Hormone replacement therapy: The right choice for your patient?

Recently published data² on the unopposed estrogen arm showed an increase in stroke (+12) and VTE (+7). No significant increase was noted for coronary heart disease (CHD) (–5) or breast cancer (–7). While there is a reduction in osteoporotic fracture (–6) in the hormone arm, there is no significant difference in colorectal cancer (+1). The CEE arm demonstrated that CEE does not significantly affect the incidence of CHD, a difference compared with the CEE/MPA arm. Potential explanations for this include the role of progestin, differences in the study populations, and the role of chance. The reduction of breast cancer incidence was unanticipated and required further investigation.

Dr Pees: The WHI data do seem to confuse the issue of when to use hormonal therapy. The answer to your question lies in knowing how the WHI risks were calculated.

The WHI authors described 2 confidence indexes (CI). The first was calculated as if looking at a single outcome variable, a nominal CI; the second, an adjusted CI, took into account multiple outcomes.

For the CEE/MPA arm, when the nominal CI is used, CHD and breast cancer risks barely reached clinical significance; colon cancer and fracture reduction reached clinical significance; DVT and strokes were also clinically significant using both CIs. When the adjusted CI is used, DVT/strokes, and to a lesser degree fracture reduction, retain clinical significance.

In the CEE-only arm (February 2004 report), DVT/strokes were the only variable that reached clinical significance. CHD/breast cancer/colorectal cancer did not reach clinical significance, and fracture reduction only a slight decrease in risk.

The WHI study introduced a new term, the Global Index, which had not been tested for validity or significance of assigning weights to various outcomes. Additionally, there were certain biases in the construct of the study. Included were 1200 individuals who had prior MI, revascularization procedures, stroke, DVT, and PE. Excluded were those with severe menopausal symptoms, prior fracture/low bone mineral density (BMD), and cognitive function deficits. Also, only 10% of the group was in the 50- to 54-year-old group; 25% were in the 70- to 79-year-old group. The dropout rate was fairly high: 42% in the CEE/MPA arm, and 38% placebo arm; only 25% of the CEE/MPA arm remained at the study termination. Unblinding occurred in 40% of the CEE/MPA arm and 5.4% in the placebo arm.

It is worth noting that CHD increased in the first year of the study but decreased in subsequent years. VTE/DVT and total fractures were the only 2 items that achieved true clinical significance. Total fracture reduction was 1.6 times greater than the increases in CHD and breast cancer; and the fact that breast cancers double every 300 days and are present for 7 to 8 years before they are detected with current diagnostic testing.

Despite the biases and limitations of the data presented by the WHI report, the study does add to our knowledge and it should be included in our discussion with patients when addressing HRT.

Dr Pees: The WHI data have altered the way in which I counsel patients. I still prescribe hormone therapy within the American College of Obstetricians and Gynecologists (ACOG) guidelines, and, after a very thorough discussion, reference the WHI data. This is to ensure that patients make informed decisions about continuing or starting hormone therapy. Following ACOG recommendations, I evaluate each patient’s risk profile. I start treatment with the lowest effective dose to relieve symptoms. I discuss the use of alternative treatments for menopausal symptoms and osteoporosis prevention, as well as evaluation and treatment for quality of life issues. Most importantly I encourage periodic re-evaluation for hormone use.

After the WHI study was terminated, the North American Menopause Society (NAMS) convened an expert Hormone Therapy Advisory Panel to examine the data and prepare a report. A position statement on the recommendations for clinical practice was released in September 2003.² The expert panel reached consensus on the following areas:

• The primary indication for systemic hormone therapy is moderate and severe menopausal symptoms: Hot flashes, night sweats/insomnia, mood swings.
• When treating moderate and severe urogenital atrophy (vaginal dryness, dyspareunia, urinary frequency, and incontinence), local estrogen preparations are preferred.
• The primary indication for progestogen is endometrial protection for all women with an intact uterus using systemic estrogen therapy (ET).
• Hormone therapy should not be used for primary or secondary prevention of CHD.
• Breast cancer risk is increased with ET and, to a greater extent, with estrogen progestogen therapy (EPT) used beyond 5 years.
• There is definitive evidence for EPT efficacy in reducing the risk of fracture. However, other effective alternate forms of therapy are readily available. Because of potential risk associated with long-term use of EPT, the risks and benefits of each option should be discussed with the patient.
• EPT is not recommended for primary prevention of dementia.
• Hormone therapy should be limited to the shortest duration to achieve treatment goals, taking into consideration the impact on quality of life.
• Lower-than-standard doses of EPT and ET should be considered (including 0.3 mg conjugated estrogen, 0.25–0.5 mg oral micronized 17 B-estradiol and 0.025 mg 17 B-estradiol patches). However, the long-term risk-benefit ratio has not been demonstrated.
• Extended use of ET/EPT is acceptable if the woman feels the benefits of symptom relief outweigh the risks, an attempt to withdraw ET has failed, or to prevent osteoporotic fracture when alternative therapies are not appropriate or usable.