Hormone-releasing silicone-rubber intrauterine contraceptive devices Effect of incorporation of various intrauterine devices in rats
LEE
LEE
Little
Rock,
compounds
on
DOYLE Arkansas
Intrauterine devices (IUD’s) containing 0, 5, or 10 per cent by weight of progesterone, melengestrol acetate (MGA), norethindrone, norgestrel, medroxyprogesterone acetate, isoxsuprine, or R2323 or wound with fine copper wire were placed in one or both horns of cycling female rats. All the progestins improved retention of the silicone-rubber IUD’s to some degree. The effects of these compounds on the estrous cycle, mating, ovulation, fertilization, tubal transport, and implantation varied among bhe compounds. Effects were local or systemic, depending upon the amount and the drug. Uterine motility studies showed clearly
could
that 5 per cent MGA decreased uterine motility; however, no be obtained using the other compounds in the same in vivo system.
E F F 0 RT S T 0 DECREASE the failure Kite, expulsion rate, and the occurrence of side effects of intrauterine contraceptive devices were first directed toward modification of size, shape, or material of the IUD. Introduction of the amazingly large number of variations on a scheme, while illustrating imaginative engineering in a limited space, only ameliorated the problems associated with the traditional IUD-it did not solve them. An alternative approach is now being investigated in which the IUD is used as a carrier for materials that are, of themselves, effective contraceptives. Two types of compounds, metals and hormones, have been studied thus far, and both appear to offer great promise. From the Department Gynecology, University Medical Center, Delta
of Obstetrics and of Arkansas Regional Primate
Supported by Population Award No. M72138/ICCR-22 Institutes of Health Grant Received Revised Accepted
for publication July July
results
Among the metals studied by Zipper, Mendel, and Prager,’ copper was shown to be most effective, and it is already being studied using the Tatum T and the Searle Cu 7 in large clinical trials.2 While some investigations have been undertaken in human beings with hormone-impregnated intrauterine devices, only two hormones, progesterone and Provera, have been used extensively. Pilot studies in animals have shown that other hormones could be utilized.3 To determine what other compounds might be equally or more effective, we have undertaken studies in laboratory rodents and primates to screen possible compounds and to compare the efficacy and delineate the characteristics of the most promising ones. The first part of the study deals with the effect that incorporation of 8 different compounds into IUD’s has on retention and reproduction in rodents.
Center.
Council Research and National HD04019. April
consistent
Material
and
methods
The hormones used included Provera,* norgestrel,? norethindrone,$
2, 1974.
4, 1974. 26, 1974.
Reprint requests: Dr. Lee Lee Doyle, Department of Obstetrics and Gynecology, University of Arkansas Medical Center, Little Rock, Arkansas 72201.
tion
405
*Courtesy
of The
Upjohn
tcourtesy Council.
of Wyeth
$Supplied
by The
Co.
Laboratories Population
progesterone, melengestrol
through Council.
the
Popula-
406
February 1, 1975 Am. J, Obstet. Gynecol.
Doyle
Tabie I. Effect of incorporation on retention
Treatment Control 5% MGA 5% Provera 5% progesterone 5 % norethindrone 5% norgestrel 5% R2323 5% isoxsuprine Coppert
Fig. 1. Silicone rubber sleeve after autoclaving,
IUD in polyethylene and ready for insertion.
sleeve,
in
acetate (MGA) ,* R2323,t and is0xsuprine.S In addition, one group of devices were wound with fine copper wire.@ The devices were made by adding 0, 5, or 10 per cent by weight of the compound to liquid silicone rubber (382 medical-grade elastomer, Dow Corning Corp.). An appropriate amount of catalyst was added, and the mixture was injected into polyethylene tubing with a 1.07 mm. inner diameter and allowed to cure at room temperature. The devices were cut into 8 mm. lengths and autoclaved for sterility and simplified removal from the polyethylene sleeve (Fig. 1) . In some early experiments, barium sulfate was included for radiopacity, and various pigments were incorporated so that the devices could be easily differentiated. As we could detect no difference in results between devices with and without pigment or barium, results were not separated. Adult Sprague Dawley rats were used in the experiments. Female rats were checked daily using vaginal lavage to make sure that all were exhibiting normal estrous cycles. At the time of the operation, done under ether the uterine horn was exposed via a anesthesia,
*Courtesy +Courtesy Council. SCourtesy $Supplied
of The Upjohn Co. of Roussel UCLAF of Mead Johnson. by The Population
through
Council.
the
Population
of various by nonmated rats
of IUD’s
hormones
Laparotomy at day 7 after surgery (no. IUD’S retained/no. IUD’S inserted* J
Autopsy at day 14 after surgery (no. IUD’S retained/ no. IUD’s inserted*)
32 26 10 10 12 12 12 30
7/15 16/18 9/10 7/to
4/23 16/16 8/10 6/10 11/12 lo/12 12/12 9/20
I
3/7
l/7
777
2/7
No. Of rats
(46.5) (89) (90) (70) -
6/10
(60)
(17.4) (100) (80) (60) (91.5) (83.5) (100) (45)
*Percentages in parentheses. tRats had a copper-wound 8 mm. segment of Tatum T in one horn and an 8 mm. segment without copper in the opposite horn.
dorsolateral incision, and a small stab wound was made in the antimesometrial wall through which the IUD was inserted. The horn was closed with a single stitch of 6-O silk in order to prevent retrograde expulsion, and the muscle and skin were closed with cotton suture. Laparotomies were done via a single ventromedial incision. All female rats used in mating studies were put into the male rats’ cages and checked daily for presence of a copulation plug or sperm, and if either was found, this was considered day 1 of pregnancy. As the timing of mating, laparotomy, and autopsy will vary, depending upon the parameter being studied, it is included with the results of the individual experiments. Results Retention studies. The effects of 5 per cent compounds on IUD retention in nonmated animals are summarized in Table I. In most cases, laparotomy was done on day 7 ; however, 3 groups were not done at this time and are indicated by the under the day 7 column. In the initial study of each compound, only one device was used per animal, and no matings were attempted. In this first screening, all 5 of the synthetic progestins appeared to be effective in enhancing IUD retention. The retention rate at 2 weeks ranged from 80 to 100 per cent, as compared to 17.4 per cent retention in controls. Progesterone ap-
Volume 121 Number 3
Hormone-releasing
Table II. Effects of mating
on retention
I Treatment
No.
