Lessons Learned From the Women's Health Initiative Study
from Southern Medical Journal
We sail within a vast sphere, ever drifting in uncertainty, driven from end to
end. When we think to attach ourselves to any point and to fasten to it, it
wavers and leaves us; and if we follow it, it eludes our grasp, slips past us,
and vanishes forever. Nothing stays for us.
Blaise Pascal (1623-1662)
French scientist, philosopher
Pensees, no. 72
Uncertainty is the refuge of hope.
Henri-Frederic Amiel (1821-1881)
Swiss philosopher, poet
Journal Intime
The estrogen/hormone replacement therapy (HRT) pendulum has been swinging for
decades. In the 1950s and before, there was little interest in estrogen therapy
for postmenopausal women. In the 1960s, estrogen therapy was viewed as a
fountain of youth. In the 1970s, concerns arose regarding the increased risk of
endometrial cancer with unopposed estrogen therapy. The 1980s saw the appearance
of "estrogen evangelists," who believed that estrogen should be given to -- even
forced upon -- all postmenopausal women (and perhaps added to the water supply)
because of its protective effects on the cardiovascular system.[1] In the 1990s,
concerns were raised about the effect of estrogen therapy on breast cancer risk,
particularly when estrogen was combined with a progestin.[2,3] The course of the
pendulum was deflected by the Heart and Estrogen/Progestin Replacement Study
(HERS).[4] Now the pendulum may have been stopped altogether by new results from
the Women's Health Initiative (WHI).[5]
We know that estrogen is an effective agent for prevention of bone loss in
recently menopausal women. Estrogen therapy has been shown to increase spinal
bone mass by 5% to 10% in numerous, randomized, placebo-controlled trials.[6-10]
Until about 5 years ago, I was saying and writing that estrogen is the treatment
of choice for established postmenopausal osteoporosis, and I suspect that most
experts in the field agreed with me at the time. This opinion was based on
observational studies that suggested that estrogen use for 7 years or longer was
associated with 25% to 50% fewer spine and hip fractures,[11-14] and the belief
that there were significant extraskeletal benefits. In 1999, however, treatment
of osteoporosis was withdrawn as an indication for estrogen therapy by the Food
and Drug Administration[15] because estrogen had not been studied in the same,
rigorous fashion required of newer medications used for treatment of
osteoporosis.
The antifracture benefit of HRT was recently confirmed in the HRT arm of the WHI,
a large, prospective, randomized, double-blind, controlled trial.[5] Good news!?
In this study of more than 16,000 postmenopausal women, HRT reduced the risk of
hip and clinical spine fractures by one third, the risk of all osteoporotic
fractures by 23%, and the total risk of fractures by 24%. This finding is
particularly remarkable in that the subjects in WHI were not selected for having
osteoporosis. The WHI is a nationwide trial being conducted at 40 centers.
Planning began in the 1980s. Recruitment of postmenopausal women, aged 50 to 79
years, with no competing health risks and no likely adherence or retention
problems, began in 1993. The trial is scheduled to be completed in 2005. More
than 161,000 women enrolled in this ambitious project. Close to 94,000 women are
in an observational study because they were either ineligible for or unwilling
to consent to one of the clinical trials. More than 68,000 women are
participating in 1 of 3 interventional studies: a dietary modification study, an
estrogen/hormone replacement study, and a calcium + vitamin D trial (which is
nested in the other 2).
Current attention is focused on the HRT trial. This study involved more than
16,000 postmenopausal women (mean age, 63 years) who had not had hysterectomy
and who agreed to be randomized to either placebo or a specific combination of
HRT (conjugated equine estrogen, 0.625 mg/day, plus medroxyprogesterone acetate,
2.5 mg/day). The Data and Safety Monitoring Board (DSMB) stopped the trial early
(after an average 5.2 years of follow-up rather than the planned 8.5 years)
because of a statistically significant 26% increase in the risk of breast
cancer, and because the overall balance of risks and benefits (global index) was
unfavorable.[5] In addition to the increase in breast cancer, there was a 41%
increase in strokes, a 29% increase in heart attacks, and a 100% increase in
venous thromboembolic events in women receiving HRT.
