S. Diaz, 0. Peralta, G. Juez, C. Herreros, M.E. Casa do, A.M. Salvat.ierra, P. Miranda and H.B. Croxatto
C0NSULT0RI0 DE PLANIFICACION FAMILIAR Centro Nacional de la Famllia J.V. Lastarria 29 - Depto. 101 Santiago, CHILE.
Abstract
Six progesterone pellets implanted subdermally were tested as contraceptive method for lactating women. One pregnancy was diagnosed in 1614 woman-months of observation, a failure rate which was similar to that observed In a contemporary group of Copper T users. Nineteen pregnancies were diagnosed in the 677 woman-months observed in untreated lactating women. The progesterone implants were effective when administered either at 30, 60 or 240 days after delivery. The duration of the effective life was 5 months and fertility was quickly restored afterwards. There were no deleterious effects upon maternal or infant health or upon lactation and the rate, of child growth. The main problem encountered was the occurrence of pellet expulsion at a variable rate which appeared related to the manufacturing procedure.
Introduction
The inhibition of fertility associated with breastfeeding is highly variable among individuals and tends to decrease as time elapses following delivery. This fact determines the need for contraceptives that can be used during lactation without interfering with milk production or child health (1, 2, 3).
In previous publications 5) we discussed the potential advantages of using the natural hormone, progesterone, as a contraceptive during breastfeeding and the preliminary results obtained with a subdermal Implant of six progesterone pellets were reported. This paper gives an updated account of the contraceptive effectiveness of progesterone administered by this means and discusses the observed advantages of such a method.
Materials and methods
Population
Volunteers enrolled in the study were selected from among healthy postpartum women who met the following requirements:ages IS to 35, parity 1 to 3, normal pregnancy and vaginal term delivery of a healthy infant with a birth weight adequate to the gestational age. They had to be regularly cohabiting, free from drug therapy and willing to nurse their infants as long as possible.
Selection was done at the maternity ward where instructions on the technique of breastfeeding on demand were given. Additional requirements at the time of Initiating treatment were normal postpartum evolution, normal physical examination of mother and child, hemoglobin values higher than 10 gZ and a normal nutritional state. All cases were required to be in exclusive breastfeeding with an adequate infant growth rate before admission.
Treatments
Progesterone was administered by subdermal insertion of six progesterone pellets. Contemporary control groups were formed by women treated with an IUD or a placebo injection. The allocation to an injectable method or an HID was at patient choice.
- Progesterone pellets:pellets were made by compressing 100 mg of the steroid into a cylinder of approximately 11.8 mm length by 3.2 mm diameter. Three different batches (I, II and III) were manufactured using different techniques. Compression was applied along the longitudinal axis in batches II and III and perpendicular to it in batch I. The progesterone used was powdered in batches I and II and precompressed in batch III. Each subject received 6 progesterone pellets that were inserted subdermally in the gluteal region through a No. 9 gauge trocar under local anesthesia. Insertion of pellets was done at day 30 or 60 after delivery. Pellets were expected to have an effective life of 5 or 6 months. Accordingly, women who were treated at day 60 were offered a second set of pellets at approximately day 260 postpartum provided they were still nursing their babies.
-Intrauterine device: The IIJD used in this study was the Copper T 200 (supplied by the AsociaciSn Chilena de Proteccion de la Familia). It was inserted at day 30 or 60 postpartum to match the contemporary groups using progesterone pellets.
-Placebo: The placebo consisted of an injection of 3 ml of distilled water, offered as a test treatment to sustain lactation and therefore to support the fertility inhibition associated with breastfeeding. Women enrolled in this group were offered non-hormonal contraceptives at the 6th postpartum month or earlier if supplementary bottle feeding was introduced. The placebo was administered at day 30 postpartum. This group was discontinued when it became evident that full nursing alone or partial nursing supplemented with barrier methods provided poor protection (6).
Follow-up
Follow-up was designed to determine the occurrence of pregnancy, breastfeeding performance, infant growth, side effects and bleeding pattern. Visits were scheduled at monthly intervals up to the sixth postpartum month and at two-month intervals thereafter. At each visit, the maternal and infant histories were recorded and both mother and child were examined. Breastfeeding performance and infant growth were carefully evaluated using the Boston curve for boys (7) as the reference standard. Special calendars were distributed to record all bleeding and spotting days and the number of breastfeeding episodes per day.
Pregnancies were diagnosed by clinical examination and urinary immunological tests for HCG. Pregnancy tests were performed when menses were delayed for 20 days or more, when amenorrheic women reached the 6th postpartum month and before the insertion of a second set of progesterone pellets at day 240 after delivery.