of rats
5% 5% 5% 5%
MGA Provera isoxsuprine norgestrel
20 13 ii 15 10 20
5%
norethindrone
20
5%
R2323
19
MGA
10 10
*These animals fPlug or sperm
were done in 2 groups: was never
recorded;
/
1 Days tion
IUD’s
407
of IUD’s
I
Control
10%
silicone-rubber
from inserto mating 1 7 7 1 1 7 7 7 7 7 7 :
in group
1, lo/IO
however, the animals
parently enhanced retention as did the nonsteroidal uterine relaxant, isoxsuprine, but neither compound appeared to be as effective as any of the synthetic steroids. Copper had no retaining effect. The copper IUD’s were of greater diameter than those of silicone rubber as they were cut from Tatum T’S In order to establish the proper control conditions, IUD’s without copper were made from the nonwrapped portion of the T and inserted into the contralateral horn. The second series of experiments involved mating the animals either immediately or 7 days after IUD insertion to determine if the hormone would inhibit mating or if mating caused expulsion of devices. At the time of laparotomy or autopsy, which varied between 2 and 14 days after breeding, all mated experimental groups had a poorer rate of device retention (Table II). In two sets of animals, one group caged with male rats on day 1 and the other on day 7, 10 per cent MGA devices were placed in one horn, and 90 per cent of these animals retained the devices until the time of autopsy; however, only a few of the animals mated. In several groups of animals, an active device was placed in one horn and a control device was placed in the contralateral horn to see if the effects of the hormone were limited to the horn containing the active IUD or if the drug exerted enough effect systematically to cause control-device retention in the opposite horn. These results can be found in Table III. All the 10 per cent devices had some effect on control-device retention, indicating a probable systemic effect. Norethindrone, R2323, ahd
No. 1 14 13 10 10 5
mated (%I
Days I
(70) (100) i55)Y (67) (50)
at ,
7to 14 7to14 7to 14 5to14 7to14 2 to 4 7 to 14 2 to4 7to17 2 to 4 7 to 14 14 14
18’+“5(90) I\
pregnant autopsy
Retention (no. IUD’s retained/ no. IUD’s ;nrcvtunl) _‘.“-’ -“-/
I
17’+3’t(85) 18”“t (96) 1 (10) 3 (30)
3/20 2/13 9/18 10/15 2/10 6/10 8/10 3/10 8/10 6/9 6/10 9/10 9/10
mated; in group 2, O/8 mated. were pregnant at autopsy.
Table III. retention
Effect of hormone-releasing of contralateral control IUD’s
Treatment 5%
MGA Control 570 Provera Control 570 norethindrone Control 570 norgestrel Control 5% R2323 Control 1070 MGA Control 1070 norethindrone Control 1070 norgestrel Control 1070 R2323 Control
No. of rats 5
Autopsy (days after insertion) 7to14
8
14
10
14
10
14
10
14
10
14
10
14
10
14
10
14
IUD’s
on
Devices retained/ devices inserted 4/5 l/5 6/8 5/8 6/10 o/10 4/10 3/1O 9/10 3/10 lO/lO lo/lo 8/1O 4/10 8/10 lO/lO lO/lO 4/10
MGA at the 5 per cent level appeared to act only locally, but 5 per cent Provera and norgestrel caused retention of some nonmedicated devices. Reproductive studies. All doses of all hormones caused aberrant vaginal cytology; however, only the 10 per cent MGA and isoxsuprine seemed to inhibit mating (see Table II). Mating was recorded as having occurred when either a copulation plug or sperm was found, but apparently, these were missed on several occasions because, at autopsy, animals that had shown no evidence of mating were found to be pregnant.
408
Februaly Am. J. Obstrt.
Doyle
Table IV. Effect of mating
on retention
Animals
de&es
Unmated f% J
Mated cc/r )
MGA Provera
100 80
50
norethindrone norgestrel R2323
91.5 83.5 100
55 70 64
Treatment
5% 5% 5% 5% 5%
retaining
of IUD’s
60
The retention rate of devices in the mated animals was less than that in the nonmated rats for all compounds at 5 per cent (Table IV). With the exception of MGA and Provera, the compounds apparently inhibited implantation to a significant degree in the control as well as the treated horns (Table V) . In all cases, the IUD, either control or active, was an effective contraceptive as only one animal showed anything remotely resembling an implantation when an IUD was present. In this single animal, there were 2 areas that could have been resorbing fetuses. To differentiate between a failure in the process of implantation and disruption of an earlier event such as fertilization or cleavage, a series of experiments was done to recover ova from the uterus and tubes. In these experiments, no treatment was given to control animals so values on normal rates of fertilization, cleavage, and recovery could be established for female rats in our colony. The treated animals had devices inserted and were put with males 7 days later. Once again, mating apparently increased device expulsion as all treated groups had 10 mated animals but only 3 to 6 had devices upon recovery 2 to 5 days after mating. The small number of animals in each group makes statistical comparisons among the treatment groups impossible, but it appears from the data in Table VI that all the hormones except norethindrone may interfere with either fertilization or development of ova to some extent. Only R2323 seems to have more effect on the treated side than on the contralateral control side. The incidence of abnormalities including nonfertilization or cleavage was increased on both sides of all treated groups. Comment No intrauterine contraceptive device can be effective if it is not well retained; thus, in our studies aimed at establishing which compounds might be promising for clinical trials in human beings, re-
1, 1975 C:ynecol.
