA, Zwidinger R, Schindler A. Sustained drug delivery system. II. Factors affecting rates of release from poly (E-caprolactone) and related biodegradable polyesters. J Pharm Sci 1979;68:1534-8. 7. Darney PD, Monroe SE, Klaisle CM, Alvarado AM. Clinical evaluation of the Capronor contraceptive implant: preliminary report. Am J Obstet Gynecol 1989;160:1292-5. 8. Myers LH, Phillips NR, Havel RJ. Mathematical evaluation of methods for estimation of the concentration of the major lipid components of human serum lipoproteins. J Lab Clin Med 1976;88:491-505. 9. Shaaban M, Sawaan I, Elwan S, Darwish HA. Effect of subdermallevonorgestrel contraceptive implants, Norplant, on serum lipids. Contraception 1984;35:413-9. 10. Affandi B, Suherman S, Djajalelana J, Hadisputra W, Moeloek FA. Serum lipids in Norplant implant users: a cross sectional study. Contraception 1987;36:429-34. 11. Viegas 0, Singh K, Liew D, Singh P, Ratnam SS. The effects of Norplant on clinical chemistry in Singaporean acceptors after one year of use: metabolic changes. Contraception 1988;38:79-89. 12. Ory SJ, Hammond CB, Yancy SG, Hendren RW, Pitt LG. The effect of a biodegradable contraceptive capsule (Capronor) containing levonorgestrel on gonadotropin, estrogen, and progesterone levels. Am J Obstet Gynecol 1983;145: 600-4. Darney et al.
Capronor contraceptive
143