- Email: [email protected]
Changes in coagulation and anticoagulation in women taking low-dose triphasic oral contraceptives: A controlled comparative 12-month clinical trial
Changes in coagulation and anticoagulation in women taking low-dose triphasic oral contraceptives: A controlled comparative 12-month clinical trial Morris Notelovitz, MD, PhD," Craig S. Kitchens, MD: and Farouk Y. Khan, MD"
c
Gainesville, Florida OBJECTIVE: The effects of two triphasic oral contraceptives on coagulation and anticoagulation factors were compared in a 12-month open-label study. STUDY DESIGN: Fifty-two women (mean age 26 years) were enrolled in and completed the study; 20 had been randomly assigned to receive levonorgestrel plus ethinyl estradiol, 24 had been randomly assigned to receive norethindrone plus ethinyl estradiol, and eight surgically sterile women acted as untreated controls. Coagulation and anticoagulation factors were measured at baseline and during the sixth and twelfth months. RESULTS: Both oral contraceptives produced significant decreases from baseline in prothrombin time and partial thromboplastin time; there were also significant changes in laboratory control times. Factor XII was significantly increased in both oral contraceptive groups after 6 and 12 months. Fibrinogen antigen was significantly increased for norethindrone plus ethinyl estradiol after 6 and 12 months and for levonorgestrel plus ethinyl estradiol after 12 months. Platelet counts were unchanged. There was a significant increase in antithrombin III activity with norethindrone plus ethinyl estradiol at 12 months. Antithrombin III antigen was unchanged with the oral contraceptives; however, significant increases existed for !X,-antitrypsin antigen and plasminogen antigen and activity after 6 and 12 months and for !X2-macroglobulin antigen after 12 months for both oral contraceptives. !X2-Antiplasmin antigen was significantly increased for norethindrone plus ethinyl estradiol at the 12-month evaluation. There were no significant differences between the oral contraceptives for any coagulation or anticoagulation factor, and mean values generally remained within reference ranges. CONCLUSIONS: Levonorgestrel plus ethinyl estradiol and norethindrone plus ethinyl estradiol had equivalent, minimal effects on hemostasis, and changes in coagulation factors appeared to be balanced by changes in anticoagulation factors. (AM J OSSTET GVNECOL 1992;167:1255-61.)
Key words: Oral contraceptives, coagulation, hemostasis Minimization of risks for cardiovascular disease, including thromboembolic events, is a continuing concern in the use of oral contraceptives. Extensive clinical experience has shown that changes in the concentration of coagulation factors and the incidence of venous thromboembolism are primarily related to the estrogen component 1 and are dose related.'-' The low-dose oral contraceptives currently used carry less cardiovascular risk than previous higher-dose formulations." In fact, some studies have suggested a cardioprotective effect associated with oral contraceptive use,6.7 particularly in relation to atherosclerosis and myocardial infarction. Although there is no conclusive evidence that progestogens alone affect hemostatic parameters,3. R the
From the Center for Climacteric Studies, Inc.," and the Departments of Internal Medicine b and Pathology,' Universitv of Florida. Supported by a grant in aid from Wyeth-Ayen-t Laboratories. Received for publication December 6, 1991; revised April 10, 1992; accepted April 15, 1992. Reprint requests: Morris Notelovitz, MD. PhD, 222 S. W. 36th Terrace, Suite C, Gainesville. FL 32607. 611 /38742
progestogen component of oral contraceptives does have a dose-related effect on arterial vascular disorders.1. 2." There are also recent indications that the progestogen component can modify the estrogenic effects of an oral contraceptive on hemostatic parameters." Therefore comparisons of oral contraceptives have been undertaken to determine which progestogens are associated with the smallest changes in coagulation factors. Often these comparisons have evaluated low-dose monophasic oral contraceptives'- '0-12 or triphasic formulations compared with monophasic ones. 1.9. 13 A few studies '·- l7 have also compared triphasic combinations of ethinyl estradiol plus levonorgestrel or gestodene. In this study we evaluated the effects of two commonly used triphasic preparations, Triphasil (WyethAyerst Laboratories, Philadelphia; levonorgestrel plus ethinyl estradiol) and Ortho Novum 71717 (Ortho Pharmaceutical Corporation, Raritan, N.J.; norethindrone plus ethinyl estradiol). Sabra and Bonnar" also studied the effects of levonorgestrel and norethindrone in combination oral contraceptives and observed that
1256
Notelovitz, Kitchens, and Khan
November 1992 Am J Obstet Gynecol
Table I. Triphasic dose regimens Tablet strength (p..g)
Oral contraceptive
50 75 125 500 750 1000
Triphasil (levonorgestrel plus ethinyl estradiol) Ortho-Novum 7/7/7 (norethindrone plus ethinyl estradiol)
an oral contraceptive containing 30 I-Lg ethinyl estradiol plus levonorgestrel changed certain hemostatic factors less than did an oral contraceptive containing 30 I-Lg ethinyl estradiol plus norethindrone. They suggested that levonorgestrel may decrease the effects of estrogen on hemostasis. Our study has addressed the suggestion that levonorgestrel might modify estrogen's effects on hemostasis. Because the two regimens that we studied contain comparable amounts of the same estrogen, it was assumed that any differences between the groups in coagulation factors could be attributed to the effects of the progestogens. It also has been suggested that changes in anticoagulation factors may work to balance any changes in coagulation factors. Therefore we have evaluated the effects of these two oral contraceptives on both coagulation and anticoagulation factors. Material and methods
Healthy women who were between 18 and 35 years old were enrolled in the study. They had to weigh within 20% of the average weight for their height and age and have regular menstrual cycles, and they could not have been pregnant within 90 days of the study start. The study received approval from the Center's institutional review board, and subjects gave written informed consent. The use of reproductive hormones was not permitted within 30 days of the start of the study. Use of anticoagulant drugs within 90 days of the study and aspirin within 10 days of baseline blood sampling also was not permitted. Participants agreed to maintain their dietary and exercise patterns during the study. They were evaluated for 12 cycles. A group of women with intact ovaries, who were surgically sterile, participated as the control group. The fertile women were randomly assigned to receive openlabel levonorgestrel plus ethinyl estradiol or norethindrone plus ethinyl estradiol. The dose regimens of the two oral contraceptives are given in Table I. Blood samples were collected after a 14-hour fast, between days 20 and 25 during the cycle preceding the study, and between days 15 and 21 during treatment cycles 6 and 12. The samples were collected in tubes
+ 30 + 40
+ 30 + 35 + 35
+ 35
Cycle days 1-6 7-11 12-21 1-7 8-16 17-21
containing 3.8% sodium citrate, and they were centrifuged immediately at 4500 rpm for 5 to 10 minutes. Plasma was separated and kept frozen at - 70° C until analysis. The coagulation factors measured were partial thromboplastin time, prothrombin time, fibrinogen antigen, platelet count, factor XllI, and fibrin degradation products. The anticoagulation factors measured were antithrombin III antigen and activity,
Fifty-two women between ~w and 33 years old completed this study; 20 received levonorgestrel plus ethinyl estradiol and 24 received norethindrone plus ethinyl estradiol. The remaining' eight women were surgically sterile and acted as the control group. There were no statistically significant differences in baseline demographic data among the three groups (Table II).