tention was the first and most important criterion to be tested. The rat was a suitable animal for retention studies as control IUD’s are very poorly retained. Three compounds-progesterone, isoxsuprine, and copper-caused an increased rate of retention compared to controls (14.3 to 60 per cent compared to 17.4 per cent) after two weeks of use but were far less effective than any of the 5 synthetic progestins, which achieved 80 to 100 per cent retention for the same time period. Most experimental groups showed a decline in the retention rate following laparotomy or mating, the latter effecting the more dramatic decrease. We considered that the difference in retention could be due to one of two causes; either the compound was being released in such quantities that there was no longer enough remaining in the device to maintain an effective release rate, or the uterine horns were responding to the trauma of laparotomy with frequent intense contractions, and the amount of hormone being released was insufficient to counteract them. In one experiment, designed to assess the importance of each factor, there were 3 experimental groups of 8 rats per group. All animals had a 5 per cent MGA device inserted in one horn, after which group 1 had no further treatment, group 2 underwent laparotomy on day 7, and group 3 was placed with male rats on day 7. At autopsy on day 14, groups 1 and 2 had 100 per cent retention, and group 3 had 90 per cent. An additional group of 10 animals, which at laparotomy on day 7 showed 8 of 10 devices in situ, was then mated on day 7, and at autopsy one week later, only 2 animals had retained devices. From these data, mating or laparotomy alone seemed to have little effect on expulsion rate; whereas, a combination of the two virtually negated the effect of the hormone. The rate of retention was not as high in other studies of mated animals treated with 5 per cent MGA or, in fact, with any of the synthetic steroids so one can probably assume that mating does influence expulsion. The 10 per cent MGA devices were not expelled in those few animals which mated. From these studies, we concluded that all the synthetic steroids were effective in enhancing retention of silicone-rubber IUD’s but that, at the 5 per cent level, they were not as effective in mated as in nonmated animals. The vaginal cytology from all animals seemed to indicate aberrant estrous cycles; however, in only one group of animals with a 5 per cent Provera device was there any decrease in the num-
Volume Number
121 3
Table
V. Effect
Hormone-releasing
of hormone-releasing
IUD’s
No. of rats
No. with de#vices in
18
2
Control
No. with devices out
MGA
10
2
5%
Provera
9
3
5 r/r
norethindrone
10
8
5%
norgestrel
10
8
5%
R2323
10
6
MGA
10
9
8 6 2 2 4 10%
1 *These
two
areas
could
Table VI. Effect ova recovery
have
been
IUD’s
on
No. of rats retaining devices with ova in
Treatment Control* Control device 5% Provera 5% norgestrel 5% norethindrone 5% R2323 *These
animals
19 5 10 10 10 10 had
no device
409
Average no. of implants 5.1 0 3.2 0 1 0 6.0 2* 5.0 0 1.3
side
Control
Average no. of corpora Average no. lutea of implants 7.8 4.0 6.8 9.0 4.0 6.5 6.0 8.0 8.0 5.0 7.0 -
5.1 8.0 5.7 5.0 4.0 7.5 8.0 7.0 5.0 8.0 5.3 -
side Average no. of corpora lutea 7.2 8.0 8.0 5.0 6.0 8.5 8.0 10.0 7.0 8.0 6.0 -
resorptions.
of hormone-releasing
No. of No. of rats rats with bred 1 device
No. with implants 0 16 1 7 1 1 2 1 1 1 1 3 0 0
16 5%
IUD’s
on implantation IUD
Treatment
silicone-rubber
Treated Control horn or horn or I tube I tube
0 6 6 3 6
19 4 3 3 3 1
19 4 4 4 3 4
inserted.
ber of rats mating. The changes in shed of vaginal cells may simply reflect a local “downstream” effect of the progestin-bearing uterine fluid that bathes the vagina rather than indicating any changes in ovarian steroid secretion. The ability of locally applied hormones to alter vaginal cytolo.gy has been demonstrated by use of estrogen on castrate mice.4 An alternative explanation is that small amounts of progestins do get into the systemic circulation, and while this low level can cause a change in vaginal cytology, it is not sufficient to inhibit mating. It appeared in our earlier studies with MGA that the 5 per cent devices acted locally on the uterus, while the 10 per cent devices affected all reproductive processes on a systemic level. In a sense, this would be analogous to the situation encountered in Hooker and Forbes’” assay for progestrone where
small amounts of the hormone caused cellular changes only in the area of the intraluminally injected progesterone, while larger amounts affected the contralateral horn as well. A possible explanation for our results with MGA is that small amounts are bound or metabolized in the uterus and only when in excess of receptor sites find their way into the systemic circulation. Another explanation is that all the MGA circulated systemically, but the amount was insufficient to interfere with reproductive parameters. In this latter case, if the progestin is acting systemically, it should be capable of eliciting the same reaction in the control as in the treated horn. If the amount being released is sufficient to cause device retention in the treated horn, it should cause device retention in the control horn. Although the retention rate varied with the compound used, in no case in which animals had a 5 per cent device in one horn and a control device in the other was the retention rate of the control device as good as that of the treated device; however, 5 per cent Provera and norgestrel did not show much difference between horns. In studies with synthetic steroids other than MGA, the distinction between local and systemic effects was not so clear cut. It may be that larger amounts of these compounds are released or that the same amount is released but less is necessary to obtain the same biologic end point. Studies on actual in vivo release rates are being done at present. In considering the question of local or systemic effect, one must clearly define the biologic end point in question. None of the 5 per cent
410
Doyle
steroids studied inhibited mating; however, all appeared to have some systemic effect on other reproductive parameters that may well be overcome by giving less of the compound. In the retention studies, the action appeared to be only local which contradicts the data from reproductive studies where, in some cases, there was evidence for systemic effect. From this, we might postulate that there is a local concentration of hormone exceeding the level found in the general circulation. If we further assume that more progestin is needed to inhibit uterine contractions than to interfere with tubal transport and implantation, we can explain the apparently contradictory results. We had hoped to accumulate additional evidence for the local-effect theory with in vivo uterine motility studies. These studies were done with several compounds, but findings were variable, inconsistent, and inconclusive in all but the 5 per cent MGA studies where a decrease in frequency and amplitude of uterine contractions was found. After much thought and discussion concerning possible sources of error and variance and how to combat them, we realized that an entirely new project would be necessary and that the expenditure of time and effort was not warranted by the amount of data we could gain. This decision left us unable to point directly to demonstrable decreases in uterine motility as the cause of enhanced IUD retention; however, our indirect evidence supports this explanation as does the work of others. The ability of intrauterine progesterone to act locally to affect uterine motility in animals has been clearly demonstrated by Porter.+* In addition, studies by Csapos and others, summarized at a conference concerning the effects of progesterone on the myometrium, have shown that the human myometrium underlying the placenta contains higher levels of progesterone than are found in the surrounding myometrium.10 Since progestins have been shown to increase the resting transmembrane potential, (I* 11-12 they could interfere with the conduction of impulses and, thereby, block the rhythmic contractions necessary for expulsion of a fetus or a foreign body. That the IUD can increase uterine contractions was demonstrated by a number of studies in animals and man.14-17 In view of these findings, a likely explanation for the steroids enhancing IUD retention is that they alter membrane potential in the area underlying the steroid-releasing device thereby blocking the prop-
February 1, 1975 Am. J. Obstet. Gynecol.