On the positive side, in addition to the reduced risk of fractures, there was a
37% reduction in colorectal cancer; however, on balance, the risks of long-term
HRT were greater than the benefits. The good news that HRT reduces the risk of
fracture is largely moot because of the bad news of the increased risks of
breast cancer and cardiovascular disease.
But (as they say in TV infomercials) wait, there's more! In the Study of
Osteoporotic Fractures, women who took HRT were more likely to have back pain,
possibly because of changes in ligaments and supporting structures of the
spine.[16] In post hoc analyses of data from the HERS trial, HRT was associated
with worsening of urinary incontinence[17] and, in women who were not
symptomatic when HRT was started, a decline in physical function and energy, and
increased fatigue.[18] The hoped-for benefit of estrogen slowing the progression
of Alzheimer's disease could not be shown in a well-designed and adequately
powered randomized trial.[19] Thus, there are several other reasons not to give
HRT to asymptomatic women for treatment of postmenopausal osteoporosis
The HRT arm of WHI only examined a single HRT regimen (0.625 mg conjugated
estrogen plus 2.5 mg medroxyprogesterone acetate daily). The results may not
pertain to other forms of estrogen therapy (oral or transdermal) used with daily
medroxyprogesterone acetate, to estrogen used with cyclic medroxyprogesterone
acetate or other progestins, or to estrogen used alone. Does that mean that
other HRT preparations are safer than the one tested in WHI? No one knows. We
will probably never know.
The estrogen-only arm of WHI was not stopped by the DSMB. Does that mean the
same problems that emerged with HRT do not occur with conjugated equine estrogen
given without medroxyprogesterone acetate (unopposed estrogen)? Not necessarily.
The estrogen-only arm of WHI is large, but with only 10,000 participants, it
still may not have the statistical power to show the same effects for another
year or two.
Is this another swing of the pendulum? I think not. I believe this is a paradigm
shift. The conclusive evidence that HRT lacks cardioprotective effects and
appears to increase the risk of heart attacks and strokes in apparently healthy
women (as well as in those with established heart disease) eliminates one of the
major reasons for prescribing HRT. The increased risk of breast cancer, though
small and consistent with observational data, strikes an emotional chord that is
hard to overcome.
What should we be doing for our postmenopausal patients? In absolute terms, the
increased risk of cardiovascular disease and breast cancer associated with HRT
is relatively small, only about 1 excess case per 1,000 women per year, or, over
10 years, about 1 excess case per 100 women. The WHI did not address
quality-of-life issues. I believe that women who need estrogen therapy for
relief of symptoms of estrogen deficiency, such as hot flashes and vaginal
dryness, should not hesitate to take estrogen or HRT, but should taper or stop
HRT every few years to find out if continued treatment is needed, and that women
currently taking HRT who are convinced that their quality of life is improved
because of HRT should probably continue taking it.
Women who were asymptomatic when they started HRT for what we now know to be the
wrong reasons, or who started HRT because of symptoms but might now be
asymptomatic if HRT were stopped, should reevaluate their options. It probably
makes sense for most of them to stop HRT. Should their treatment automatically
be changed to something else? Not so fast! We should stop thinking of menopause
as a disease. Women who stop HRT and are at risk for cardiovascular disease
should consider taking medications such as hydroxymethylglutaryl co-enzyme A
reductase inhibitors (statins), agents that have proven effectiveness against
cardiovascular disease. Women who stop HRT and have osteoporosis should receive
medications such as raloxifene or a bisphosphonate, agents that have proven
effectiveness in reducing the risk of fracture. Women who stop HRT but seem
otherwise healthy do not need medication.
If, in the process of reconsidering HRT, bone status is not known, bone mineral
density (BMD) testing should be done to determine if adding a pharmacologic
agent is appropriate. It is less expensive to test for osteoporosis than it is
to treat it. Based on BMD results, a rational decision can be made to start a
preventive regimen (in a patient whose BMD is borderline-low who has risk
factors), a treatment regimen (in a patient who has developed osteoporosis,
despite taking HRT), or just a healthful lifestyle (in a person whose BMD is
normal or borderline-low without risk factors).
As the old cigarette ad used to say, "You've come a long way, baby." The
swinging of the pendulum has stopped. We can now regroup and provide rational
advice for our patients.
Nelson B. Watts, MD