Coagulation and triphasic oral contraceptives
Volume 167 Number 5
1257
Table II. Baseline demographic data Control
Attribute
(n = 8)
Levonorgestrel plus ethinyl estradiol (n = 20)
Norethindrone plus ethinyl estradiol (n = 24)
Age (yr, mean ± SD) Parity (mean ± SEM) Race (No. and %) White Black Other Height, (in, mean ± SD) Weight (lb, mean ± SD) Cigarette use (no. and %) Yes No Cigarettes/day (No., mean ± SD) Alcohol use (No. of drinks/wk, mean ± SEM)
27.1 ± 3.0 1.1 ± 0.4
26.0 ± 3.2 0.8 ± 0.2
25.9 ± 3.9 0.7 ± 0.2
7 (88%) 1 (13%) 0(0%) 65.0 ± 1.0 133.1 ± 13.5
20 (100%) 0(0%) 0(0%) 65.0 ± 0.8 135.1 ± 16.7
20 (83%) 2 (8%) 2 (8%) 66.1 ± 0.6 133.7 ± 18.1
0(0%) 8 (100%) 0.0 ± 0.0 0.9 ± 0.4
2 (10%) 18 (90%) 0.6 ± 2.3 1.7 ± 0.4
1 (4%) 23 (96%) 0.3 ± 1.4 1.8 ± 0.4
Table III. Summary of coagulation factors in women taking low-dose triphasic oral contraceptives and in untreated control subjects Treatment group (sec, mean ± SEM) Coagulation factor with reference range
Control (n = 8)
Prothrombin time, subjects (9.5-12.0 sec)
Levonorgestrel plus ethinyl estradiol (n = 20)
Norethindrone plus ethinyl estradiol (n = 24)
Baseline 6 mo 12 mo
12.6 ± 0.1 12.5 ± 0.1 11.5 ± 0.2t
12.7 ± 0.1 12.4 ± 0.1* 11.2 ± 0.1 *
12.6 ± 0.1 12.2 ± O.lt 11.1 ± 0.1 *
Baseline 6 mo 12 mo
27.4 ± 0.8 27.0 ± 0.8 27.6 ± 2.0
28.5 ± 0.6 24.7 ± 0.6+ 25.2 ± LOt
26.5 ± 0.7 24.0 ± 0.5t, + 25.1 ± 1.0
Baseline 6 mo 12 mo
106.4 ± 5.9 103.6 ± 6.5 85.5 ± 3.2t
89.6 ± 5.6 127.4 ± 7.3§ 105.2 ± 7.6+, II
97.9 ± 4.8 131.4 ± 5.2t 110.4 ± 4.7t, +
Baseline 6 mo 12 mo
328.0 ± 48.6 266.6 ± 44.3 328.8 ± 20.4
305.3 ± 20.6 319.8 ± 19.7 350.5 ± 10.611
309.1 ± 17.3 355.6 ± 26.411 375.4 ± 13.3*, ~
Baseline 6 mo 12 mo
192.0 ± 23.7 202.8 ± 16.0 201.4 ± 11.4
282.6 ± 15.0 270.5 ± 19.8 292.3 ± 16.6
293.0 ± 17.4 297.8 ± 17.5 271.9 ± 19.2
Partial thromboplastin time, subjects «35 sec)
Factor XlI (60%-200% of normal)
Fibrinogen antigen (200-450 mg/dl)
Platelets (150-450 thousands/mm')#
*Significant difference from baseline, p :s 0.001. tSignificant difference from baseline, p :s 0.01. +Significant difference from control group, p:s 0.01. §Significant difference from control group, p :s 0.001. IISignificant difference from baseline, p :s 0.05. ~Significant difference from control, p :s 0.05. #Complete platelet data were available for only 5, 19, and 21 patients in the control, levonorgesterel plus ethinyl estradiol, and norethindrone plus ethinyl estradiol groups, respectively.
Results of prothrombin time and partial thromboplastin time testing are summarized in Table III. There were significant decreases from baseline in prothrombin time after 6 and 12 months for levonorgestrel plus ethinyl estradiol and norethindrone plus ethinyl estradiol and after 12 months for the control group. There was, however, no difference at any of the sampling intervals between the treatment groups and the un-
treated controls. Further, the values were well within the reference range of this laboratory. Partial thromboplastin time values were significantly shorter for both oral contraceptives compared with baseline and the control group at the 6-month evaluation and for levonorgestrel plus ethinyl estradiol compared with baseline after 12 months. Mean values for partial thromboplastin time remained within normal
1258
Notelovitz, Kitchens, and Khan
November 1992 Am J Obstet Gynecol
Table IV. Summary of anticoagulation parameters Control Anticoagulation factor and reference range
(n = 8) (mean ± SEM)
Antithrombin III antigen (22-39 mgldl)
Baseline 24.3 ± 0.8 6 mo 27.6 ± 1.3 12 mo 29.1 ± 1.2t Antithrombin III activity (80%-200% of normal) Baseline 9S.1 ± 3.9 6 mo 91.9 ± 4.6 12 mo 92.1 ± S.4 a.-Macroglobulin antigen (175-450 mgldl) Baseline 208.8 ± IS.O 6 mo 222.8 ± 9.3 12 mo 262.S ± 17.St ai-Antitrypsin antigen (200-400 mgldl) Baseline 223.1 ± IS.3 246.9 ± 22.2 6 mo 12 mo 23S.0 ± 13.4 Plasminogen activity (2.4 to 3.8 coherent time average unitslml) Baseline 2.9 ± 0.1 6 mo 3.3 ± 0.2 12 mo 3.S ± 0.2 a2-Antiplasmin antigen (80%-200% of normal) Baseline 83.4 ± 6.2 6 mo 77.0 ± 4.6 12 mo 77.8 ± 3.3
Levonorgestrel plus ethinyl estradiol (n = 20) (mean ± SEM)
2S.0 ± O.S 24.8 ± 0.6* 2S.1 ± 90.7*
Norethindrone plus ethinyl estradiol (n = 24) (mean ± SEM)
26.8 ± 0.6 2S.6 ± 0.6* 26.3 ± 0.6*
97.4 ± 3.7 90.7 ± 4.2 96.4 ± 2.9
91.6 ± 2.S 102.9 ± 4.2t 94.8 ± 3.4
21S.2 ± 10.8 220.9 ± 10.S 2S3.3 ± IS.O§
226.2 ± 9.6 236.8 ± 11.7 272.9 ± IS. 111
249.0 ± 10.4 333.8 ± 13.6*11 312.0 ± 14.0*11
260.8 ± 7.3 346.S ± 11.0*11 341.9 ± 11.911,~
3.1 ± 0.1 4.7 ± 0.211~
3.4 ± 0.1 S.O ± 0.111~ S.O ± 0.211~
s.o
± 0.211~
78.3 ± 3.2 76.3 ± 3.4 84.1 ± 3.7
77.6 ± 2.7 76.7 ± 3.S 86.0 ± 3.4t
*Significant difference from control, p :::; O.OS. tSignificant difference from baseline, p :::; O.OS. *Significant difference from control, p :::; 0.01. §Significant difference from baseline, p:::; 0.01. IISignificant difference from baseline, p:::; 0.001. ~Significant difference from control, p :::; 0.001.