agation of rhythmic contractions necessary for expulsion. Once it was established that the 5 synthetic progestins which increased the retention rates of IUD’s did not inhibit mating, their effects on subsequent reproductive processes were assessed. The IUD, control or treated, is always an effective contraceptive in the horn containing it.lR One animal with a 5 per cent norgestrel device in situ at autopsy had 2 small swollen areas in which there was old blood (reported in Table V as possible resorbtions). In no other case was there any evidence of an implantation site in the horn with an IUD in situ. Early studies with MGA devices (not included in Table V since the exact amount of drug could only be estimated at approximately 5 per cent) had indicated that the implantation inhibition was local. Six animals with active devices in situ at autopsy had implantations in the contralateral horn. In these animals however, the number of implantations in the control horn (3.4) was less than might be expected as the average number of corpora lutea from these animals was 9. Unfortunately, in later studies with 5 per cent MGA and with Provera, the number of mated animals with retained devices was so small that no conclusions could be drawn. The more recent studies with 5 per cent norethindrone, norgestrel, and R2323 show a clear-cut effect on implantation in both horns as, at best, only 2 of 8 animals had implants in the control horn. Because the time of expulsion, if known, varied, it was impossible to estimate the effect of devices prior to expulsion. There was evidence that 2 compounds might impair reproductive processes prior to implantation as, in norgestrel and R2323 rats, tubal transport appeared to be altered. Only in animals with R2323 devices, however, was this more pronounced on the treated side. The variability of response within groups and between different experiments on the same treatment group made statistical evaluation impossible; however, definite trends could be seen. All 5 synthetic progestins enhanced device retention, and enough animals retained a treated device and became pregnant in the opposite horn to give support to our basic theory that, although we may not have determined the amounts, there indeed are amounts of these hormones that will enhance retention of IUD’s while not interfering with other highly sensitive reproductive processes. More evidence favoring this hypothesis comes from a small clinical study
Volume Number
121 3
in which women with Provera-releasing tained the devices while the hormone and had normal cyclic menses and incidence of the bothersome side effects and cramping.lvl 2o While these clinical
Hormone-releasing
IUD’s rewas present a very low of bleeding studies with
silicone-rubber
IUD’s
411
Provera are very promising and support the theofurther studies with the other retical concept, synthetic progestins are necessary before we can be sure we have used the ideal steroid at the optimum dose.
REFERENCES
1.
2. 3. 4. 5. 6. 7.
8. 9.
10.
Zipper, J. A., Mendel, M., and Prager, R.: Sixth World Congress on Fertility, Tel Aviv, May, 1968. New York, 1970, Gordon & Brech, p. 154. (Abst.) Tatum, H. J.: AM. J. OBSTET. GYNECOL. 117: 602, 1973. Doyle, L. L., and Clewe, T. H.: AM. J. OBSTET. GYNECOL. 101: 564, 1968. Emmens, C. W.: Hormone Assay, New York, 1950, Academic Press, Inc., pp. 391-418. Hooker, C. W., and Forbes, T. R.: Endocrinology: 41: 159, 1947. Porter, D. G.: J. Reprod. Fertil. 15: 437, 1968. Porter, D. G.: Progesterone: Its Regulatory Effect on the Myometrium, Ciba Foundation Study Group No. 34, London, 1969, J. A. Churchill, Ltd., pp. 79-88. Porter, D. G.: J, Endocrinol. 46: 425, 1970. Csa.po, A.: Mechanism of Action of Steroid Hormones, New York, 1961, Macmillan Publishing Company, Inc., pp. 126-147. Csapo, A.: Progesterone: Its Regulatory Effect on the Myometrium, Ciba Foundation Study Group No.
11. 12. 13. 14. 15. 16. 17. 18. 19.
20.
34, London, 1969, J. & A. Churchill, Ltd., pp. 13-55. Thiersch, J. B., Landa, J. F., and West, T. G.: Am. J. Physiol. 196: 901, 1958. Daniel, E. E.: AM. J. OBSTET. GYNECOL. 80: 229, 1961. Kumar, D., and Barnes, A. C.: AM. J. OBSTET. GYNECOL. 82: 737, 1961. Brinsfield, T. H., and Hawk, H. W.: J, Reprod. Fertil. 16: 129, 1968. Marcus, S. L., Marcus, Cl. C., and Wilson, K. H.: Obstet. Gynecol. 27: 238, 1966. Bengtsson, L. P., and Moawad, A. H.: Lancet 1: 146, 1966. Bengtsson, L. P., and Moawad, A. H.: AM. J. OBSTET. GYNECOL. 98: 957, 1967. Doyle, L. L., and Margolis, A. J.: Science 139: 833, 1963. Stryker, J. C., Doyle, L. L., Clewe, T. H., and Lippes, J.: Advances in Planned Parenthood, VII Excerpta Med. Cong. Ser. 246: 100, 197i. Stryker, J. C.: Personal communication.
