CONTRACEPTION
EFFECT'SOF VARIOUS IUDsONTHECOYF'OSITION OFCERVICALMUCUS B jonsctm. P-M Landgren and P Eneroth Department of Obstetrics and Gynecology, Earolinska Hospital, S-1 040IStockholm. Sweden
ABSTRACT The mfiuence cd three different Intrauterine devices on the compos~tic~ncd cervical mucus was studied. The amount of mucin, albumin and immunoglobulin G was estimated. After the insertion of an inert IlID, a decrease in mucin was observed. During copper-Km use the content of mucin. albumin and IgG was increased in cervical mucus, while weight was not affected. In the levonorgestrel-IEID users, ovulatitm was inhibited in 2 out of 8 women. Mucus weight was increased. The amounts of mucin, albumin and IgGwere not changed. In an in vitro experiment the effect of copper-JUDs on autocrxidaticmd cholesterol was studled. There was an extensive cnnverslon of cholesterol but addition of albumin quenched the oxidation of cholesterol. It is suggested that the increased secretion of albumin induced by copper-Il:D users may offer protection against copper-induced cell damage.
Submitted for publication November 1, 1990 Accepted for publication February 25, 1991
MAY 1991 VOL. 43 NO. 5
447
CONTRACEPTION
INTRODUCTION In a previous study we have demonstrated that the ccjmposition of cervical mucus differed between women using copper IEJDsand normal controls, Comparatively larger amounts of lmmunoglobulin G, albumin and complement (3 were found in the IlJil-users LI 1. It was assumed that copper induced an increase of the local transudation of mucus. To eiucldate if the changes in the composition of cervical mucus was an effect of the copper or of the foreign body itself, this study was undertaken, Since gestagens are known to affect the formation of cervical mucus, we have al::) included the use of a Ievonorgestrel-loaded KJD in the study i 2, 3 1
MATERIALS AND METHODS
Ethical aspects The study was aptlroved by the ethical committee of the Kafolinska Hospital.
Clinical Material Twenty-seven apparently healthy women participated in the study, ail with a history 01 regular menstruations and a cycle length of 26 to 32 days, None had used an IUD or steroidal contraceptive for at least three months before entering the study. AlJ women were parous ! 1-3 children I and between 26 and 40 years of age.
Study design The women were randomly assigned to three treatment groups of 9 subjecrs each. The first group had an empty skeleton of a Nova-T (inert ItJIll Inserted. ‘The second group received a copper-containing Nova-T, and the third group a ievonorgesrrel-loaded Nova-T fall IUDs were fabricated bv Leira, Finland). During a pretreatment control cycle and after three months with the IUD in situ, blood samples and cervical mucus were collected daily from day 10 of the menstrual cycle until estimated day of ovulation. 0ne week after estimated ovulation. three additional daily blood samples were collected. After centnfugatlon of the blood sample; serum was stored at -20°C until anaIy.zed. Cervical mucus was collected in baby-ieeding tubes, which were immediately closed and stored at -20°C until analyzed. The day of ovulation was estimated from assays of urinary LH @Qvustickl, estradiol in serum and by clinical evaluation of the mucus.
