CONTRACEPTION
SERUM LIPID VARIATIONS IN WOMENON COPPER INTRAUTERINEDEVICES, TRIPHASIC ORAL CONTRACEPTIVES ANDMONTHLY INJECTABLE CONTRACEPTIVES E. Kesserii, H. Albornoz, M. Diaz, R. Socolsky 1st Gynaecological Clinic University of Buenos Aires Argentina
ABSTRACT This is a longitudinal, prospective, comparative controlled study. Patient selection excluded smokers, overweighted, hypertensive, and aged ) 34 years. 168 women were distributed by random list into four groups; A) fitted with inert IUD (control); B) with Nova T (Cu-IUD);Cl received triphasic gestodene (OC);and D) monthly norethisterone enanthate 50 mg + estradiol valerate 5 mg (injectable). Serum lipid patterns were assessed at the beginning of contraceptive use and at 1, 6 and 12 months. The results of the Cu-IUDgroup did not show any difference with the control group. The OCgroup showed an increase of HDLand a significant lowering of LDLcholesterol at 6 months but this reverted by 12 months. On the other hand, in the injectable group, a sharp decrease of total lipids, total cholesterol and LDLwas observed from the first month, lasting throughout 12 months, yielding significant difference with controls and also with pfe-treatment values. In sum, the monthly injectable showed quite favorable patterns that would deserve further study, especially in view of the scarcity of hitherto available information concerning this formulation.
Address for Correspondence: la. C&edra de Ginecologia,Hospital de Clinicas Cordoba 235 1, ( 11201 Buenos Aires, Argentina Submitted for publication September 17, 1990 Accepted for publication July 10, 1991
SEPTEMBER 1991 VOL. 44 NO. 3
235
CONTRACEPTION
INTRODUCTION The relationship between lipid metabolism disturbances and the increase of the risk of arterial thromboembolism is nowadays almost universally accepted (1,2,3). Likewise, it is generally accepted that contraceptive steroids may affect, in several ways, the lipid metabolism and thus play a role in increasing thromboembolism risk I4,5,b). Nevertheless, this point is still under dispute in both physiopathological and epidemiological aspects (7,8X However, it became evident that synthetic progestogens may effect a lowering of HDLcholesterol, thus producing an imbalance of the HDL/LDL ratio (9,101, which has been directly related to subsequent atheroma and connected accident (3,4,10). Estrogens would exert an opposite action upon this mechanism ( 11). That gave momentum to the pill-lipid research with the result of a great many publications dealing with various aspects (formulations, doses, time of use, consumers’ status, etc.1 that may influence the effects of OCon lipids (3,12,13,14,15). Contrasting with the large amount of work done with oral contraceptives, there is a surprising scarcity of information using injectable preparations ( lb, 171,in spite of the fact that pharmacokinetics of oral and parenteral contraceptive steroids are quite different ItSI. Besides, although IUDshave been used in some studies as controls for OCs(5,19,20).there is no information so far whether copper-releasing IUDsmay exert any influence, e.g. whether there is any difference between copper IUDs and inert IUDs. In the present study, the behavior of serum lipid parameters was assessed in a prospective, randomized controlled design over one year’s duration. Inert Lippes IUDs were used as controls to compare the effects of copper IUDs,triphasic gestogen OCs, and monthly injectable progestogen plus estrogen contraceptives. MATERIAL 200 women from our family planning outpatient clinic were initially allotted by random list to one of the four contraceptive regimens. The criteria for the inclusion were: age between 20 and 34 years; proven fertility; no systemic contraceptives for the last b months; no smoking; no obesity &ss than +20X of ideal body weight); no hypertension, and with normal liver function as well as glucose tolerance tests. Only those patients who completed one year of contraception were used for the evaluation of results (n- 1661. The final integration and characteristics of the four groups are shown on Table I.
236
SEPTEMBER 1991 VOL. 44 NO. 3
CONTRACEPTION
TABLEI
STUDIED MATERIAL A
GROUPS
Contraceptive
, agent I ! I ncases age(years)x(S.D.) parityY(S.D.1 abortionsY?u(S.DJ
1 av.sexual frequency/ , week 51 (SD.)