ranges for the oral contraceptive-treated women, with a trend to reversal of the time at 12 months. The mean values for the other tested coagulation factors did not differ significantly between the two oral contraceptive groups, and none of the mean values were outside the reference ranges (Table III). The individual values also generally remained within the reference ranges. There were significant increases from baseline in factor XII values with both oral contraceptives at both ontherapy evaluations. Increases were greatest at 6 months and returned toward baseline values after 12 months. This trend mirrors the partial thromboplastin time results. The factor XII values in both treated groups at 6 and 12 months were significantly greater than those of the control group; the control group showed a significant decrease from baseline factor XII values after 12 months. The baseline values differed among the three groups, but the differences were not statistically significant. Levels of fibrinogen antigen increased significantly from baseline after 12 months for levonorgestrel plus ethinyl estradiol and after 6 and 12 months for norethindrone plus ethinyl estradiol. There were no significant changes in mean platelet counts for any group.
The baseline platelet value in the control groups was statistically less than that of the treatment groups (P < 0.01) but was still within the normal biologic range. Only one patient, who received norethindrone plus ethinyl estradiol, had a positive titer for fibrin degradation products. She had a titer of 1 : 40 at the baseline and 6-month evaluations, but she had a negative titer at the 12-month evaluation. It was concluded that, although there were some indicators of procoagulant change in this subject, there was no laboratory evidence of intravascular coagulation. Results of tests for anticoagul.ation factors are summarized in Table IV. There were no significant differences between oral contraceptive treatment groups for any of the anticoagulation factors tested. Mean values for the anticoagulation factors, except for plasminogen activity, remained within normal ranges. Individual values also generally remained within normal ranges except for plasminogen activity; for plasminogen activity individual values were often higher than the reference range (Fig. 1). Antithrombin III activity and antithrombin III antigen generally showed no significant changes from baseline in the treated women, although antithrombin III activity was significantly increased after 6 months
Coagulation and triphasic oral contraceptives
Volume 167 \':umber 5
CONTROL (n=8)
NET/EE (n=24)
LNG/EE (n=20)
7
6 6
...J
--
E en
6 t--
;:)
5
IZ
,
6 6 6 6(3) 6 6(2) 6 6(5)
-
IU
>I> IU
0
"z 0
8(2)
4 r-------------------------------------------0
3 r-
z w
0
g!~!
0(2) 0
0 0 0(2)
0 0(2) 0(2) 0
6
6(2)
t(2) t(2) 6(2)
6(2)
6(2)
6
6
6(2)
6
:(3)
6(3)
6
~
6 6(3)
-------4------------4----------... ---------
6(4) 6(4)
6 6 6 6(2) 6 6 6(2) 6(4) 6(2)
6
6
~(5) 6
~(4)
6
6(8) 6 6(4) 6
1(4) 6(3) 6(2)
6 6
---------------------.--------------------- -------~-(~)--------------------------------- -------------------------------------------_ .. -
2 -
0
6
6
6
:2:
....J
6
----------6----------------------£:::-----------
en
a..
1259
0
1 '-
0
6 6 I
o
I
I
6
12
I
o
I
I
6
12
I
o
CONTROL LNG/EE NET/EE
I
I
6
12
MONTHS Fig. 1. Alteration in plasminogen activity in women taking low-dose triphasic oral contraceptives and in untreated control women relative to reference range (2.4 to 3.8 coherent time average units per milliliter). 0, Control subjects; A, subjects taking levonorgestrel (LNG) plus ethinyl estradiol (EE); /'" subjects taking norethindrone (NET) plus ethinyl estradiol.
(but not later) for norethindrone plus ethinyl estradiol. Both treatment groups had significantly lower antithrombin II antigen levels at 6 and 12 months when compared with those of the control group. Both treatment groups showed statistically significant increases from baseline values in the less potent plasmin inhibitors, a"-macroglobulin antigen (12 months only) and ai-antitrypsin antigen (6 and 12 months). In addition, both oral contraceptive-treated groups had statistically significant higher ai-antitrypsin antigen values than did the control group at 6 and 12 mOllths. Plasminogen activity was significantly increased. a"-Antiplasmin antigen was unchanged at 6 months for both oral contraceptives, but it was significantly increased from baseline for norethindrone plus ethinyl estradiol at 12 months. Comment
Studies evaluating the effect of sex steroids on hemostasis aU have the same deficiencies: The factors measured are remote from the site of thrombosis; there are no known tests that reliably predict coagulation"l; some of the factors (e.g., factor XII) may initiate both coagulation and fibrinolysis; and the changes that occur
may be statistically significant but, with few exceptions, the mean values do not exceed the reference laboratories' normal ranges. Any hemostatic effect by these agents is therefore theoreticallv doubtful. Given these limitations, the results of tests for coagulation and anticogulation factors in the current study indicated no significant difference between the levonorgestrel plus ethinyl estradiol and norethindrone plus ethinyl estradiol combination triphasic oral contraceptives. It appears that the progestogens in these oral contraceptives do not differ significantly in their effect on these factors, either by direct impact on hemostasis or by influencing estrogen's effects on hemostasis. The results of this study do not support the suggestion that the progestogen modifies the effect of estrogen on hemostasis." Further, although both oral contraceptives induced changes in coagulation and anticoagulation factors (that were often statistically significant), mean values usually remained within normal reference ranges, with the exception of their effect on plasminogen activity. The prothrombin time is a clinical marker of activation of the extrinsic compartment of the coagulation cascade, because it reflects factor VII activity. According to some researchers/"O higher levels of factor VII
1260 Notelovitz, Kitchens, and Khan
and fibrinogen are associated with an increased tendency to arterial disease and thrombosis. Mean prothrombin time levels decreased significantly for both oral contraceptive groups after 6 and 12 months. However, the clinical significance of changes in prothrombin time is questionable because the mean prothrombin time for the control group also was significantly decreased after 12 months and there were no statistically significant differences between the control and treatment groups. Furthermore, all values were within the laboratory range. The partial thromboplastin time is reflective of a more complex interaction of segments of the intrinsic system leading eventually to the activation of factor X. The partial thromboplastin time values for the control group remained constant, which suggests the validity of oral contraceptive-induced shortening of the partial thromboplastin time. These decreases also are consistent with the results of an earlier evaluation of contraceptives containing norethindrone l2 ; these findings need to be interpreted within the context of the wide biologic range of partial thromboplastin time function and the simultaneous changes in anticoagulation and fibrinolysis. For example, factor XII (which stimulates the intrinsic system) and fibrinogen antigen values increased significantly with both oral contraceptives; in contrast, factor XII activity decreased significantly in the control group. Gevers Leuven et al. II also noted significant increases in factor XII in a study comparing monophasic levonorgestrel plus ethinyl estradiol and Iynestrenol plus ethinyl estradiol. However, the relevance of this with regard to procoagulation is questionable, because factor XII stimulates fibrinolysis as well.