Laboratory methods Commercial radioimmunoassay kits purchased from Radioimmunoassay Sysrems Lahora~orry G.,os Angeles, CA,‘USAI were used for estradml and 448
MAY 1991 VOL. 43 NO. 5
CONTRACEPTION
gr(ge:tertjne measurements. LH was asca:?ed using kits from Diagnostic Products Inc. rLos Angeles, CA.USAr.Serum concentration of levonorgestrel was determined in the 819 women who received a levonorgestrel IUDt4 I The cervical mucus samples were analyzed for immunogiobulin G, albumin and muctn using tmmunonephelometrtc methods. The soluble proteins were separated from cervtcal mucus using continuous extraction with sodium chlortde of high iontc strength !0,66Mt on an inert chromatographic bed ( 1 ml of washed seasand covered with glass wooPackard Instrument Co., Zurich. Switzerland 1~The procedure has been demonstrated to yield the proteins in the NaClfraction, as shown previously !5i. For quantitation of the soluble proteins, albumins and IgG, immunonephelometrtc methods w-ere used !hallestad LSX 240 instrument nephelometer, Chaska, Mtnnesota. USA!.Commeraally avatlable antibodies !Dako, Copenhagen. Denmark I were used. Reference preparations of albumin and IgG were purchased from Gahi !Stockholm, Sweden) r5 1,For quantitation of mucina method developed by us was used (6). Lectin from Helix Pomatia obtained from Pharmacia Phine Chemicals Co. (Uppsala. Sweden) was used. Leclin reacted with N-acetyl-galactosamine \Sigma, St. Louis, USA! residues, The reactton u-as enhanced by high polyethylene-glycol concentrations IKeho. Spinga. Sweden) and the complex formed was quantitated by light scattering using a nephelometer, As a standard, muctn from bovine submaxtlhar>- gland type I obtained from Sigma iSt. Louis. US.-Iiwas used. The intra- and interassa)- variations for albumin was 11~10%(SD.n=5Or and 232 15%{SD:n=35t. and for lectin j4z2Ox t’SD:n=471 and 64t48% (SD: n= 19 I, respectively. For 1gGthe intraassay variation was 17x2 1%{SD;n=50! and t.he interassay variatron was 1%4,4X iSD:n-50!. Xutooxtdatton of l$C-cholesterol was studted in TRIS-HClbuffer with saline contatning three copper IUDs.Following Folch extraction with chloroform/ methanol 2t 1 iv/v) and solvent. evaporation, the reaction products were separated by thin layer chromatography using petroleum ether/ether O/l !v/v! as solvent. Reaction products were localized by autoradiography and each compound was extracted from the gel by chloroform/methanol l/ 1 1v/v) and its content of 14Cwas determined. The identity of the compounds formed during cholesterol autooxidation has been described before (7 I Statistical
methods
The statistical methods used were Wilcoxon matched pairs signed-rank test, the Mann-Whitney U-test and Eruskai-Wallis one-way analysis of variance (8).
MAY 1991 VOL. 43 NO. 5
449
CONTRACEPTION
Estradiol (pmot/ll
Inert
IUD
Copper-IUD
III
before
RI
after
Levonorgestrei-IUD
Fig. 1 Mean serum estradiol concentration (pmol/l) and standard deviation before and after insertion of different IUDs (9 women in each group) Estradiol levels determined during three days (LH peak day and the two preceding days) were averaged for each woman.
450
MAY 1991 VOL. 43 NO. 5
CONTRACEPTION
RESULTS
The estradiol levels in the three groups of women are shown in Figure 1, There were no significant differences neither in the control period nor after insertion of the IUDbetween the three groups of women. The progesterone levels before and after insertjon of the IUDsare shown in Figure 2. A mean of the three luteal phase values was calculated for each individual. All women ovulated during the control period when the characterization of a normal ovulatory cycle by Landgren et al. was used f,1 1 1. In the third treatment month, 2 of 8 women using the Ievonorgestrelreleasing IUDwere anovulatory. The remainmg 6 women ovulated but the progesterone levels were significantly lower than in the control month 1p 0.00 1 I and when compared to those of the women usmg an Inert IUD Ip~li.051or acctpper-II!D!,pcll.lfli.
nmol/f
cl q After
Refore
Inert
I LID
Copper -I LID
Levonorgestrei-IUD
Fig. 2. Mean serum progesterone (nmol/li concentration and standard deviation before and after insertion of diiferent 117Ds.
MAY 1991 VOL. 43 NO. 5
451
rJl
5
E
B
!k!