Progesterone blood levels
Blood samples were drawn at 10-or 15~day intervals from a subsample of users of the 3 different batches of progesterone pellets during the first segment of use and from control subjects. Progesterone levels were determined by RIA as previously described (8) using the procedures and reagents provided by the Programme for the Provision of Matched Assay Reagents for the RIA of Hormones in Reproductive Physiology of the World Health Organization.
Data analysis
The month of conception was estimated by adding 14 days to the first day of the last bleeding run observed. It was confirmed or adjusted when possible by reference to the estimated gestational age of the neonate. The Pearl Index was used to estimate the probability of pregnancy at various intervals during treatment.
Lactation was classified up to the 6th month of age as full nursing or exclusive breastfeeding if the breast was the only source of nutrients for the baby. After the 6th month, non-dairy food was prescribed routinely and lactation was still classified as full nursing if the breast was the only source of milk. The proportion of cases remaining in the full nursing condition was used to assess the effect of treatment upon lactation. Supplement was prescribed by the pediatrician when the rate of infant growth indicated inadequate milk supply. In several cases, the supplement was self- prescribed by the mother based upon her subjective evaluation of milk production and child satisfaction.
The occurrence of severe intercurrent disease, prolonged or repeated separation of mother and child, discontinuation of treatment or los t-to-follow-up determined exclusion of the case from analysis of lactation and infant growth following the last visit prior to the event.
In order to assess the effect of treatment upon infant growth, the average absolute weight at each month of life and the average weight increase at monthly and daily Intervals were calculated for each group. Only cases in exclusive breastfeeding were included. Calculations were done up to the 8th month of age.
The occurrence of uncomfortable or prolonged bleeding was analyzed through the proportion of women who experienced more than 6 days of bleeding plus spotting per 30-day interval after treatment.
Contingency table analysis and student's"t" test were used to assess the statistical significance of differences in distribution and mean values, respectively. Only p values < 0.05 were considered significant.
Results
The number of subjects enrolled in each treatment group is shown in Table I. Hinety-two out of 112 women (82.1%) accepted the reinsertion of progesterone pellets that was offered at day 240 postpartum.
Table I. Distribution of Subjects According to Treatment and Day of Administration.
The plasma progesterone levels achieved with the 3 different batches of pellets are shown in Figure 1. They reached their maximum within ten days of insertion and declined gradually thereafter remaining elevated at least for 5 months in comparison to the control group. Levels obtained with the 3 hatches were of similar magnitude.
Figure 1. Plasma progesterone levels (X ± S.E.) in lactating women treated with progesterone pellets from Batch I (□) Batch II ( Δ ) and Batch III (o) and in control cases (•).
The local tolerance to progesterone pellets was poor. Expulsion occurred in 2.4% of subjects from hatch I, in 31.7% of subjects from hatch II and in 9.9% of subjects from batch III. The time elapsed between insertion and expulsion varied from several days to five months.
Table II shows the occurrence of pregnancy in women treated with progesterone pellets according to the breastfeeding status and the time of treatment. This table includes only the first 5 months after each progesterone implant insertion. For this reason, the 8th postpartum month is not included. No pregnancies were observed in the first 5 months of treatment in full nursing women when treatment was given at day 30, 60 or 240 postpartum. Only one pregnancy was diagnosed in partially nursing women and it occurred in the 4th month after insertion. The overall incidence of pregnancy was 1 in 1614 months. This figure Is comparable to the fertility inhibition obtained with the Copper T in the same postpartum intervals. Both treatment groups showed a significantly lower incidence of pregnancy as compared to the placebo group in which 19 pregnancies were diagnosed in the 677 woman-months observed in comparable postpartum intervals.
Table II. Contraceptive Effectiveness of Progesterone Pellets versus Copper T and Placebo According to the Breastfeeding Status and the Observation Period.
Table III. shows the overall contraceptive effectiveness of six progesterone pellets for different treatment lengths, regardless of the time of insertion’. Treatment was highly effective during the first five months of use. The effectiveness decreased when the observation was extended to months 6 and 7 where 3 and 6 pregnancies were diagnosed in 239 and 72 woman-months of exposure, respectively.
Table III. Duration of the Contraceptive Effectiveness of Six Progesterone Pellets.
The cumulative distribution of women according to their breastfeeding status at the 6th postpartum month is shown in Table IV. Treatment groups initiated at day 30 or 60 postpartum had breastfeeding evolutions similar to their contemporary control groups. Both series initiated at day 60 showed a higher proportion of supplemented cases at the 6th month when compared to the series started at day 30.
Table IV. Cummulative Distribution of Women According to their Breastfeeding Condition at the 6th Postpartum Month.
The growth of exclusively breastfed infants was similar in treated and control groups when analyzed through the average absolute body weight at each month of age and the average monthly and daily weight increase. Table V shows the average weight increase-of exclusively breastfed infants from birth to the 6th month of age. Values for all groups were within the normal range according to the standard used for comparison.