· Fibrinogen is the substrate that is converted by thrombin to fibrin and has been found to be increased in most (if not all) studies evaluating exogenous estrogen. Although an increased fibrinogen level is an epidemiologic risk factor for arterial disease and thrombosis;' 20 it is important to differentiate endogenous evaluations in untreated subjects with exogenously stimulated values in response to oral contraceptives. The latter response is pharmacologically balanced by enhanced fibrinolysis. Decreases in antithrombin III previously have been associated with estrogen use 9 ; however, the values for antithrombin III antigen and activity in this study generally remained unchanged. These results are consistent with other studies of levonorgestrel- and norethindrone-containing oral contraceptives,12. 14-18 although we did see decreases in antithrombin III antigen with norethindrone plus ethinyl estradiol in an earlier study.12 The unchanged antithrombin III activity levels in both oral contraceptive-treated groups are (theoretically at least) capable of inhibiting any increase in coagulability, as suggested by the partial thromboplas-
November 1992 Am J Obstet Gynecol
tin time change. Prime actions among the actions of antithrombin III is inhibition factor Xa activity. Increases in concentration of other anticoagulation factors may also compensate for the increases in coagulation factors; for example, values for plasminogen activity increased significantly with both oral contraceptives. The increase in plasminogen activity is especially encouraging. Although elevated levels of plasminogen activity do not necessarily imply increased fibrinolysis in vivo, there are two points that merit emphasis: the plasminogen activity levels were the only changes that exceeded the laboratory reference range, and this finding may reflect a clinically meaningful pharmacologic effect. This is made likely by extrapolation from the recent study of Jespersen et al. 4 They compared the fibrinolytic potential of two oral contraceptives containing third-generation progestogens (desogestrel and gestodene) combined with ethinyl estradiol and noted a highly significant increase in plasminogen antigen. This was accompanied by a simultaneous and significant decrease (p < 0.01) in histidine-rich glycoprotein (the natural inhibitor of plasminogen synthesis) and a trend toward decreases in plasminogen activity inhibitor type I, another highly potent inhibitor of the conversion of plasminogen to plasmin. These changes were accompanied by a significant (p < 0.05) increase in tissue plasminogen activator activity. Changes in fibrinolytic activity and anticoagulation factors consistent with the hypothesis that hemostatic balance is normally maintained with norethindroneand levonorgestrel-containing low-dose oral contraceptives were seen in other studies l2 . 17 and are supported by the conclusions of this study: Triphasic levonorgestrel plus ethinyl estradiol and norethindrone plus ethinyl estradiol produced equivalent changes in coagulation parameters, but these changes were offset by increases in anticoagulation activity. This is supported by the observation that only one oral contraceptivetreated patient had a positive titer for fibrin degradation products and at only one measurement point. There was no indication that the progestogen modified the effect of estrogen on hemostasis. We therefore conclude that both tested oral contraceptives can be expected to have a minimal net effect on the laboratory evaluation of these hemostatic factors in healthy women. These findings are consistent with the observed clinical safety of these compounds.
REFERENCES 1. Bonnar J. Coagulation effects of oral contraception. AM ] OSSTET GYNECOL 1987;157:1042-8. 2. Kelleher CC. Clinical aspects of the relationship between oral contraceptives and abnormalities of the hemostatic system: relation to the development of cardiovascular disease. AM] OSSTET GYNECOL 1990;163:392-5. 3. Beller FK, Ebert C. Effects of oral contraceptives on blood
Coagulation and triphasic oral contraceptives
Volume 167 Number 5
4.
5.
6. 7. 8.
9. 10.
II.
12. 13.
coagulation. A review. Obstet Gynecol Surv 1985;40;42536. Jespersen J. Petersen KR. Skouby SO. Effects of newer oral contraceptives on the inhibition of coagulation and fibrinolysis in relation to dosage and type of steroid. AM J OBSTET GYNECOL 1990;163:396-403. Vessey MP. The Jephcott lecture. 1989: an overview of the benefits and risks of combined oral contraceptives. Oral contraceptives and breast cancer. Proceedings of a symposium held at The Royal Society of Medicine. London. July 1989. London: Royal Society of Medicine. 1989:121-35. Adams MR. Clarkson TB. Koritnik DR. Nash HA. Contraceptive steroids and coronary artery atherosclerosis in cynomolgus macaques. Fertil Steril 1987;47: 1010-8. Croft p. Hannaford P. Risk factors for acute myocardial infarction in women. BMJ 1989;298:674. Conard J. Samana M. The effects of progestins on coagulation. In: Bardin CWo Milgrom E. Manvais-Jarvis p. eds. Progesterone and progestins. New York: Raven Press. 1983:447-52. Bonnar J. Sabra AM. Oral contraceptives and blood coagulation. J Reprod Med 1986;3 I (suppl):55 1-6. Daly L. Bonnar J. Comparative studies of 30 ILg ethinyl estradiol combined with gestodene and desogestrel on blood coagulation. fibrinolysis. and platelets. AM J OBSTET GYNECOL 1990;163:430-7. Gevers Leuven JA. Kluft C. Bertina RM. Hessel LW. Effects of two low-dose oral contraceptives on circulating components of the coagulation and fibrinolytic systems. J Lab Clin Med 1987;109:631-6. Notelovitz M. Kitchens CS. Coone L. McKenzie L. Carter R. Low-dose oral contraceptive usage and coagulation. AM J OBSTET GYNECOL 1981;141:71-5. DavidJL. Gaspard UJ. Gillian D. Raskinet R. Lepot MR.
14.
15.
16.
17. 18.
19.
20. 21.
1261
Hemostasis profile in women taking low-dose oral contraceptives. AM J OBSTET GYNECOL 1990; 163 :420-3. Refn H. Kjaer A. Lebech A-M. Borggaard B. Schierup L. Bremmelgaard A. Metabolic changes during treatment with two different progestogens. AM J OBSTET GYNECOL 1990;163:374-7. Omsjo IH. Oian p. Maltau JM. Osterud B. Effects of two triphasic oral contraceptives containing ethinylestradiol plus levonorgestrel or gestodene on blood coagulation and fibrinolysis. Acta Obstet Gynecol Scand 1989;68:2730. Kjaer A. Lebech A-M. Borggaard B. et al. Lipid metabolism and coagulation of two contraceptives: correlation to serum concentrations of levonorgestrel and gestodene. Contraception 1989;40:665-73. Bonnar J. Daly L. Carroll E. Blood coagulation with a combination pill containing gestodene and ethinyl estradiol. Int J Ferti! 1987;32:21-8. Sabra A. Bonnar J. Hemostatic system changes induced by 50 ILg and 30 ILg estrogen/progestogen oral contraceptives. Modification of estrogen effects by levonorgestrel. J Reprod Med 1983;28:85-91. Smith LG. Kitchens CS. A comparison between two commercially available activators for determining the partial thromboplastic time. Arch Pathol Lab Med 1985;109; 243-6. Merskey C. Kleiner GJ. Johnson AJ. Quantitative estimation of split products of fibrinogen in human serum. Relation to diagnosis and treatment. Blood 1966;28: I. Kluft C. Disorders of the hemostasis system and the risk of the development of thrombotic and cardiovascular diseases: limitations of laboratory diagnosis. AM J OBSTET GYNECOL 1990;163:305-12.