0 1293
0 4050
0 08
!O 18871 ’
0.1860
(0.29841 \n.s
(0 5009;) (0.2 147! ‘pi0.05’
0.092
0.044
0.1680
0.058
p
EFFECT'SOF VARIOUS IUDsONTHECOYF'OSITION OFCERVICALMUCUS B jonsctm. P-M Landgren and P Eneroth Department of Obstetrics and Gynecology, Earolinska Hospital, S-1 040IStockholm. Sweden
ABSTRACT The mfiuence cd three different Intrauterine devices on the compos~tic~ncd cervical mucus was studied. The amount of mucin, albumin and immunoglobulin G was estimated. After the insertion of an inert IlID, a decrease in mucin was observed. During copper-Km use the content of mucin. albumin and IgG was increased in cervical mucus, while weight was not affected. In the levonorgestrel-IEID users, ovulatitm was inhibited in 2 out of 8 women. Mucus weight was increased. The amounts of mucin, albumin and IgGwere not changed. In an in vitro experiment the effect of copper-JUDs on autocrxidaticmd cholesterol was studled. There was an extensive cnnverslon of cholesterol but addition of albumin quenched the oxidation of cholesterol. It is suggested that the increased secretion of albumin induced by copper-Il:D users may offer protection against copper-induced cell damage.
Submitted for publication November 1, 1990 Accepted for publication February 25, 1991
MAY 1991 VOL. 43 NO. 5
447
CONTRACEPTION
INTRODUCTION In a previous study we have demonstrated that the ccjmposition of cervical mucus differed between women using copper IEJDsand normal controls, Comparatively larger amounts of lmmunoglobulin G, albumin and complement (3 were found in the IlJil-users LI 1. It was assumed that copper induced an increase of the local transudation of mucus. To eiucldate if the changes in the composition of cervical mucus was an effect of the copper or of the foreign body itself, this study was undertaken, Since gestagens are known to affect the formation of cervical mucus, we have al::) included the use of a Ievonorgestrel-loaded KJD in the study i 2, 3 1
MATERIALS AND METHODS
Ethical aspects The study was aptlroved by the ethical committee of the Kafolinska Hospital.
Clinical Material Twenty-seven apparently healthy women participated in the study, ail with a history 01 regular menstruations and a cycle length of 26 to 32 days, None had used an IUD or steroidal contraceptive for at least three months before entering the study. AlJ women were parous ! 1-3 children I and between 26 and 40 years of age.
Study design The women were randomly assigned to three treatment groups of 9 subjecrs each. The first group had an empty skeleton of a Nova-T (inert ItJIll Inserted. ‘The second group received a copper-containing Nova-T, and the third group a ievonorgesrrel-loaded Nova-T fall IUDs were fabricated bv Leira, Finland). During a pretreatment control cycle and after three months with the IUD in situ, blood samples and cervical mucus were collected daily from day 10 of the menstrual cycle until estimated day of ovulation. 0ne week after estimated ovulation. three additional daily blood samples were collected. After centnfugatlon of the blood sample; serum was stored at -20°C until anaIy.zed. Cervical mucus was collected in baby-ieeding tubes, which were immediately closed and stored at -20°C until analyzed. The day of ovulation was estimated from assays of urinary LH @Qvustickl, estradiol in serum and by clinical evaluation of the mucus.