Table V. Average Weight Increase of Exclusively Breastfed Infants from Birth to the 6th Month of Age.
No major side effects were detected. The bleeding pattern showed disruption during the first month of use of pellets comparable to that observed after insertion of a Copper T. Table VI shows the proportion of full nursing women who experienced 6 or more days of bleeding plus spotting per 30- day interval. Excessive bleeding was infrequent in all groups after the first month of treatment. After the insertion of the second set of implants at day 240 postpartum, the proportion of women who reported more than 6 days of bleeding per interval ranged from 6% (third month of treatment) to 16% (fifth month). In the same period, the proportion of Copper T users who experienced more than 6 days of bleeding ranged from 21% to 31%.
Table VI. Percentage of Full Nursing Women who Experienced More than 6 Days of Bleeding or Spotting per 30-Day Interval.
Discussion
The results of this study show that the continuous administration of progesterone by means of an implant inhibits -fertility in lactating women and that it can be used for contraceptive purposes. The pregnancy rate observed during the period of elevated plasma levels of the steroid is highly satisfactory and within the range of a Copper T IUD. The high incidence of pregnancies observed in the placebo group allows the conclusion that the low fertility of the treated groups was not determined exclusively by breastfeeding.
The duration of elevated plasma progesterone levels and of an acceptable contraceptive efficacy was limited to 5 months. After this period, fertility was quickly restored as can be inferred from the number of pregnancies observed at the 6th and 7th month after insertion. This led us to offer a second set of implants after the exhaustion of the first reservoir to women who were still lactating at that time. The low incidence of pregnancies observed in the second segment is reassuring of the contraceptive effectiveness of progesterone implants during lactation and shows that its efficacy applies also to long-term lactation.
The use of progesterone induced no adverse side effects on lactation as judged by the proportion of cases in exclusive breastfeeding and the growth rate of exclusively breastfed infants up to the 6th postpartum month. Treatment with progesterone pellets or a Copper T resulted in a higher proportion of subjects who experienced more than 6 days of bleeding as compared to the untreated women. Nevertheless, this was a transient problem, limited to the postinsertion period and did not represent a reason for discontinuation or a health risk for the subjects. No deleterious effects upon infant health were detected.
The percentage of women who were willing to use a second set of progesterone implants indicates that this is a well accepted alternative.
The main problem encountered during this experience was the high incidence of expulsion of the progesterone pellets which appears to be partially related to the manufacturing technique. The results reported justify new efforts to develop an improved system for the long-term delivery of progesterone.
Acknowledgements
The authors are grateful to the authorities of the hospital Paula Jaraquemada and the Central Health Area of Santiago for the facilities provided for this study, to the Instituto Bioqulmico Beta for the provision of progesterone pellets, to the WHO Programme for the Provision of Matched Assay Reagents for the RTA of Hormones in Reproductive Physiology, and to the International Development Research Centre of Canada and to the International Committee for Contraception Research of The Population Council for their support.
References
Van Ginneken, J.K. The chance of conception during lactation. J Biosoc Sci, Supp 4:41-54 (1977).
Laukaran, V.H. Contraceptive choices for lactating women: suggestions for postpartum family planning. Stud Pam Plann 12:156-163 (1981).
McCann, M.F. , Liskin, L.S., Piotrow, P.T., Rinehart, W. and Fox, G. La lactancia materna, la fecundidad y la planificaci6n familiar. Population Reports J 24, October 1982.
Croxatto, H.B., Diaz, S., Peralta, 0., Juez, G., Casado, M.E., Salvatierra, A.M. and Duran E. Fertility regulation in nursing women, II, Comparative performance of progesterone implants versus placebo and Copper T. Am J Obstet Gynecol 144:201-208 (1982).
Croxatto, H.B., Diaz, S., Peralta, 0., Juez, C., Herreros, C. , Casado, M.E., Salvatierra, A.M. and Miranda P. Subdermal progesterone Implants for fertility regulation in nursing women. Presented at the International Workshop on Long-Acting Contraceptive Delivery Systems. PAJRFR, May 1983.
Diaz, S., Peralta, 0., Juez, G., Salvatierra, A.M. , Casado, M.E. , Duran, E, and Croxatto, H.B. Fertility regulation in nursing women. I. The probability of conception in full nursing women living in an urban setting. J Biosoc Sci 14:329-341 (1982).
Nelson, W.E., Editor. Textbook of Pediatrics, 8th ed. U.B. Saunders Co., Philadelphia P.A., 1964. p. 42.
Croxatto, H.B., Diaz, S., Peralta, 0., Salvatierra, A.M. and Brandeis, A. Plasma progesterone levels following subdermal implantation of progesterone pellets in lactating women. Acta Endocrinol 100:630-633 (1982).