c
Gainesville, Florida OBJECTIVE: The effects of two triphasic oral contraceptives on coagulation and anticoagulation factors were compared in a 12-month open-label study. STUDY DESIGN: Fifty-two women (mean age 26 years) were enrolled in and completed the study; 20 had been randomly assigned to receive levonorgestrel plus ethinyl estradiol, 24 had been randomly assigned to receive norethindrone plus ethinyl estradiol, and eight surgically sterile women acted as untreated controls. Coagulation and anticoagulation factors were measured at baseline and during the sixth and twelfth months. RESULTS: Both oral contraceptives produced significant decreases from baseline in prothrombin time and partial thromboplastin time; there were also significant changes in laboratory control times. Factor XII was significantly increased in both oral contraceptive groups after 6 and 12 months. Fibrinogen antigen was significantly increased for norethindrone plus ethinyl estradiol after 6 and 12 months and for levonorgestrel plus ethinyl estradiol after 12 months. Platelet counts were unchanged. There was a significant increase in antithrombin III activity with norethindrone plus ethinyl estradiol at 12 months. Antithrombin III antigen was unchanged with the oral contraceptives; however, significant increases existed for !X,-antitrypsin antigen and plasminogen antigen and activity after 6 and 12 months and for !X2-macroglobulin antigen after 12 months for both oral contraceptives. !X2-Antiplasmin antigen was significantly increased for norethindrone plus ethinyl estradiol at the 12-month evaluation. There were no significant differences between the oral contraceptives for any coagulation or anticoagulation factor, and mean values generally remained within reference ranges. CONCLUSIONS: Levonorgestrel plus ethinyl estradiol and norethindrone plus ethinyl estradiol had equivalent, minimal effects on hemostasis, and changes in coagulation factors appeared to be balanced by changes in anticoagulation factors. (AM J OSSTET GVNECOL 1992;167:1255-61.)
Key words: Oral contraceptives, coagulation, hemostasis Minimization of risks for cardiovascular disease, including thromboembolic events, is a continuing concern in the use of oral contraceptives. Extensive clinical experience has shown that changes in the concentration of coagulation factors and the incidence of venous thromboembolism are primarily related to the estrogen component 1 and are dose related.'-' The low-dose oral contraceptives currently used carry less cardiovascular risk than previous higher-dose formulations." In fact, some studies have suggested a cardioprotective effect associated with oral contraceptive use,6.7 particularly in relation to atherosclerosis and myocardial infarction. Although there is no conclusive evidence that progestogens alone affect hemostatic parameters,3. R the
From the Center for Climacteric Studies, Inc.," and the Departments of Internal Medicine b and Pathology,' Universitv of Florida. Supported by a grant in aid from Wyeth-Ayen-t Laboratories. Received for publication December 6, 1991; revised April 10, 1992; accepted April 15, 1992. Reprint requests: Morris Notelovitz, MD. PhD, 222 S. W. 36th Terrace, Suite C, Gainesville. FL 32607. 611 /38742
progestogen component of oral contraceptives does have a dose-related effect on arterial vascular disorders.1. 2." There are also recent indications that the progestogen component can modify the estrogenic effects of an oral contraceptive on hemostatic parameters." Therefore comparisons of oral contraceptives have been undertaken to determine which progestogens are associated with the smallest changes in coagulation factors. Often these comparisons have evaluated low-dose monophasic oral contraceptives'- '0-12 or triphasic formulations compared with monophasic ones. 1.9. 13 A few studies '·- l7 have also compared triphasic combinations of ethinyl estradiol plus levonorgestrel or gestodene. In this study we evaluated the effects of two commonly used triphasic preparations, Triphasil (WyethAyerst Laboratories, Philadelphia; levonorgestrel plus ethinyl estradiol) and Ortho Novum 71717 (Ortho Pharmaceutical Corporation, Raritan, N.J.; norethindrone plus ethinyl estradiol). Sabra and Bonnar" also studied the effects of levonorgestrel and norethindrone in combination oral contraceptives and observed that
1256
Notelovitz, Kitchens, and Khan
November 1992 Am J Obstet Gynecol
Table I. Triphasic dose regimens Tablet strength (p..g)
Oral contraceptive
50 75 125 500 750 1000
Triphasil (levonorgestrel plus ethinyl estradiol) Ortho-Novum 7/7/7 (norethindrone plus ethinyl estradiol)
an oral contraceptive containing 30 I-Lg ethinyl estradiol plus levonorgestrel changed certain hemostatic factors less than did an oral contraceptive containing 30 I-Lg ethinyl estradiol plus norethindrone. They suggested that levonorgestrel may decrease the effects of estrogen on hemostasis. Our study has addressed the suggestion that levonorgestrel might modify estrogen's effects on hemostasis. Because the two regimens that we studied contain comparable amounts of the same estrogen, it was assumed that any differences between the groups in coagulation factors could be attributed to the effects of the progestogens. It also has been suggested that changes in anticoagulation factors may work to balance any changes in coagulation factors. Therefore we have evaluated the effects of these two oral contraceptives on both coagulation and anticoagulation factors. Material and methods
Healthy women who were between 18 and 35 years old were enrolled in the study. They had to weigh within 20% of the average weight for their height and age and have regular menstrual cycles, and they could not have been pregnant within 90 days of the study start. The study received approval from the Center's institutional review board, and subjects gave written informed consent. The use of reproductive hormones was not permitted within 30 days of the start of the study. Use of anticoagulant drugs within 90 days of the study and aspirin within 10 days of baseline blood sampling also was not permitted. Participants agreed to maintain their dietary and exercise patterns during the study. They were evaluated for 12 cycles. A group of women with intact ovaries, who were surgically sterile, participated as the control group. The fertile women were randomly assigned to receive openlabel levonorgestrel plus ethinyl estradiol or norethindrone plus ethinyl estradiol. The dose regimens of the two oral contraceptives are given in Table I. Blood samples were collected after a 14-hour fast, between days 20 and 25 during the cycle preceding the study, and between days 15 and 21 during treatment cycles 6 and 12. The samples were collected in tubes
+ 30 + 40
+ 30 + 35 + 35
+ 35
Cycle days 1-6 7-11 12-21 1-7 8-16 17-21
containing 3.8% sodium citrate, and they were centrifuged immediately at 4500 rpm for 5 to 10 minutes. Plasma was separated and kept frozen at - 70° C until analysis. The coagulation factors measured were partial thromboplastin time, prothrombin time, fibrinogen antigen, platelet count, factor XllI, and fibrin degradation products. The anticoagulation factors measured were antithrombin III antigen and activity,
Fifty-two women between ~w and 33 years old completed this study; 20 received levonorgestrel plus ethinyl estradiol and 24 received norethindrone plus ethinyl estradiol. The remaining' eight women were surgically sterile and acted as the control group. There were no statistically significant differences in baseline demographic data among the three groups (Table II).