Laboratory methods Commercial radioimmunoassay kits purchased from Radioimmunoassay Sysrems Lahora~orry G.,os Angeles, CA,‘USAI were used for estradml and 448
MAY 1991 VOL. 43 NO. 5
CONTRACEPTION
gr(ge:tertjne measurements. LH was asca:?ed using kits from Diagnostic Products Inc. rLos Angeles, CA.USAr.Serum concentration of levonorgestrel was determined in the 819 women who received a levonorgestrel IUDt4 I The cervical mucus samples were analyzed for immunogiobulin G, albumin and muctn using tmmunonephelometrtc methods. The soluble proteins were separated from cervtcal mucus using continuous extraction with sodium chlortde of high iontc strength !0,66Mt on an inert chromatographic bed ( 1 ml of washed seasand covered with glass wooPackard Instrument Co., Zurich. Switzerland 1~The procedure has been demonstrated to yield the proteins in the NaClfraction, as shown previously !5i. For quantitation of the soluble proteins, albumins and IgG, immunonephelometrtc methods w-ere used !hallestad LSX 240 instrument nephelometer, Chaska, Mtnnesota. USA!.Commeraally avatlable antibodies !Dako, Copenhagen. Denmark I were used. Reference preparations of albumin and IgG were purchased from Gahi !Stockholm, Sweden) r5 1,For quantitation of mucina method developed by us was used (6). Lectin from Helix Pomatia obtained from Pharmacia Phine Chemicals Co. (Uppsala. Sweden) was used. Leclin reacted with N-acetyl-galactosamine \Sigma, St. Louis, USA! residues, The reactton u-as enhanced by high polyethylene-glycol concentrations IKeho. Spinga. Sweden) and the complex formed was quantitated by light scattering using a nephelometer, As a standard, muctn from bovine submaxtlhar>- gland type I obtained from Sigma iSt. Louis. US.-Iiwas used. The intra- and interassa)- variations for albumin was 11~10%(SD.n=5Or and 232 15%{SD:n=35t. and for lectin j4z2Ox t’SD:n=471 and 64t48% (SD: n= 19 I, respectively. For 1gGthe intraassay variation was 17x2 1%{SD;n=50! and t.he interassay variatron was 1%4,4X iSD:n-50!. Xutooxtdatton of l$C-cholesterol was studted in TRIS-HClbuffer with saline contatning three copper IUDs.Following Folch extraction with chloroform/ methanol 2t 1 iv/v) and solvent. evaporation, the reaction products were separated by thin layer chromatography using petroleum ether/ether O/l !v/v! as solvent. Reaction products were localized by autoradiography and each compound was extracted from the gel by chloroform/methanol l/ 1 1v/v) and its content of 14Cwas determined. The identity of the compounds formed during cholesterol autooxidation has been described before (7 I Statistical
methods
The statistical methods used were Wilcoxon matched pairs signed-rank test, the Mann-Whitney U-test and Eruskai-Wallis one-way analysis of variance (8).
MAY 1991 VOL. 43 NO. 5
449
CONTRACEPTION
Estradiol (pmot/ll
Inert
IUD
Copper-IUD
III
before
RI
after
Levonorgestrei-IUD
Fig. 1 Mean serum estradiol concentration (pmol/l) and standard deviation before and after insertion of different IUDs (9 women in each group) Estradiol levels determined during three days (LH peak day and the two preceding days) were averaged for each woman.
450
MAY 1991 VOL. 43 NO. 5
CONTRACEPTION
RESULTS
The estradiol levels in the three groups of women are shown in Figure 1, There were no significant differences neither in the control period nor after insertion of the IUDbetween the three groups of women. The progesterone levels before and after insertjon of the IUDsare shown in Figure 2. A mean of the three luteal phase values was calculated for each individual. All women ovulated during the control period when the characterization of a normal ovulatory cycle by Landgren et al. was used f,1 1 1. In the third treatment month, 2 of 8 women using the Ievonorgestrelreleasing IUDwere anovulatory. The remainmg 6 women ovulated but the progesterone levels were significantly lower than in the control month 1p 0.00 1 I and when compared to those of the women usmg an Inert IUD Ip~li.051or acctpper-II!D!,pcll.lfli.
nmol/f
cl q After
Refore
Inert
I LID
Copper -I LID
Levonorgestrei-IUD
Fig. 2. Mean serum progesterone (nmol/li concentration and standard deviation before and after insertion of diiferent 117Ds.
MAY 1991 VOL. 43 NO. 5
451
rJl
5
E
B
!k!
0 1293
0 4050
0 08
!O 18871 ’
0.1860
(0.29841 \n.s
(0 5009;) (0.2 147! ‘pi0.05’
0.092
0.044
0.1680
0.058
p