LEE
LEE
Little
Rock,
compounds
on
DOYLE Arkansas
Intrauterine devices (IUD’s) containing 0, 5, or 10 per cent by weight of progesterone, melengestrol acetate (MGA), norethindrone, norgestrel, medroxyprogesterone acetate, isoxsuprine, or R2323 or wound with fine copper wire were placed in one or both horns of cycling female rats. All the progestins improved retention of the silicone-rubber IUD’s to some degree. The effects of these compounds on the estrous cycle, mating, ovulation, fertilization, tubal transport, and implantation varied among bhe compounds. Effects were local or systemic, depending upon the amount and the drug. Uterine motility studies showed clearly
could
that 5 per cent MGA decreased uterine motility; however, no be obtained using the other compounds in the same in vivo system.
E F F 0 RT S T 0 DECREASE the failure Kite, expulsion rate, and the occurrence of side effects of intrauterine contraceptive devices were first directed toward modification of size, shape, or material of the IUD. Introduction of the amazingly large number of variations on a scheme, while illustrating imaginative engineering in a limited space, only ameliorated the problems associated with the traditional IUD-it did not solve them. An alternative approach is now being investigated in which the IUD is used as a carrier for materials that are, of themselves, effective contraceptives. Two types of compounds, metals and hormones, have been studied thus far, and both appear to offer great promise. From the Department Gynecology, University Medical Center, Delta
of Obstetrics and of Arkansas Regional Primate
Supported by Population Award No. M72138/ICCR-22 Institutes of Health Grant Received Revised Accepted
for publication July July
results
Among the metals studied by Zipper, Mendel, and Prager,’ copper was shown to be most effective, and it is already being studied using the Tatum T and the Searle Cu 7 in large clinical trials.2 While some investigations have been undertaken in human beings with hormone-impregnated intrauterine devices, only two hormones, progesterone and Provera, have been used extensively. Pilot studies in animals have shown that other hormones could be utilized.3 To determine what other compounds might be equally or more effective, we have undertaken studies in laboratory rodents and primates to screen possible compounds and to compare the efficacy and delineate the characteristics of the most promising ones. The first part of the study deals with the effect that incorporation of 8 different compounds into IUD’s has on retention and reproduction in rodents.
Center.
Council Research and National HD04019. April
consistent
Material
and
methods
The hormones used included Provera,* norgestrel,? norethindrone,$
2, 1974.
4, 1974. 26, 1974.
Reprint requests: Dr. Lee Lee Doyle, Department of Obstetrics and Gynecology, University of Arkansas Medical Center, Little Rock, Arkansas 72201.
tion
405
*Courtesy
of The
Upjohn
tcourtesy Council.
of Wyeth
$Supplied
by The
Co.
Laboratories Population
progesterone, melengestrol
through Council.
the
Popula-
406
February 1, 1975 Am. J, Obstet. Gynecol.
Doyle
Tabie I. Effect of incorporation on retention
Treatment Control 5% MGA 5% Provera 5% progesterone 5 % norethindrone 5% norgestrel 5% R2323 5% isoxsuprine Coppert
Fig. 1. Silicone rubber sleeve after autoclaving,
IUD in polyethylene and ready for insertion.
sleeve,
in
acetate (MGA) ,* R2323,t and is0xsuprine.S In addition, one group of devices were wound with fine copper wire.@ The devices were made by adding 0, 5, or 10 per cent by weight of the compound to liquid silicone rubber (382 medical-grade elastomer, Dow Corning Corp.). An appropriate amount of catalyst was added, and the mixture was injected into polyethylene tubing with a 1.07 mm. inner diameter and allowed to cure at room temperature. The devices were cut into 8 mm. lengths and autoclaved for sterility and simplified removal from the polyethylene sleeve (Fig. 1) . In some early experiments, barium sulfate was included for radiopacity, and various pigments were incorporated so that the devices could be easily differentiated. As we could detect no difference in results between devices with and without pigment or barium, results were not separated. Adult Sprague Dawley rats were used in the experiments. Female rats were checked daily using vaginal lavage to make sure that all were exhibiting normal estrous cycles. At the time of the operation, done under ether the uterine horn was exposed via a anesthesia,
*Courtesy +Courtesy Council. SCourtesy $Supplied
of The Upjohn Co. of Roussel UCLAF of Mead Johnson. by The Population
through
Council.
the
Population
of various by nonmated rats
of IUD’s
hormones
Laparotomy at day 7 after surgery (no. IUD’S retained/no. IUD’S inserted* J
Autopsy at day 14 after surgery (no. IUD’S retained/ no. IUD’s inserted*)
32 26 10 10 12 12 12 30
7/15 16/18 9/10 7/to
4/23 16/16 8/10 6/10 11/12 lo/12 12/12 9/20
I
3/7
l/7
777
2/7
No. Of rats
(46.5) (89) (90) (70) -
6/10
(60)
(17.4) (100) (80) (60) (91.5) (83.5) (100) (45)
*Percentages in parentheses. tRats had a copper-wound 8 mm. segment of Tatum T in one horn and an 8 mm. segment without copper in the opposite horn.
dorsolateral incision, and a small stab wound was made in the antimesometrial wall through which the IUD was inserted. The horn was closed with a single stitch of 6-O silk in order to prevent retrograde expulsion, and the muscle and skin were closed with cotton suture. Laparotomies were done via a single ventromedial incision. All female rats used in mating studies were put into the male rats’ cages and checked daily for presence of a copulation plug or sperm, and if either was found, this was considered day 1 of pregnancy. As the timing of mating, laparotomy, and autopsy will vary, depending upon the parameter being studied, it is included with the results of the individual experiments. Results Retention studies. The effects of 5 per cent compounds on IUD retention in nonmated animals are summarized in Table I. In most cases, laparotomy was done on day 7 ; however, 3 groups were not done at this time and are indicated by the under the day 7 column. In the initial study of each compound, only one device was used per animal, and no matings were attempted. In this first screening, all 5 of the synthetic progestins appeared to be effective in enhancing IUD retention. The retention rate at 2 weeks ranged from 80 to 100 per cent, as compared to 17.4 per cent retention in controls. Progesterone ap-
Volume 121 Number 3
Hormone-releasing
Table II. Effects of mating
on retention
I Treatment
No.