Coagulation and triphasic oral contraceptives
Volume 167 Number 5
1257
Table II. Baseline demographic data Control
Attribute
(n = 8)
Levonorgestrel plus ethinyl estradiol (n = 20)
Norethindrone plus ethinyl estradiol (n = 24)
Age (yr, mean ± SD) Parity (mean ± SEM) Race (No. and %) White Black Other Height, (in, mean ± SD) Weight (lb, mean ± SD) Cigarette use (no. and %) Yes No Cigarettes/day (No., mean ± SD) Alcohol use (No. of drinks/wk, mean ± SEM)
27.1 ± 3.0 1.1 ± 0.4
26.0 ± 3.2 0.8 ± 0.2
25.9 ± 3.9 0.7 ± 0.2
7 (88%) 1 (13%) 0(0%) 65.0 ± 1.0 133.1 ± 13.5
20 (100%) 0(0%) 0(0%) 65.0 ± 0.8 135.1 ± 16.7
20 (83%) 2 (8%) 2 (8%) 66.1 ± 0.6 133.7 ± 18.1
0(0%) 8 (100%) 0.0 ± 0.0 0.9 ± 0.4
2 (10%) 18 (90%) 0.6 ± 2.3 1.7 ± 0.4
1 (4%) 23 (96%) 0.3 ± 1.4 1.8 ± 0.4
Table III. Summary of coagulation factors in women taking low-dose triphasic oral contraceptives and in untreated control subjects Treatment group (sec, mean ± SEM) Coagulation factor with reference range
Control (n = 8)
Prothrombin time, subjects (9.5-12.0 sec)
Levonorgestrel plus ethinyl estradiol (n = 20)
Norethindrone plus ethinyl estradiol (n = 24)
Baseline 6 mo 12 mo
12.6 ± 0.1 12.5 ± 0.1 11.5 ± 0.2t
12.7 ± 0.1 12.4 ± 0.1* 11.2 ± 0.1 *
12.6 ± 0.1 12.2 ± O.lt 11.1 ± 0.1 *
Baseline 6 mo 12 mo
27.4 ± 0.8 27.0 ± 0.8 27.6 ± 2.0
28.5 ± 0.6 24.7 ± 0.6+ 25.2 ± LOt
26.5 ± 0.7 24.0 ± 0.5t, + 25.1 ± 1.0
Baseline 6 mo 12 mo
106.4 ± 5.9 103.6 ± 6.5 85.5 ± 3.2t
89.6 ± 5.6 127.4 ± 7.3§ 105.2 ± 7.6+, II
97.9 ± 4.8 131.4 ± 5.2t 110.4 ± 4.7t, +
Baseline 6 mo 12 mo
328.0 ± 48.6 266.6 ± 44.3 328.8 ± 20.4
305.3 ± 20.6 319.8 ± 19.7 350.5 ± 10.611
309.1 ± 17.3 355.6 ± 26.411 375.4 ± 13.3*, ~
Baseline 6 mo 12 mo
192.0 ± 23.7 202.8 ± 16.0 201.4 ± 11.4
282.6 ± 15.0 270.5 ± 19.8 292.3 ± 16.6
293.0 ± 17.4 297.8 ± 17.5 271.9 ± 19.2
Partial thromboplastin time, subjects «35 sec)
Factor XlI (60%-200% of normal)
Fibrinogen antigen (200-450 mg/dl)
Platelets (150-450 thousands/mm')#
*Significant difference from baseline, p :s 0.001. tSignificant difference from baseline, p :s 0.01. +Significant difference from control group, p:s 0.01. §Significant difference from control group, p :s 0.001. IISignificant difference from baseline, p :s 0.05. ~Significant difference from control, p :s 0.05. #Complete platelet data were available for only 5, 19, and 21 patients in the control, levonorgesterel plus ethinyl estradiol, and norethindrone plus ethinyl estradiol groups, respectively.
Results of prothrombin time and partial thromboplastin time testing are summarized in Table III. There were significant decreases from baseline in prothrombin time after 6 and 12 months for levonorgestrel plus ethinyl estradiol and norethindrone plus ethinyl estradiol and after 12 months for the control group. There was, however, no difference at any of the sampling intervals between the treatment groups and the un-
treated controls. Further, the values were well within the reference range of this laboratory. Partial thromboplastin time values were significantly shorter for both oral contraceptives compared with baseline and the control group at the 6-month evaluation and for levonorgestrel plus ethinyl estradiol compared with baseline after 12 months. Mean values for partial thromboplastin time remained within normal
1258
Notelovitz, Kitchens, and Khan
November 1992 Am J Obstet Gynecol
Table IV. Summary of anticoagulation parameters Control Anticoagulation factor and reference range
(n = 8) (mean ± SEM)
Antithrombin III antigen (22-39 mgldl)
Baseline 24.3 ± 0.8 6 mo 27.6 ± 1.3 12 mo 29.1 ± 1.2t Antithrombin III activity (80%-200% of normal) Baseline 9S.1 ± 3.9 6 mo 91.9 ± 4.6 12 mo 92.1 ± S.4 a.-Macroglobulin antigen (175-450 mgldl) Baseline 208.8 ± IS.O 6 mo 222.8 ± 9.3 12 mo 262.S ± 17.St ai-Antitrypsin antigen (200-400 mgldl) Baseline 223.1 ± IS.3 246.9 ± 22.2 6 mo 12 mo 23S.0 ± 13.4 Plasminogen activity (2.4 to 3.8 coherent time average unitslml) Baseline 2.9 ± 0.1 6 mo 3.3 ± 0.2 12 mo 3.S ± 0.2 a2-Antiplasmin antigen (80%-200% of normal) Baseline 83.4 ± 6.2 6 mo 77.0 ± 4.6 12 mo 77.8 ± 3.3
Levonorgestrel plus ethinyl estradiol (n = 20) (mean ± SEM)
2S.0 ± O.S 24.8 ± 0.6* 2S.1 ± 90.7*
Norethindrone plus ethinyl estradiol (n = 24) (mean ± SEM)
26.8 ± 0.6 2S.6 ± 0.6* 26.3 ± 0.6*
97.4 ± 3.7 90.7 ± 4.2 96.4 ± 2.9
91.6 ± 2.S 102.9 ± 4.2t 94.8 ± 3.4
21S.2 ± 10.8 220.9 ± 10.S 2S3.3 ± IS.O§
226.2 ± 9.6 236.8 ± 11.7 272.9 ± IS. 111
249.0 ± 10.4 333.8 ± 13.6*11 312.0 ± 14.0*11
260.8 ± 7.3 346.S ± 11.0*11 341.9 ± 11.911,~
3.1 ± 0.1 4.7 ± 0.211~
3.4 ± 0.1 S.O ± 0.111~ S.O ± 0.211~
s.o
± 0.211~
78.3 ± 3.2 76.3 ± 3.4 84.1 ± 3.7
77.6 ± 2.7 76.7 ± 3.S 86.0 ± 3.4t
*Significant difference from control, p :::; O.OS. tSignificant difference from baseline, p :::; O.OS. *Significant difference from control, p :::; 0.01. §Significant difference from baseline, p:::; 0.01. IISignificant difference from baseline, p:::; 0.001. ~Significant difference from control, p :::; 0.001.
ranges for the oral contraceptive-treated women, with a trend to reversal of the time at 12 months. The mean values for the other tested coagulation factors did not differ significantly between the two oral contraceptive groups, and none of the mean values were outside the reference ranges (Table III). The individual values also generally remained within the reference ranges. There were significant increases from baseline in factor XII values with both oral contraceptives at both ontherapy evaluations. Increases were greatest at 6 months and returned toward baseline values after 12 months. This trend mirrors the partial thromboplastin time results. The factor XII values in both treated groups at 6 and 12 months were significantly greater than those of the control group; the control group showed a significant decrease from baseline factor XII values after 12 months. The baseline values differed among the three groups, but the differences were not statistically significant. Levels of fibrinogen antigen increased significantly from baseline after 12 months for levonorgestrel plus ethinyl estradiol and after 6 and 12 months for norethindrone plus ethinyl estradiol. There were no significant changes in mean platelet counts for any group.
