An Overview
Robert L. Barbieri
Department of Obstetrics, Gynecology and Reproductive Biology, Brigham and
Women's Hospital, Harvard Medical School, Boston, Massachusetts 02115
 |
Introduction
|
---|
IN THIS Clinical Controversy, Drs. Speroff, Walker, and McPherson review
the possible association between the use of oral contraceptives
containing the synthetic progestins gestodene or desogestrel and the
risk of venous thromboembolism (VTE). Speroff interprets the available
data as indicating that all oral contraceptives are associated with a
small increase in the relative risk of VTE, and that there are no
differences in the risk of VTE among various contraceptives containing
second (norgestrel) or third (gestodene or desogestrel) generation
progestins. In contrast, Walker and McPherson interpret the data as
indicating that the use of oral contraceptives containing the third
generation progestins are associated with a greater risk of DVT. Two
major issues were discussed by all three experts: 1) how to assess the
quality of the evidence; and 2) developing a meaningful framework for
integrating an analysis of the benefits and risks of a contraceptive
hormone. In this overview, I would like to discuss these two issues and
then reflect on the leadership role the endocrinologist must take to
better define the pharmaco-endocrinology of gestodene and
desogestrel.
 |
The quality of the study design influences the quality of the
data
|
---|
Most clinical investigators believe that, for any experiment, the
research design is the single, most critically important variable in
determining the quality of the data generated and the power of the
study to isolate cause-effect mechanisms. Most clinical investigators
believe that the prospective, controlled clinical trial with
randomization of subjects is the study design most likely to generate
reliable, high quality data that can yield generalizable, cause-effect
insights. Unfortunately, no prospective randomized studies have been
performed that explore the relative risk of VTE in users of oral
contraceptives that contain gestodene or desogestrel compared with
those containing norgestrel or norethindrone. The absence of data from
high quality clinical trials forces clinicians to rely on information
from epidemiological studies to explore this association.
As reviewed by Walker and McPherson, case control studies have
suggested an association between the use of oral contraceptives with
gestodene or desogestrel (third generation progestins) and an increased
risk of DVT compared with oral contraceptives that contain norgestrel
(second generation progestin). Walker and McPherson believe that the
reported case-control studies adequately control for both confounding
and bias and that the data is reliable. In contrast, Speroff suggests
that the case-control studies reported in 1995 and 1996 had design
problems that limit the reliability of the data. Speroff believes that
the initial case-control studies did not control for important clinical
differences between the users of second and third generation oral
contraceptives. Speroff cites subsequent case-control studies that
controlled for duration of oral contraceptive use and/or that were
focused on first time users. These studies found no differences between
second and third generation oral contraceptives and the risk of VTE.
For example, using the German MediPlus data base, Farmer and colleagues
(1) reported that, among women on oral contraceptives who were treated
with an anticoagulant for VTE, there was no increased use of the third
generation progestins. Supporting these results are findings by Suissa
and colleagues (2), which indicate that, in a case-control study
limited to first time users and controlled for the duration of oral
contraceptive use, there was no difference in the risk of VTE with
second or third generation oral contraceptives. Interestingly, Suissa
and colleagues (2) and Lidegaard and colleagues (3) both reported that
the risk of VTE among oral contraceptive users was greatest in the
first year of use and decreased markedly with extended use. These
studies demonstrate the importance of controlling for duration of
use in studies that explore the association between hormones and VTE.
My assessment of the literature is that there is no convincing evidence
of an increase of VTE in users of third generation versus
second generation oral contraceptives.
 |
Benefit-risk assessments are extremely complex
|
---|
A second major focus of the commentaries of Speroff, Walker, and
McPherson is a discussion of the proper framework for analyzing the
benefits and risks of contraceptive hormones. Benefit-risk analysis is
inherently complex because it is difficult for the human mind to
balance the total effect of a large number of benefits and a small
number of risks. Many studies suggest that humans tend to overestimate
the probability that a rare event (such as winning the lottery, or a
major asteroid striking the earth) might actually occur to them. As
noted by McPherson, the potential risk of VTE associated with the use
of oral contraceptives containing gestodene or desogestrel is far less
than the risks associated with unintended pregnancy or the risks
associated with the use of tobacco products. The oral contraceptives
that contain gestodene or desogestrel have many health benefits,
including a reduced risk of pregnancy, ovarian cancer, and endometrial
cancer. In one cohort study, long-term use of oral contraceptives
reduced the risk of endometrial cancer by 90% and the risk of ovarian
cancer by 60% (4). From my perspective, the benefits of oral
contraceptives that contain gestodene or desogestrel clearly far
outweigh the risks of the agents. Speroff discusses the importance of
clinical judgment in weighing the benefits and risks of an oral
contraceptive. For example, he notes that clinicians should be careful
about prescribing oral contraceptives to women with a previous history
of idiopathic VTE or a close family history (parent or sibling) of
VTE.
It is important for all clinical investigators to minimize the
possibility that a rare potential risk is overemphasized by the media,
resulting in unintended adverse public health consequences. The British
Pregnancy Advisory Service reported that media coverage of a potential
link between the third generation progestins and VTE was followed by a
10% increase in the number of abortions performed (5). Public health
would have been better served if the media had reported the harmful
effects of tobacco and emphasized the beneficial effects of all oral
contraceptives.
 |
The leadership role of the endocrinologist
|
---|
During the recent debate on the effects of the second and third
generation oral contraceptives, epidemiologists published far more
reports than clinical endocrinologists. However, it is the clinical
endocrinologist, not the epidemiologist, who has the depth and breadth
of knowledge of steroid biochemistry, molecular and cellular
endocrinology, and endocrine physiology to scientifically design the
critical experiments in this arena. Clinical endocrinologists need to
be leaders both in investigating the potential physiological effects of
combination estrogen-progestin medications and in communicating the
findings of their studies to the public. The liver is the largest
endocrine target organ, and given the powerful research tools available
to the endocrinologist, additional investigations of the effects of
novel estrogen-progestin combinations on liver production of
clotting factors, both in vitro and in vivo, is
timely and necessary.
 |
Footnotes
|
---|
Address correspondence and requests for reprints to: Robert L.
Barbieri, Department of Obstetrics and Gynecology, ASB1-3, Brigham and
Womens Hospital, 75 Francis Street, Boston, Massachusetts 02115.
Accepted March 3, 1999.
 |
References
|
---|
-
Farmer RD, Todd JC, Lewis MA, MacRae KD, Williams
TJ. 1998 The risks of venous thromboembolic disease among German
women using oral contraceptives: a database study. Contraception. 57:6770.[CrossRef][Medline]
-
Suissa S, Blais L, Spitzer WO, Cusson J, Lewis M,
Heinemann L. 1997 First-time use of newer oral contraceptives and
the risk of venous thromboembolism. Contraception. 56:1416.[CrossRef][Medline]
-
Lidegaard O, Edstrom B, Kreiner S. 1998 Oral
contraceptives and venous thromboembolism. A case control study. Contraception. 57:291301.[CrossRef][Medline]
-
Vessey MP, Painter R. 1995 Endometrial and ovarian
cancer and oral contraceptivesfindings in a large cohort study. Br J Cancer. 71:13402.[Medline]
-
Dillner L. 1996 Pill scare linked to rise in
abortions. BMJ. 312:996.[Free Full Text]