of rats
5% 5% 5% 5%
MGA Provera isoxsuprine norgestrel
20 13 ii 15 10 20
5%
norethindrone
20
5%
R2323
19
MGA
10 10
*These animals fPlug or sperm
were done in 2 groups: was never
recorded;
/
1 Days tion
IUD’s
407
of IUD’s
I
Control
10%
silicone-rubber
from inserto mating 1 7 7 1 1 7 7 7 7 7 7 :
in group
1, lo/IO
however, the animals
parently enhanced retention as did the nonsteroidal uterine relaxant, isoxsuprine, but neither compound appeared to be as effective as any of the synthetic steroids. Copper had no retaining effect. The copper IUD’s were of greater diameter than those of silicone rubber as they were cut from Tatum T’S In order to establish the proper control conditions, IUD’s without copper were made from the nonwrapped portion of the T and inserted into the contralateral horn. The second series of experiments involved mating the animals either immediately or 7 days after IUD insertion to determine if the hormone would inhibit mating or if mating caused expulsion of devices. At the time of laparotomy or autopsy, which varied between 2 and 14 days after breeding, all mated experimental groups had a poorer rate of device retention (Table II). In two sets of animals, one group caged with male rats on day 1 and the other on day 7, 10 per cent MGA devices were placed in one horn, and 90 per cent of these animals retained the devices until the time of autopsy; however, only a few of the animals mated. In several groups of animals, an active device was placed in one horn and a control device was placed in the contralateral horn to see if the effects of the hormone were limited to the horn containing the active IUD or if the drug exerted enough effect systematically to cause control-device retention in the opposite horn. These results can be found in Table III. All the 10 per cent devices had some effect on control-device retention, indicating a probable systemic effect. Norethindrone, R2323, ahd
No. 1 14 13 10 10 5
mated (%I
Days I
(70) (100) i55)Y (67) (50)
at ,
7to 14 7to14 7to 14 5to14 7to14 2 to 4 7 to 14 2 to4 7to17 2 to 4 7 to 14 14 14
18’+“5(90) I\
pregnant autopsy
Retention (no. IUD’s retained/ no. IUD’s ;nrcvtunl) _‘.“-’ -“-/
I
17’+3’t(85) 18”“t (96) 1 (10) 3 (30)
3/20 2/13 9/18 10/15 2/10 6/10 8/10 3/10 8/10 6/9 6/10 9/10 9/10
mated; in group 2, O/8 mated. were pregnant at autopsy.
Table III. retention
Effect of hormone-releasing of contralateral control IUD’s
Treatment 5%
MGA Control 570 Provera Control 570 norethindrone Control 570 norgestrel Control 5% R2323 Control 1070 MGA Control 1070 norethindrone Control 1070 norgestrel Control 1070 R2323 Control
No. of rats 5
Autopsy (days after insertion) 7to14
8
14
10
14
10
14
10
14
10
14
10
14
10
14
10
14
IUD’s
on
Devices retained/ devices inserted 4/5 l/5 6/8 5/8 6/10 o/10 4/10 3/1O 9/10 3/10 lO/lO lo/lo 8/1O 4/10 8/10 lO/lO lO/lO 4/10
MGA at the 5 per cent level appeared to act only locally, but 5 per cent Provera and norgestrel caused retention of some nonmedicated devices. Reproductive studies. All doses of all hormones caused aberrant vaginal cytology; however, only the 10 per cent MGA and isoxsuprine seemed to inhibit mating (see Table II). Mating was recorded as having occurred when either a copulation plug or sperm was found, but apparently, these were missed on several occasions because, at autopsy, animals that had shown no evidence of mating were found to be pregnant.
408
Februaly Am. J. Obstrt.
Doyle
Table IV. Effect of mating
on retention
Animals
de&es
Unmated f% J
Mated cc/r )
MGA Provera
100 80
50
norethindrone norgestrel R2323
91.5 83.5 100
55 70 64
Treatment
5% 5% 5% 5% 5%
retaining
of IUD’s
60
The retention rate of devices in the mated animals was less than that in the nonmated rats for all compounds at 5 per cent (Table IV). With the exception of MGA and Provera, the compounds apparently inhibited implantation to a significant degree in the control as well as the treated horns (Table V) . In all cases, the IUD, either control or active, was an effective contraceptive as only one animal showed anything remotely resembling an implantation when an IUD was present. In this single animal, there were 2 areas that could have been resorbing fetuses. To differentiate between a failure in the process of implantation and disruption of an earlier event such as fertilization or cleavage, a series of experiments was done to recover ova from the uterus and tubes. In these experiments, no treatment was given to control animals so values on normal rates of fertilization, cleavage, and recovery could be established for female rats in our colony. The treated animals had devices inserted and were put with males 7 days later. Once again, mating apparently increased device expulsion as all treated groups had 10 mated animals but only 3 to 6 had devices upon recovery 2 to 5 days after mating. The small number of animals in each group makes statistical comparisons among the treatment groups impossible, but it appears from the data in Table VI that all the hormones except norethindrone may interfere with either fertilization or development of ova to some extent. Only R2323 seems to have more effect on the treated side than on the contralateral control side. The incidence of abnormalities including nonfertilization or cleavage was increased on both sides of all treated groups. Comment No intrauterine contraceptive device can be effective if it is not well retained; thus, in our studies aimed at establishing which compounds might be promising for clinical trials in human beings, re-
1, 1975 C:ynecol.