The baseline platelet value in the control groups was statistically less than that of the treatment groups (P < 0.01) but was still within the normal biologic range. Only one patient, who received norethindrone plus ethinyl estradiol, had a positive titer for fibrin degradation products. She had a titer of 1 : 40 at the baseline and 6-month evaluations, but she had a negative titer at the 12-month evaluation. It was concluded that, although there were some indicators of procoagulant change in this subject, there was no laboratory evidence of intravascular coagulation. Results of tests for anticoagul.ation factors are summarized in Table IV. There were no significant differences between oral contraceptive treatment groups for any of the anticoagulation factors tested. Mean values for the anticoagulation factors, except for plasminogen activity, remained within normal ranges. Individual values also generally remained within normal ranges except for plasminogen activity; for plasminogen activity individual values were often higher than the reference range (Fig. 1). Antithrombin III activity and antithrombin III antigen generally showed no significant changes from baseline in the treated women, although antithrombin III activity was significantly increased after 6 months
Coagulation and triphasic oral contraceptives
Volume 167 \':umber 5
CONTROL (n=8)
NET/EE (n=24)
LNG/EE (n=20)
7
6 6
...J
--
E en
6 t--
;:)
5
IZ
,
6 6 6 6(3) 6 6(2) 6 6(5)
-
IU
>I> IU
0
"z 0
8(2)
4 r-------------------------------------------0
3 r-
z w
0
g!~!
0(2) 0
0 0 0(2)
0 0(2) 0(2) 0
6
6(2)
t(2) t(2) 6(2)
6(2)
6(2)
6
6
6(2)
6
:(3)
6(3)
6
~
6 6(3)
-------4------------4----------... ---------
6(4) 6(4)
6 6 6 6(2) 6 6 6(2) 6(4) 6(2)
6
6
~(5) 6
~(4)
6
6(8) 6 6(4) 6
1(4) 6(3) 6(2)
6 6
---------------------.--------------------- -------~-(~)--------------------------------- -------------------------------------------_ .. -
2 -
0
6
6
6
:2:
....J
6
----------6----------------------£:::-----------
en
a..
1259
0
1 '-
0
6 6 I
o
I
I
6
12
I
o
I
I
6
12
I
o
CONTROL LNG/EE NET/EE
I
I
6
12
MONTHS Fig. 1. Alteration in plasminogen activity in women taking low-dose triphasic oral contraceptives and in untreated control women relative to reference range (2.4 to 3.8 coherent time average units per milliliter). 0, Control subjects; A, subjects taking levonorgestrel (LNG) plus ethinyl estradiol (EE); /'" subjects taking norethindrone (NET) plus ethinyl estradiol.
(but not later) for norethindrone plus ethinyl estradiol. Both treatment groups had significantly lower antithrombin II antigen levels at 6 and 12 months when compared with those of the control group. Both treatment groups showed statistically significant increases from baseline values in the less potent plasmin inhibitors, a"-macroglobulin antigen (12 months only) and ai-antitrypsin antigen (6 and 12 months). In addition, both oral contraceptive-treated groups had statistically significant higher ai-antitrypsin antigen values than did the control group at 6 and 12 mOllths. Plasminogen activity was significantly increased. a"-Antiplasmin antigen was unchanged at 6 months for both oral contraceptives, but it was significantly increased from baseline for norethindrone plus ethinyl estradiol at 12 months. Comment
Studies evaluating the effect of sex steroids on hemostasis aU have the same deficiencies: The factors measured are remote from the site of thrombosis; there are no known tests that reliably predict coagulation"l; some of the factors (e.g., factor XII) may initiate both coagulation and fibrinolysis; and the changes that occur
may be statistically significant but, with few exceptions, the mean values do not exceed the reference laboratories' normal ranges. Any hemostatic effect by these agents is therefore theoreticallv doubtful. Given these limitations, the results of tests for coagulation and anticogulation factors in the current study indicated no significant difference between the levonorgestrel plus ethinyl estradiol and norethindrone plus ethinyl estradiol combination triphasic oral contraceptives. It appears that the progestogens in these oral contraceptives do not differ significantly in their effect on these factors, either by direct impact on hemostasis or by influencing estrogen's effects on hemostasis. The results of this study do not support the suggestion that the progestogen modifies the effect of estrogen on hemostasis." Further, although both oral contraceptives induced changes in coagulation and anticoagulation factors (that were often statistically significant), mean values usually remained within normal reference ranges, with the exception of their effect on plasminogen activity. The prothrombin time is a clinical marker of activation of the extrinsic compartment of the coagulation cascade, because it reflects factor VII activity. According to some researchers/"O higher levels of factor VII
1260 Notelovitz, Kitchens, and Khan
and fibrinogen are associated with an increased tendency to arterial disease and thrombosis. Mean prothrombin time levels decreased significantly for both oral contraceptive groups after 6 and 12 months. However, the clinical significance of changes in prothrombin time is questionable because the mean prothrombin time for the control group also was significantly decreased after 12 months and there were no statistically significant differences between the control and treatment groups. Furthermore, all values were within the laboratory range. The partial thromboplastin time is reflective of a more complex interaction of segments of the intrinsic system leading eventually to the activation of factor X. The partial thromboplastin time values for the control group remained constant, which suggests the validity of oral contraceptive-induced shortening of the partial thromboplastin time. These decreases also are consistent with the results of an earlier evaluation of contraceptives containing norethindrone l2 ; these findings need to be interpreted within the context of the wide biologic range of partial thromboplastin time function and the simultaneous changes in anticoagulation and fibrinolysis. For example, factor XII (which stimulates the intrinsic system) and fibrinogen antigen values increased significantly with both oral contraceptives; in contrast, factor XII activity decreased significantly in the control group. Gevers Leuven et al. II also noted significant increases in factor XII in a study comparing monophasic levonorgestrel plus ethinyl estradiol and Iynestrenol plus ethinyl estradiol. However, the relevance of this with regard to procoagulation is questionable, because factor XII stimulates fibrinolysis as well.· Fibrinogen is the substrate that is converted by thrombin to fibrin and has been found to be increased in most (if not all) studies evaluating exogenous estrogen. Although an increased fibrinogen level is an epidemiologic risk factor for arterial disease and thrombosis;' 20 it is important to differentiate endogenous evaluations in untreated subjects with exogenously stimulated values in response to oral contraceptives. The latter response is pharmacologically balanced by enhanced fibrinolysis. Decreases in antithrombin III previously have been associated with estrogen use 9 ; however, the values for antithrombin III antigen and activity in this study generally remained unchanged. These results are consistent with other studies of levonorgestrel- and norethindrone-containing oral contraceptives,12. 14-18 although we did see decreases in antithrombin III antigen with norethindrone plus ethinyl estradiol in an earlier study.12 The unchanged antithrombin III activity levels in both oral contraceptive-treated groups are (theoretically at least) capable of inhibiting any increase in coagulability, as suggested by the partial thromboplas-