B CU-IUD
Controls
Nova-T
LippesC
47
’
48
c TriphasicOC Gestodenet Ethinylestradiol (SH415G)* 33
I
Monthly D injectable Morethisterone enanthate 5Dmg + Estradiol valet-ate Smg (SH329AB) l 40
27.4 (4.8)
26.8 (5.6) 1
26.1 (5.4)
26.4t5.5)
1.5(I.21
1.7(1.4)
1.5(1.4)
1.6ll.3)
0.9 (1.1)
0.7(1.0)
0.4(0.6)
0.8 (1.1)
2.9(1.1)
2.8(1.3)
2.5(1.4)
2.8(1.2)
METHODS In each patient, four blood samples were taken, i.e. at the beginning of contraception, at one month on contraception, at six months on contraception, and at twelve months on contraception. The blood samples were taken between the 1st and 7th days of the cycle. All subjects were instructed to fast overnight and refrain from alcohol for three days pnor to giving the blood samples. In each blood sample the following parameters were assessed: total serum lipids (2 11, total cholesterol (221, HDL cholesterol, LDL cholesterol G$, and triglycerides (22). Multiple-comparison statistical analyses adjusted for the number of comparisons were performed by using Bonferroni’s procedure (241, with a level of significance of 0.05. Each group was compared with the control group at 1, 6 and 12 months after starting treatments. Also, within-group comparisons were performed against the own pf e-treatment values. RESULTS Fig. 1 shows the evolution of total sefum lipid values in the four studied groups. There were no differences between the values of the control group (A), the copper IUD group (B), and the OC group 0. On the other hand, the injectable group (D) showed a sharp decrease with high statistical significance when compared to the control group, and also with statistical significance against their own pfe-treatment values. This statistically significant lowering continued at six months of observation but lost statistical significance at twelve months, thus narrowing to the control group values. On Fig. 2 the evolution of total cholesterol values are shown. Here, again, the evolution of the copper IUDgroup and the OCgroup was similar to the controls with mean values and 95%confidence limit values almost totally overlapping. However, the
SEPTEMBER 1991 VOL. 44 NO. 3
237
CONTRACEPTION
mg/dl
CONT.
1 MO
6 MO
ugure 1. RESULTS OF TOTAL LIPIDS for tbe nunbor of comparisons).
12Mo
(Confidoaccintervals 95!5, adjusted
Legends A -
Control (inert IUD)group = Copper IUDgroup = Oral triphaslc group = MOnthly injectable group = p ,. 0.05 vs. control group ** = p s 0.01 vs. control group 0 = p
SERUM LIPID VARIATIONS IN WOMENON COPPER INTRAUTERINEDEVICES, TRIPHASIC ORAL CONTRACEPTIVES ANDMONTHLY INJECTABLE CONTRACEPTIVES E. Kesserii, H. Albornoz, M. Diaz, R. Socolsky 1st Gynaecological Clinic University of Buenos Aires Argentina
ABSTRACT This is a longitudinal, prospective, comparative controlled study. Patient selection excluded smokers, overweighted, hypertensive, and aged ) 34 years. 168 women were distributed by random list into four groups; A) fitted with inert IUD (control); B) with Nova T (Cu-IUD);Cl received triphasic gestodene (OC);and D) monthly norethisterone enanthate 50 mg + estradiol valerate 5 mg (injectable). Serum lipid patterns were assessed at the beginning of contraceptive use and at 1, 6 and 12 months. The results of the Cu-IUDgroup did not show any difference with the control group. The OCgroup showed an increase of HDLand a significant lowering of LDLcholesterol at 6 months but this reverted by 12 months. On the other hand, in the injectable group, a sharp decrease of total lipids, total cholesterol and LDLwas observed from the first month, lasting throughout 12 months, yielding significant difference with controls and also with pfe-treatment values. In sum, the monthly injectable showed quite favorable patterns that would deserve further study, especially in view of the scarcity of hitherto available information concerning this formulation.