tention was the first and most important criterion to be tested. The rat was a suitable animal for retention studies as control IUD’s are very poorly retained. Three compounds-progesterone, isoxsuprine, and copper-caused an increased rate of retention compared to controls (14.3 to 60 per cent compared to 17.4 per cent) after two weeks of use but were far less effective than any of the 5 synthetic progestins, which achieved 80 to 100 per cent retention for the same time period. Most experimental groups showed a decline in the retention rate following laparotomy or mating, the latter effecting the more dramatic decrease. We considered that the difference in retention could be due to one of two causes; either the compound was being released in such quantities that there was no longer enough remaining in the device to maintain an effective release rate, or the uterine horns were responding to the trauma of laparotomy with frequent intense contractions, and the amount of hormone being released was insufficient to counteract them. In one experiment, designed to assess the importance of each factor, there were 3 experimental groups of 8 rats per group. All animals had a 5 per cent MGA device inserted in one horn, after which group 1 had no further treatment, group 2 underwent laparotomy on day 7, and group 3 was placed with male rats on day 7. At autopsy on day 14, groups 1 and 2 had 100 per cent retention, and group 3 had 90 per cent. An additional group of 10 animals, which at laparotomy on day 7 showed 8 of 10 devices in situ, was then mated on day 7, and at autopsy one week later, only 2 animals had retained devices. From these data, mating or laparotomy alone seemed to have little effect on expulsion rate; whereas, a combination of the two virtually negated the effect of the hormone. The rate of retention was not as high in other studies of mated animals treated with 5 per cent MGA or, in fact, with any of the synthetic steroids so one can probably assume that mating does influence expulsion. The 10 per cent MGA devices were not expelled in those few animals which mated. From these studies, we concluded that all the synthetic steroids were effective in enhancing retention of silicone-rubber IUD’s but that, at the 5 per cent level, they were not as effective in mated as in nonmated animals. The vaginal cytology from all animals seemed to indicate aberrant estrous cycles; however, in only one group of animals with a 5 per cent Provera device was there any decrease in the num-
Volume Number
121 3
Table
V. Effect
Hormone-releasing
of hormone-releasing
IUD’s
No. of rats
No. with de#vices in
18
2
Control
No. with devices out
MGA
10
2
5%
Provera
9
3
5 r/r
norethindrone
10
8
5%
norgestrel
10
8
5%
R2323
10
6
MGA
10
9
8 6 2 2 4 10%
1 *These
two
areas
could
Table VI. Effect ova recovery
have
been
IUD’s
on
No. of rats retaining devices with ova in
Treatment Control* Control device 5% Provera 5% norgestrel 5% norethindrone 5% R2323 *These
animals
19 5 10 10 10 10 had
no device
409
Average no. of implants 5.1 0 3.2 0 1 0 6.0 2* 5.0 0 1.3
side
Control
Average no. of corpora Average no. lutea of implants 7.8 4.0 6.8 9.0 4.0 6.5 6.0 8.0 8.0 5.0 7.0 -
5.1 8.0 5.7 5.0 4.0 7.5 8.0 7.0 5.0 8.0 5.3 -
side Average no. of corpora lutea 7.2 8.0 8.0 5.0 6.0 8.5 8.0 10.0 7.0 8.0 6.0 -
resorptions.
of hormone-releasing
No. of No. of rats rats with bred 1 device
No. with implants 0 16 1 7 1 1 2 1 1 1 1 3 0 0
16 5%
IUD’s
on implantation IUD
Treatment
silicone-rubber
Treated Control horn or horn or I tube I tube
0 6 6 3 6
19 4 3 3 3 1
19 4 4 4 3 4
inserted.
ber of rats mating. The changes in shed of vaginal cells may simply reflect a local “downstream” effect of the progestin-bearing uterine fluid that bathes the vagina rather than indicating any changes in ovarian steroid secretion. The ability of locally applied hormones to alter vaginal cytolo.gy has been demonstrated by use of estrogen on castrate mice.4 An alternative explanation is that small amounts of progestins do get into the systemic circulation, and while this low level can cause a change in vaginal cytology, it is not sufficient to inhibit mating. It appeared in our earlier studies with MGA that the 5 per cent devices acted locally on the uterus, while the 10 per cent devices affected all reproductive processes on a systemic level. In a sense, this would be analogous to the situation encountered in Hooker and Forbes’” assay for progestrone where
small amounts of the hormone caused cellular changes only in the area of the intraluminally injected progesterone, while larger amounts affected the contralateral horn as well. A possible explanation for our results with MGA is that small amounts are bound or metabolized in the uterus and only when in excess of receptor sites find their way into the systemic circulation. Another explanation is that all the MGA circulated systemically, but the amount was insufficient to interfere with reproductive parameters. In this latter case, if the progestin is acting systemically, it should be capable of eliciting the same reaction in the control as in the treated horn. If the amount being released is sufficient to cause device retention in the treated horn, it should cause device retention in the control horn. Although the retention rate varied with the compound used, in no case in which animals had a 5 per cent device in one horn and a control device in the other was the retention rate of the control device as good as that of the treated device; however, 5 per cent Provera and norgestrel did not show much difference between horns. In studies with synthetic steroids other than MGA, the distinction between local and systemic effects was not so clear cut. It may be that larger amounts of these compounds are released or that the same amount is released but less is necessary to obtain the same biologic end point. Studies on actual in vivo release rates are being done at present. In considering the question of local or systemic effect, one must clearly define the biologic end point in question. None of the 5 per cent
410
Doyle
steroids studied inhibited mating; however, all appeared to have some systemic effect on other reproductive parameters that may well be overcome by giving less of the compound. In the retention studies, the action appeared to be only local which contradicts the data from reproductive studies where, in some cases, there was evidence for systemic effect. From this, we might postulate that there is a local concentration of hormone exceeding the level found in the general circulation. If we further assume that more progestin is needed to inhibit uterine contractions than to interfere with tubal transport and implantation, we can explain the apparently contradictory results. We had hoped to accumulate additional evidence for the local-effect theory with in vivo uterine motility studies. These studies were done with several compounds, but findings were variable, inconsistent, and inconclusive in all but the 5 per cent MGA studies where a decrease in frequency and amplitude of uterine contractions was found. After much thought and discussion concerning possible sources of error and variance and how to combat them, we realized that an entirely new project would be necessary and that the expenditure of time and effort was not warranted by the amount of data we could gain. This decision left us unable to point directly to demonstrable decreases in uterine motility as the cause of enhanced IUD retention; however, our indirect evidence supports this explanation as does the work of others. The ability of intrauterine progesterone to act locally to affect uterine motility in animals has been clearly demonstrated by Porter.+* In addition, studies by Csapos and others, summarized at a conference concerning the effects of progesterone on the myometrium, have shown that the human myometrium underlying the placenta contains higher levels of progesterone than are found in the surrounding myometrium.10 Since progestins have been shown to increase the resting transmembrane potential, (I* 11-12 they could interfere with the conduction of impulses and, thereby, block the rhythmic contractions necessary for expulsion of a fetus or a foreign body. That the IUD can increase uterine contractions was demonstrated by a number of studies in animals and man.14-17 In view of these findings, a likely explanation for the steroids enhancing IUD retention is that they alter membrane potential in the area underlying the steroid-releasing device thereby blocking the prop-
February 1, 1975 Am. J. Obstet. Gynecol.