November 1992 Am J Obstet Gynecol
tin time change. Prime actions among the actions of antithrombin III is inhibition factor Xa activity. Increases in concentration of other anticoagulation factors may also compensate for the increases in coagulation factors; for example, values for plasminogen activity increased significantly with both oral contraceptives. The increase in plasminogen activity is especially encouraging. Although elevated levels of plasminogen activity do not necessarily imply increased fibrinolysis in vivo, there are two points that merit emphasis: the plasminogen activity levels were the only changes that exceeded the laboratory reference range, and this finding may reflect a clinically meaningful pharmacologic effect. This is made likely by extrapolation from the recent study of Jespersen et al. 4 They compared the fibrinolytic potential of two oral contraceptives containing third-generation progestogens (desogestrel and gestodene) combined with ethinyl estradiol and noted a highly significant increase in plasminogen antigen. This was accompanied by a simultaneous and significant decrease (p < 0.01) in histidine-rich glycoprotein (the natural inhibitor of plasminogen synthesis) and a trend toward decreases in plasminogen activity inhibitor type I, another highly potent inhibitor of the conversion of plasminogen to plasmin. These changes were accompanied by a significant (p < 0.05) increase in tissue plasminogen activator activity. Changes in fibrinolytic activity and anticoagulation factors consistent with the hypothesis that hemostatic balance is normally maintained with norethindroneand levonorgestrel-containing low-dose oral contraceptives were seen in other studies l2 . 17 and are supported by the conclusions of this study: Triphasic levonorgestrel plus ethinyl estradiol and norethindrone plus ethinyl estradiol produced equivalent changes in coagulation parameters, but these changes were offset by increases in anticoagulation activity. This is supported by the observation that only one oral contraceptivetreated patient had a positive titer for fibrin degradation products and at only one measurement point. There was no indication that the progestogen modified the effect of estrogen on hemostasis. We therefore conclude that both tested oral contraceptives can be expected to have a minimal net effect on the laboratory evaluation of these hemostatic factors in healthy women. These findings are consistent with the observed clinical safety of these compounds.
REFERENCES 1. Bonnar J. Coagulation effects of oral contraception. AM ] OSSTET GYNECOL 1987;157:1042-8. 2. Kelleher CC. Clinical aspects of the relationship between oral contraceptives and abnormalities of the hemostatic system: relation to the development of cardiovascular disease. AM] OSSTET GYNECOL 1990;163:392-5. 3. Beller FK, Ebert C. Effects of oral contraceptives on blood
Coagulation and triphasic oral contraceptives
Volume 167 Number 5
4.
5.
6. 7. 8.
9. 10.
II.
12. 13.
coagulation. A review. Obstet Gynecol Surv 1985;40;42536. Jespersen J. Petersen KR. Skouby SO. Effects of newer oral contraceptives on the inhibition of coagulation and fibrinolysis in relation to dosage and type of steroid. AM J OBSTET GYNECOL 1990;163:396-403. Vessey MP. The Jephcott lecture. 1989: an overview of the benefits and risks of combined oral contraceptives. Oral contraceptives and breast cancer. Proceedings of a symposium held at The Royal Society of Medicine. London. July 1989. London: Royal Society of Medicine. 1989:121-35. Adams MR. Clarkson TB. Koritnik DR. Nash HA. Contraceptive steroids and coronary artery atherosclerosis in cynomolgus macaques. Fertil Steril 1987;47: 1010-8. Croft p. Hannaford P. Risk factors for acute myocardial infarction in women. BMJ 1989;298:674. Conard J. Samana M. The effects of progestins on coagulation. In: Bardin CWo Milgrom E. Manvais-Jarvis p. eds. Progesterone and progestins. New York: Raven Press. 1983:447-52. Bonnar J. Sabra AM. Oral contraceptives and blood coagulation. J Reprod Med 1986;3 I (suppl):55 1-6. Daly L. Bonnar J. Comparative studies of 30 ILg ethinyl estradiol combined with gestodene and desogestrel on blood coagulation. fibrinolysis. and platelets. AM J OBSTET GYNECOL 1990;163:430-7. Gevers Leuven JA. Kluft C. Bertina RM. Hessel LW. Effects of two low-dose oral contraceptives on circulating components of the coagulation and fibrinolytic systems. J Lab Clin Med 1987;109:631-6. Notelovitz M. Kitchens CS. Coone L. McKenzie L. Carter R. Low-dose oral contraceptive usage and coagulation. AM J OBSTET GYNECOL 1981;141:71-5. DavidJL. Gaspard UJ. Gillian D. Raskinet R. Lepot MR.
14.
15.
16.
17. 18.
19.
20. 21.
1261
Hemostasis profile in women taking low-dose oral contraceptives. AM J OBSTET GYNECOL 1990; 163 :420-3. Refn H. Kjaer A. Lebech A-M. Borggaard B. Schierup L. Bremmelgaard A. Metabolic changes during treatment with two different progestogens. AM J OBSTET GYNECOL 1990;163:374-7. Omsjo IH. Oian p. Maltau JM. Osterud B. Effects of two triphasic oral contraceptives containing ethinylestradiol plus levonorgestrel or gestodene on blood coagulation and fibrinolysis. Acta Obstet Gynecol Scand 1989;68:2730. Kjaer A. Lebech A-M. Borggaard B. et al. Lipid metabolism and coagulation of two contraceptives: correlation to serum concentrations of levonorgestrel and gestodene. Contraception 1989;40:665-73. Bonnar J. Daly L. Carroll E. Blood coagulation with a combination pill containing gestodene and ethinyl estradiol. Int J Ferti! 1987;32:21-8. Sabra A. Bonnar J. Hemostatic system changes induced by 50 ILg and 30 ILg estrogen/progestogen oral contraceptives. Modification of estrogen effects by levonorgestrel. J Reprod Med 1983;28:85-91. Smith LG. Kitchens CS. A comparison between two commercially available activators for determining the partial thromboplastic time. Arch Pathol Lab Med 1985;109; 243-6. Merskey C. Kleiner GJ. Johnson AJ. Quantitative estimation of split products of fibrinogen in human serum. Relation to diagnosis and treatment. Blood 1966;28: I. Kluft C. Disorders of the hemostasis system and the risk of the development of thrombotic and cardiovascular diseases: limitations of laboratory diagnosis. AM J OBSTET GYNECOL 1990;163:305-12.