Address for Correspondence: la. C&edra de Ginecologia,Hospital de Clinicas Cordoba 235 1, ( 11201 Buenos Aires, Argentina Submitted for publication September 17, 1990 Accepted for publication July 10, 1991
SEPTEMBER 1991 VOL. 44 NO. 3
235
CONTRACEPTION
INTRODUCTION The relationship between lipid metabolism disturbances and the increase of the risk of arterial thromboembolism is nowadays almost universally accepted (1,2,3). Likewise, it is generally accepted that contraceptive steroids may affect, in several ways, the lipid metabolism and thus play a role in increasing thromboembolism risk I4,5,b). Nevertheless, this point is still under dispute in both physiopathological and epidemiological aspects (7,8X However, it became evident that synthetic progestogens may effect a lowering of HDLcholesterol, thus producing an imbalance of the HDL/LDL ratio (9,101, which has been directly related to subsequent atheroma and connected accident (3,4,10). Estrogens would exert an opposite action upon this mechanism ( 11). That gave momentum to the pill-lipid research with the result of a great many publications dealing with various aspects (formulations, doses, time of use, consumers’ status, etc.1 that may influence the effects of OCon lipids (3,12,13,14,15). Contrasting with the large amount of work done with oral contraceptives, there is a surprising scarcity of information using injectable preparations ( lb, 171,in spite of the fact that pharmacokinetics of oral and parenteral contraceptive steroids are quite different ItSI. Besides, although IUDshave been used in some studies as controls for OCs(5,19,20).there is no information so far whether copper-releasing IUDsmay exert any influence, e.g. whether there is any difference between copper IUDs and inert IUDs. In the present study, the behavior of serum lipid parameters was assessed in a prospective, randomized controlled design over one year’s duration. Inert Lippes IUDs were used as controls to compare the effects of copper IUDs,triphasic gestogen OCs, and monthly injectable progestogen plus estrogen contraceptives. MATERIAL 200 women from our family planning outpatient clinic were initially allotted by random list to one of the four contraceptive regimens. The criteria for the inclusion were: age between 20 and 34 years; proven fertility; no systemic contraceptives for the last b months; no smoking; no obesity &ss than +20X of ideal body weight); no hypertension, and with normal liver function as well as glucose tolerance tests. Only those patients who completed one year of contraception were used for the evaluation of results (n- 1661. The final integration and characteristics of the four groups are shown on Table I.
236
SEPTEMBER 1991 VOL. 44 NO. 3
CONTRACEPTION
TABLEI
STUDIED MATERIAL A
GROUPS
Contraceptive
, agent I ! I ncases age(years)x(S.D.) parityY(S.D.1 abortionsY?u(S.DJ
1 av.sexual frequency/ , week 51 (SD.)
B CU-IUD
Controls
Nova-T
LippesC
47
’
48
c TriphasicOC Gestodenet Ethinylestradiol (SH415G)* 33
I
Monthly D injectable Morethisterone enanthate 5Dmg + Estradiol valet-ate Smg (SH329AB) l 40
27.4 (4.8)
26.8 (5.6) 1
26.1 (5.4)
26.4t5.5)
1.5(I.21
1.7(1.4)
1.5(1.4)
1.6ll.3)
0.9 (1.1)
0.7(1.0)
0.4(0.6)
0.8 (1.1)
2.9(1.1)
2.8(1.3)
2.5(1.4)
2.8(1.2)
METHODS In each patient, four blood samples were taken, i.e. at the beginning of contraception, at one month on contraception, at six months on contraception, and at twelve months on contraception. The blood samples were taken between the 1st and 7th days of the cycle. All subjects were instructed to fast overnight and refrain from alcohol for three days pnor to giving the blood samples. In each blood sample the following parameters were assessed: total serum lipids (2 11, total cholesterol (221, HDL cholesterol, LDL cholesterol G$, and triglycerides (22). Multiple-comparison statistical analyses adjusted for the number of comparisons were performed by using Bonferroni’s procedure (241, with a level of significance of 0.05. Each group was compared with the control group at 1, 6 and 12 months after starting treatments. Also, within-group comparisons were performed against the own pf e-treatment values. RESULTS Fig. 1 shows the evolution of total sefum lipid values in the four studied groups. There were no differences between the values of the control group (A), the copper IUD group (B), and the OC group 0. On the other hand, the injectable group (D) showed a sharp decrease with high statistical significance when compared to the control group, and also with statistical significance against their own pfe-treatment values. This statistically significant lowering continued at six months of observation but lost statistical significance at twelve months, thus narrowing to the control group values. On Fig. 2 the evolution of total cholesterol values are shown. Here, again, the evolution of the copper IUDgroup and the OCgroup was similar to the controls with mean values and 95%confidence limit values almost totally overlapping. However, the
SEPTEMBER 1991 VOL. 44 NO. 3
237
CONTRACEPTION
mg/dl
CONT.
1 MO
6 MO
ugure 1. RESULTS OF TOTAL LIPIDS for tbe nunbor of comparisons).
12Mo
(Confidoaccintervals 95!5, adjusted
Legends A -
Control (inert IUD)group = Copper IUDgroup = Oral triphaslc group = MOnthly injectable group = p ,. 0.05 vs. control group ** = p s 0.01 vs. control group 0 = p