agation of rhythmic contractions necessary for expulsion. Once it was established that the 5 synthetic progestins which increased the retention rates of IUD’s did not inhibit mating, their effects on subsequent reproductive processes were assessed. The IUD, control or treated, is always an effective contraceptive in the horn containing it.lR One animal with a 5 per cent norgestrel device in situ at autopsy had 2 small swollen areas in which there was old blood (reported in Table V as possible resorbtions). In no other case was there any evidence of an implantation site in the horn with an IUD in situ. Early studies with MGA devices (not included in Table V since the exact amount of drug could only be estimated at approximately 5 per cent) had indicated that the implantation inhibition was local. Six animals with active devices in situ at autopsy had implantations in the contralateral horn. In these animals however, the number of implantations in the control horn (3.4) was less than might be expected as the average number of corpora lutea from these animals was 9. Unfortunately, in later studies with 5 per cent MGA and with Provera, the number of mated animals with retained devices was so small that no conclusions could be drawn. The more recent studies with 5 per cent norethindrone, norgestrel, and R2323 show a clear-cut effect on implantation in both horns as, at best, only 2 of 8 animals had implants in the control horn. Because the time of expulsion, if known, varied, it was impossible to estimate the effect of devices prior to expulsion. There was evidence that 2 compounds might impair reproductive processes prior to implantation as, in norgestrel and R2323 rats, tubal transport appeared to be altered. Only in animals with R2323 devices, however, was this more pronounced on the treated side. The variability of response within groups and between different experiments on the same treatment group made statistical evaluation impossible; however, definite trends could be seen. All 5 synthetic progestins enhanced device retention, and enough animals retained a treated device and became pregnant in the opposite horn to give support to our basic theory that, although we may not have determined the amounts, there indeed are amounts of these hormones that will enhance retention of IUD’s while not interfering with other highly sensitive reproductive processes. More evidence favoring this hypothesis comes from a small clinical study
Volume Number
121 3
in which women with Provera-releasing tained the devices while the hormone and had normal cyclic menses and incidence of the bothersome side effects and cramping.lvl 2o While these clinical
Hormone-releasing
IUD’s rewas present a very low of bleeding studies with
silicone-rubber
IUD’s
411
Provera are very promising and support the theofurther studies with the other retical concept, synthetic progestins are necessary before we can be sure we have used the ideal steroid at the optimum dose.
REFERENCES
1.
2. 3. 4. 5. 6. 7.
8. 9.
10.
Zipper, J. A., Mendel, M., and Prager, R.: Sixth World Congress on Fertility, Tel Aviv, May, 1968. New York, 1970, Gordon & Brech, p. 154. (Abst.) Tatum, H. J.: AM. J. OBSTET. GYNECOL. 117: 602, 1973. Doyle, L. L., and Clewe, T. H.: AM. J. OBSTET. GYNECOL. 101: 564, 1968. Emmens, C. W.: Hormone Assay, New York, 1950, Academic Press, Inc., pp. 391-418. Hooker, C. W., and Forbes, T. R.: Endocrinology: 41: 159, 1947. Porter, D. G.: J. Reprod. Fertil. 15: 437, 1968. Porter, D. G.: Progesterone: Its Regulatory Effect on the Myometrium, Ciba Foundation Study Group No. 34, London, 1969, J. A. Churchill, Ltd., pp. 79-88. Porter, D. G.: J, Endocrinol. 46: 425, 1970. Csa.po, A.: Mechanism of Action of Steroid Hormones, New York, 1961, Macmillan Publishing Company, Inc., pp. 126-147. Csapo, A.: Progesterone: Its Regulatory Effect on the Myometrium, Ciba Foundation Study Group No.
11. 12. 13. 14. 15. 16. 17. 18. 19.
20.
34, London, 1969, J. & A. Churchill, Ltd., pp. 13-55. Thiersch, J. B., Landa, J. F., and West, T. G.: Am. J. Physiol. 196: 901, 1958. Daniel, E. E.: AM. J. OBSTET. GYNECOL. 80: 229, 1961. Kumar, D., and Barnes, A. C.: AM. J. OBSTET. GYNECOL. 82: 737, 1961. Brinsfield, T. H., and Hawk, H. W.: J, Reprod. Fertil. 16: 129, 1968. Marcus, S. L., Marcus, Cl. C., and Wilson, K. H.: Obstet. Gynecol. 27: 238, 1966. Bengtsson, L. P., and Moawad, A. H.: Lancet 1: 146, 1966. Bengtsson, L. P., and Moawad, A. H.: AM. J. OBSTET. GYNECOL. 98: 957, 1967. Doyle, L. L., and Margolis, A. J.: Science 139: 833, 1963. Stryker, J. C., Doyle, L. L., Clewe, T. H., and Lippes, J.: Advances in Planned Parenthood, VII Excerpta Med. Cong. Ser. 246: 100, 197i. Stryker, J. C.: Personal communication.
